SUGAR SERIES Vol. Vol. Vol. Vol.
1. 2. 3. 4.
Standard Fabrication Practices for Cane Sugar Mills (Delden) Manufacture and Refining of Raw Cane Sugar (Baikow) By-Products of the Cane Sugar Industry (Paturau) Unit Operations in Cane Sugar Production (Payne)
sugar series, 4
unit operations in cane sugar production JOHN HOWARD PAYNE
ELSEVIER SCIENTIFIC PUBLISHING COMPANY Amsterdam — O x f o r d — N e w Y o r k
1982
ELSEVIER SCIENTIFIC PUBLISHING COMPANY Molenwerf 1 , P.O. Box 2 1 1 , 1000 A E Amsterdam, The Netherlands
Distributors for the United States and Canada: E L S E V I E R SCIENCE P U B L I S H I N G C O M P A N Y INC. 5 2 , Vanderbilt Avenue New York, N.Y. 10017
Library
of C o n g r e s s
Cataloging
in P u b l i c a t i o n
Data
Payne, John Howard, 1906Unit operations in cane sugar production. (Sugar series ; k) Bibliography: p. Includes index. 1. Sugar—Manufacture and refining. I. Title. II. Series. f TP37T.P38 1982 661* . 122 82-11373 ISBN 0-UUU-U210U-1
ISBN 0-444-42104-1 (Vol. 4) ISBN 0-444-41897-0 (Series) © Elsevier Scientific Publishing Company, 1982 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher, Elsevier Scientific Publishing Company, P.O. Box 3 3 0 , Amsterdam, The Netherlands Printed in The Netherlands
ν
PREFACE This compilation is intended as a guide for the efficient performance of the several unit operations obtaining in a factory processing sugar cane. It is not a textbook, engineering handbook, or equipment operating manual. The purpose is only to present in simple form the basic principles involved and some of the means of reaching optimum results. The reader is assumed to be grounded in the fundamental chemistry, physics, engineering, biology, economics and common sense involved in sugar factory routine. He is also assumed to have had a reasonable length of experience in the industry. These practices are designed for an Hawaiian type of raw material. That is, two-year cane, rake harvested under both wet and dry conditions, requiring the use of wet cleaning plants. The product is a sugar averaging 99 pol, designed for delivery to a refinery. Although some of the Hawaiian conditions are not common, most of the practices are applicable to all cane sugar areas. The author salutes with respect his predecessors who developed the vast reservoir of knowledge which forms the basis of present technology. He thanks his contemporaries who have inspired as well as contributed. He greets the upcoming with the admonition that if they do not improve on this, we both have failed.
vi
ACKNOWLEDGMENTS
Acknowledgment goes to Amfac Inc. for initially suggesting the writing and close cooperation in the period of its development, to the Hawaiian Sugar Planters' Association for permission to use some unpublished material and to John W . Herkes who assisted in the preparation of the chapter in Equipment Maintenance. The comments of Hideo Idehara and J. R. Albert-Thenet are appreciated. Drawings are the work of James Kawamura. Specific recognition is gratefully extended to Misue Okino Sakamoto who assisted in editing and proofreading the manuscript and did much of the detailed assembly work.
vii
1
When Omer smote 'is bloomin
lyre,
,
He'd eard men sing by land an' sea; 1
An -what he thought 'e might
require,
Έ went an' took - the same as me. Rudyard
Kipling
Chapter 1
FACTORY CONTROL The monitoring, measuring, sampling and analytical procedures necessary for operational control and material accounting are outlined in the Schedule for Measuring, Sampling and Analysis at the end of this chapter. These are considered to be the minimum that will provide adequate information for all concerned in the factory operation. It is assumed that automatic devices are used wherever practicable. The laboratory load is reduced to a level where only one analyst is used per shift, exclusive of the analytical work necessary for field distribution and agricultural control. No provision is included for special analytical work of an experimental nature. The procedures are those of The Official Methods of the Hawaiian 1 Sugar Technologists. PROCEDURES Field Cane Received Measuring Weigh transport units at tolerance of ± 0.25%.
the
factory on a beam scale with a
Calibrate the scale annually, using known weights. Check tares each shift. No accumulation of dirt on the platform and dirt and water in the scale pit is permitted. Report the weight of cane received as the difference between the weights of the transport unit before and after discharging the cane. This weight is reported with the field area designation and the transport unit number. Sampling Core sample the cane on the transport. Equipment. A horizontal core sampling machine is used. The coring tube of 25 cm in diameter and revolves at 100 rpm. The tube enters 2 m into the cane through an opening provided in the transport approximately midway vertically in the load. Upon retraction, the sample in the tube is discharged into a conveyor leading to a 25-cm Rietz Prebreaker which chops the sample and discharges it into a revolving disc subsampler. This drops a portion into a subsample tray and discharges the remainder onto a disposal conveyor.
2
Procedure. The coring tube must enter the full distance into the load and the entire contents of the tube must go through the Prebreaker. The subsampler should approximately 250 g (0.5 lb).
be
adjusted
to
give
a
subsample
of
The entire subsample is placed in an airtight plastic container. Four cores from the sample area are composited in the container and constitute one sample for analysis.
same
Frequency. The number of cores taken is based upon the requirement to obtain a number of samples for analysis equal to the square root of the number of deliveries per unit area per week. Four cores are composited to make one sample for analysis in all cases where the number of deliveries exceeds four. The number of cores and samples needed is derived from the following table: No. of loads
Cores
Samples for analysis
1 2 3 4 5-16 17-30 31-42 43-56 57-72 73-90 91-110 111-132 133-156 157-182 183-210
1 2 3 4 5-16 20 24 28 32 36 40 44 48 52 56
1 2 3 4 4 5 6 7 8 9 10 11 12 13 13
900
120
30
Analysis Disintegrator Method for Fiber. Extract with the following variations:
Pol and Refractometer Solids %
1. The frozen sample is placed in a tared 12-gauge polyvinyl chloride (pvc) bag and weighed. The sample is broken up by beating with a mallet on the total contents of the bag (unless over 1000 g) are added directly to the disintegrator. The sample is not allowed to thaw prior to disintegration. With this precaution, mercuric chloride and sodium bicarbonate are not necessary. 2.
Pol is run on each sample.
3. Fiber is determined only on the basis of the square root of the number of samples, making sure that at least one fiber analysis is made on each area each day.
3 Prepared Cane Measuring Weigh prepared cane on a belt weighing device, with a precision of ± 2.0%, just ahead of the extracting unit. In diffuser installations, this is immediately following the Fiberizer. Sampling In the absence of a chute-gate time-cycle automatic sampling device, grab sampling is used. Samples are taken over a 5-minute period and held in a closed container until analyzed. No preservative is used, and the sample must be analyzed within 2 hours. Analysis Disintegrator Method for Fiber, Pol and Refractometer Solids, except that no sodium bicarbonate and mercuric chloride are used as preservatives. Stalk Cane Sampling 1. Cut and pull two stools of cane at benchmark locations in the area (not exceeding 12 ha). 2. Remove trash and weigh. Sample should include all the live stalks in the stools. A t this point, data on pest damage and cane condition may also be recorded. 3. Reduce the cane in the Prebreaker at the core sampling station, or in an ensilage cutter. 4.
Subsample the reduced material by quartering to approximately
500 g. 5. Place the subsample in an airtight container and preserve in a freezer until analyzed. Analysis Disintegrator Method for Fiber, Pol and Refractometer Solids % Extract. No preservatives are used. Bagasse Measuring Weigh on a belt weighing device, with a precision of ± 2.0%. Sampling Sample in the feed chute to the conveyor or to the pneumatic feeder. The sampling device consists of a time-cycle actuated, hydraulic slide gate which allows bagasse to drop into a container. The recommended timing is every 30 minutes, giving eight openings in a 4-hour period. The total amount collected should approximate 25 liters. The sample is taken to the laboratory and quartered to provide two 100 g samples.
4
Analysis Moisture Oven Method or Moisture Teller Method. Fiber Disintegrator Method. No preservatives are used. Pol Disintegrator Method. First Expressed Juice Sampling Collect separate samples from the crusher and the discharge roll of the first mill by hand, or by a continuous sampler. Analysis Crusher sample for Refractometer Solids and Pol. Purity of First Expressed Juice is calculated from these results. First mill sample for Refractometer Solids. This figure is reported for First Expressed Juice since it is less affected by extraneous water carried from the cleaning plant. Determine pH of the first mill sample. Absolute Juice Calculated from First Expressed Juice Pol, Refractometer Solids, Purity and Absolute Juice Factor. Analyzed as Juice in Prepared Cane Pol, Refractometer Solids and Purity. Last Expressed Juice Sampling Sample by hand from the last two rolls of the tandem, or from the dewatering device of the diffuser. Analysis Pol and Refractometer Solids. Press Return Measuring * Measure volume by means of a proportional weir at the point at which the press return enters the diffuser. Sampling Sample by hand from the press return weir box on top of the diffuser.
5
Analysis Pol, Refractometer Solids and Insoluble Solids. Mixed Juice Measuring Weigh on automatic beam scales with a precision of ± 0.1%. In order to maintain this precision, the rate of flow of juice into the scale tank must be uniform to insure a constant overshoot, i.e., the quantity of juice flowing into the tank after the scale is balanced and before the valve shuts off. Inlet and discharge valves are interlocked so that both will not be open at the same time. The scales are equipped with a recorder registering tank dumps. Tare is taken at least once a shift and the scales are calibrated annually and after any major maintenance. Sampling Sample automatically and proportional to the juice flow. In cases where hot diffuser juice is weighed, the sample must be protected from evaporation. Analysis Pol, Refractometer Solids, Insoluble Solids and pH. Filtrate Sampling Collect a sample from the filtrate return by hand. Analysis Pol, Refractometer Solids and pH. Clarified Juice Sampling Collect a sample from the evaporator supply tank by hand. Analysis Pol, Refractometer Solids and pH. Filter Cake Measuring and Sampling Cut measured areas of filter cake from two spots on the filter drum, just before the discharge blade. Weigh immediately. From the total filter area and the number of revolutions, calculate the total weight of cake discharged. Use the weighed samples for analysis. Analysis Fiber, Moisture and Pol.
6
Syrup Sampling Sample syrup continuously by collecting a drip sample from discharge of the syrup pump.
the
Analysis Pol, Refractometer Solids and pH. Massecuite Measuring Calibrate each pan for volume at various striking levels. Record the volume for each strike and estimate the weight from the calculated density. Sampling Collect a sample by hand from the flow of massecuite from the pan. Sample at intervals during the steady flow period of discharge. Analysis Pol and Refractometer Solids. Calculate Crystal content. Molasses, A , Β and C Sampling A - and B-molasses: Collect sample by hand from the centrifugal discharge trough as flow enters the pump feed tank. The sample should be taken at about the mid-point of centrifuging. C-molasses: Two samples are obtained, before and after the crystallizer. 1. Filter by pressure a portion of the C-massecuite obtained when the strike is dropped into the crystallizer.
sample
2. Filter by pressure a sample of the massecuite entering the heater after the crystallizer. Analysis Pol and Refractometer Solids. Molasses, Final Measuring Measure molasses in the storage at the factory with a Pneumercator with a precision of ± 0.5%. Calibration of Pneumercator: The average net area of tank is determined from circumferential measurements and the weight is calculated for a standard pressure of 2 2 0.71 g / c m (10 lb/in. ). The vertical height of the corresponding reading
7
on the manometer scale is then determined using a special calibrating jig. It is compared with the correct height of a column of mercury at this pressure to determine percent error. The manometer is then adjusted if necessary to bring the actual height into agreement with the correct height. Official weights are obtained from the terminal where molasses deliveries are weighed on beam scales with a precision of ± 0.1%. Sampling At the factory, sampling is by means of a continuous drip sample from the discharge of the pump handling the molasses to storage. At the terminal, sampling is conducted on each delivery. Analysis Samples collected Refractometer Solids.
at
the
factory
are
analyzed
for
Pol and
A weekly composite of the factory molasses is analyzed for Pol, Refractometer Solids, Sucrose, Reducing Substances and Ash, Carbonate. Purity Expected is calculated. The samples from the terminal are composited weekly and analyzed for Brix 1-1 dilution. Sugar - Remelt Sampling Sample continuously by collecting a drip sample from the discharge of the pump delivering the remelt to storage on the pan floor. Analysis Pol and Refractometer Solids. Sugar - Commercial Measuring Sugar is weighed at the terminal on beam scales with a precision of ± 0.1%. Sampling Factory: Each strike is sampled from the Composite for analysis on an 8-hour basis.
belt after
the
centrifugals.
Terminal: Each delivery is sampled automatically by a bucket sampler from the belt conveyor to the storage area. Analysis Factory and Terminal samples are analyzed for Pol and Moisture.
8 Condensate Sampling and Analysis Individual sources of condensate, evaporator cells, vacuum pans and juice heaters are monitored continuously by electrical conductivity. Response to high conductivity is to divert flow from boiler feedwater supply to house hot water supply - either automatically or manually. In the meantime, a sample should be collected and checks made with alpha naphthol to determine if high reading is caused by sugar carryover. A high reading may be caused by too low pH of clarified juice in the case of evaporator condensate. Feedwater Sampling and Analysis Feedwater is monitored continuously by electrical High conductivity water is discarded.
conductivity.
Boiler Water Sampling A sample is collected mannually every 8 hours from a sample line fitted with a condenser assembly. Analysis Alkalinity, pH, Phosphate Sulphite and Total Dissolved Solids. For high pressure desirable.
boilers,
Silica and Dissolved Oxygen are also
Instead of using Hawaiian Sugar Technologists Methods, standard procedures of companies specializing in boiler water control may be used. Condenser Water and Drains Sampling Condenser water from evaporator and pans, collection areas, are sampled manually once an hour.
and
drains
from
Analysis Colorimetric method for sugar. A suggested Schedule for Measuring, Sampling and Analysis is given in the following pages. Examples of Daily, Weekly and Recovery and Loss reports are also shown. REFERENCES 1
Payne, John H . , (ed.) Sugar Cane Factory Analytical Control, The Official Methods of the Hawaiian Sugar Technologists, Elsevier, Amsterdam, 1968 Revised Edition.
9
SCHEDULE FOR MEASURING, SAMPLING AND ANALYSIS Stream Cane Field Prepared Stalk Bagasse Final Juice First Expr. Absolute Last Expr. Mixed Filtrate Clarified Filter Cake
Measurement/Method Wt/Beam scale Wt/Belt scale
Core sampler Intermittent Grab
Wt/Belt scale
Intermittent
Wt/Beam scale
Prep, cane or calc. Grab Cont.(grab for insol.sol.) Grab Grab
Wt/calculated
Grab
Grab
Syrup Massecuite A, B, C
Sampling Method
Continuous Pan volume
Grab
Molasses A , B, C Final (Factory) Pneumercator Final (Terminal) Wt/Terminal scale
Grab Continuous Grab
Re melt
Grab
Sugar Factory Terminal Water Condensate Boiler Feed Boiler Water Drains
Wt/Terminal scale
Grab Continuous Monitor Monitor Grab Grab
10
Schedule for Measuring, Sampling and Analysis (continued) Analysis
Analytical Method
1/24 hr
Composite 4 cores Composite 1/4 hr Daily/area
Ref.Sol., Pol, Fiber Ref. Sol., Pol, Fiber Ref. Sol., Pol, Fiber
HST Pol Ratio HST Pol Ratio HST Pol Ratio
Bagasse Final
1/1 hr
Composite
Ref. Sol, Pol, Moist.
HST
Juice First Expr.
1/8 hr
1/8 hr
HST
Prep, cane or calc. 1/8 hr 1/4 hr
1/4 hr
Ref. Sol., Pol, pH Prep, cane or calc. Ref. Sol., Pol Ref. Sol., Pol, pH Insol. Sol. Ref. Sol., Pol, pH Ref. Sol., Pol, pH
HST
Stream Cane Field Prepared Stalk
Absolute Last Expr. Mixed
Sampling Frequency
Γ
\J Trucks/area/ day 1/1 hr
1/4 hr
Analysis Frequency
1/8 hr 1/4 hr
HST HST HST
Filtrate
1/24 hr
Composite 1/24 hr 1/24 hr
Clarified
1/8 hr
1/8 hr
Filter Cake
1/8 hr 1/8 hr
1/8 hr Composite 1/24 hr
Pol Fiber, Moist.
HST HST
Syrup
1/8 hr
Composite 1/8 hr
Ref. Sol., Pol, pH
HST
Massecuite A , B, C
Each strike
Each strike
Ref. Sol., Pol
HST
Remelt
1/8 hr
1/8 hr
Ref. Sol., Pol
HST
Each strike
Each strike
Ref. Sol., Pol
HST
1/8 hr
Composite 1/8 hr Composite weekly
Ref. Sol., Pol
HST
Ref. Sol., Pol, Sue, Red. Sugar, Ash, Brix 1-1
HST
Molasses A , B, C Final, Factory Terminal
Each truck
HST
HST
11
Schedule for Measuring, Sampling and Analysis (continued)
Stream Sugar Factory Terminal
Sampling Frequency
Analysis Frequency
Each strike
Composite 1/8 hr Composite weekly
Pol, Moist.
HST
Pol, Moist.
HST
Monitor Monitor 1/8 hr
Conductivity Conductivity pH, Alk., Phos. Suif., TDS Sugar
Cond. Cond. Boiler Control
Each truck
Water Condensate Boiler feed Boiler Water 1/8 hr Drains
1/24 hr
1/24 hr
Analysis
Analytical Method
HST
12 DAILY
FACTORY
REPORT
AMFAC FORM 1045B 2/71 Data & Tim* Started
[Data & Time E n d a d ( S I G N E D
PERFORMANCE PREPARED CANE NET lAVAIL.
DELAYS
(HOURS)
PRODUCTION
TIME
EXTRACT.
BOILING
POWER
EXTRACTING
AVAILABLE
PLANT
HOUSE
[GENERATION
TIME
TIME
SOLIDS REF. FIELD CANE
REC
IPREPARED CANE [STALK FILTER
CANE CAKE
1ST E X P . JUICE [LAST E X P . J U I C E ICLARIFIED IMIXED
JUICE
JUICE
[FINAL MOLASSES
DATE SAMPLED
FIELD NUMBER
Fig.
FIELD CANE RECEIVED
1-1. Daily factory report.
PREPARED CANE
13 WEEKLY FACTORY AM F AC FORM 1613 11/70
REPORT
FACTORY
FIELD
CANE
REPORT NO.
WEEK ENDING
SOLIDS REF.
PURITY
SIGNED
RECEIVED
TONS
PREPARED
TONS POL
POL %
FIBER %
CANE
TONS
TONS POL
SOLIDS REF.
POL %
FIBER %
This Week To Date EXTRACTING Extraction
CANE DILUTION %
Abs. Juka Prep.Cane|
Fiber
This Week
TONS PER HOUR E DAYS HOURS ieldCane Cetondar) Fiber p>rep.Cane
HOURS LOST
%
AREA HARV.
V A R I E T I ES %
%
%
I
To Date BAGASSE
DELAYS
TONS
POL %
TONS POL
Moisture %
FIBER %
NO CANE
Acres
EXTRACTING TIME
(HOURS) EXTN. Cleaner PLANT.
SOILING POWER GEN. HOUSE
HOURS Effciency AVAIL. I %
MISC.
This Week To Date MIXED
CLARIFICATION AGENTS
JUICE
GROSS TONS
NET TONS
TONS SOLIDS REF.
SOLIDS REF.
TONS POL
POL %
PURITY
Insoluble Solids %
LIME TONS
MgO TONS
Soda Ash TONS
This Week To Date JUICES FIRST EXPRESSED iofdsRef
POL %
LAST EXPRESSED
IPURITY Saids Ref. POL %
CLARIFIED
PURITY Saids Ref POL %
FILTERED PURITY Sofds Ref PURITY
ABSOLUTE (PREPARED CANE) Sofds Ref. P O L *
PURITY FACTOR
Γ I
This Week
I
To Dete SYRUP
pH
SOLIDS REF.
POL %
First
PURITY
FILTER Mixed J. Liming
Evap. Supply
SYRUP
CAKE
TONS
TONS POL
POL%
Moisture FIBER % X PREP. % SOLIDS CANE
This Week
I
To Date STRIKE S
FUEL A
Β
MASSECUITE eu. ft./ t. sua.
Sofds Ref PURITY
LOW GRADE
MASSECUITE MOL. PURITY Sofds Ref PURITY
J
?m
This Week
MOL. PURITY
MASSECUI ΓΕ eu. HJ
Sofds Ref. PURITY
MOL. PURITY
Remelt
Barrels
Purity
v#n Π
t
To Date F I N A L M O L A S S ES TONS
TONS SOLIDS REF.
TONS SUCROSE
SOLIDS REF.
TONS POL
POL %
Sucrose %
Per 100 Solids Ref. Sucrose Pol Pur RJSJAah Purity tfeleeded) R.S. ASH
HSJ Cond.
This Week To Date SUGAR
MANUFACTURED
TONS COM'L
TONS 96 DA
FINAL MOLASSES T O TERMIN AL
This Week Previously To Date
Fig.
1-2. Weekly factory report.
Moisture DET*N % FACTOR
TONS
BRIX 1 - 1
TONS 85 BRIX
Purity Above Expected
14 RECOVERY AND LOSSES REPORT STOCK
O N
Solids Ref %
H A N D
Purity
Pol %
C u Ft
W t Per C u Ft
X X X
X X X
Tons Solids R e f
Tons
Tons Pol
1 Juice 2 Syrup 3 Molasses 4 Remelt 5 Massecuite 6 7 8 Totals a n d A v e r a g e s RECOVERY
O N
STOCK
9 Pol
m
i(»- ) Tons
Tons Solids R e f
Solids R e f %
Pol %
Purity
Solids R e f %
Pol %
Tons
Tons Solids R e f
Tons Pol
10 Sugar, C o m m e r c i a l 1 1 Molasses FINAL
MOLASSES
12 W e i g h e d For C r o p / P e r i o d 13 A v a i l a b l e in Procoss This D a t e 14 Subtotal ι 15 In Process Beginning of C r o p / P e r i o d 16 Total D u e For These
Weeks Tons
BALANCE ENTERED I N T O M A N U F A C T U R E : 17 Cane, N e t 18 M i x e d Juice 19 A v a i l a b l e in Process Beginning ο
Crop/Period
X X X X X X
X X X
X X X
X X X Tons Pol In C a n e Out In
X X X
Pol %
X X X
2 0 C o m m e r c i a l Sugar P r o d u c e d 21 A v a i l a b l e in Process This D a t e 22 Loss in Manufacture; 23 Totals LOSSES
X X X X X X X X X
X X X
X X X X X X X X X
X X X X X X X X X
Tons
Pol %
Tons Pol in Mixed Juice Out In X X X
X X X X X X X X X
X X X X X X X X X
X X X X X X X X X Total Tons Pol
7.
24 Bagasse
%
Boiling House Tons Pol
X X X
X X X
25 26 27 28 29
Filter C a k e Molasses Undetermined Sum or T o t a l
X X X X X X
X X X X X X
RECOVERY 3 0 Total a n d Boiling H o u s e
m
31 A v a i l a b l e , s - j - m
*(i- ) i(s-m)
_
( I
) )
X X X
32 A c t u a l % of C a l c u l a t e d
X X X Period
GENERAL
To Date
33 Tons Sugar Produced a n d A v a i l a b l e 34 Tons 9 6 D A Sugar Produced a n d A v a i l a b l e
AMRAC FORM 1907 3/76
R & L Report N o .
.Factory .
For
the_
.Weeks
ending.
_I9 _Chemist
Fig.
1-3. Recovery and losses report.
15
Chapter 2
CANE CLEANING The essential steps in effective cane cleaning of pushrake harvested whole stalk cane are: 1. 2. 3· 4.
Thinning Rock, gravel and sand removal Washing Fibrous trash removal
A schematic diagram of a typical cleaning plant is shown in Figure 2-1. THINNING A thin mat of cane, 2 to 3 stalks thick, is necessary to obtain good cleaning. Thinning action is usually effected with a carding drum over which the cane passes. The drum, fitted with tines 30 to 40 cm (12 to 16 in.) in length, is set somewhat above and beyond the carrier and turns in the direction of cane flow at about 300 m (1000 ft) per minute peripheral speed. Carding is reasonably good if the cane is not presented in large bundles, some of which tend to be thrown over the top of the drum without carding. Partial preliminary load distribution is therefore necessary. An alternative to the carding drum is a tumbling conveyor. This is of pipe-slat construction hanging in a catenary arc of 55°. The angle is such that cane in some depth starts tumbling back near the top of the conveyor and will only pass over the top when the blanket is thin enough to give a low center of gravity. This type of action becomes less effective with unburned cane where tops and leaves bind the cane together. ROCK, GRAVEL AND SAND REMOVAL Rock, gravel and sand are the most deleterious materials in rakeharvested cane. Only by an hydraulic bath can an acceptable separation be made. Two types of equipment are suitable. In a velocity bath, a uniform rapid flow of water carries the cane in a flat arc through the bath to a drag-slat conveyor. Rock, gravel and sand (also tramp iron), being of higher density, drop in a steeper arc through the water and fall onto a conveyor in front and at right angles to the cane conveyor and here are conveyed to a truck for disposal. The importance of a thin blanket of cane is apparent, as a bundle of cane will raft the heavier materials across with the cane. In a sink-float bath a slurry of mud and water is maintained at a density that the cane will float and the dense materials sink. Mud density
Fig.
2-1. Cane cleaner.
16
17 3
3
should average about 1200 k g / m (75 l b / f t ) . Cane fed into the bath is floated across by some flow of slurry assisted by a revolving dunking drum. The same kind of cane and rock conveyors as outlined previously are used, and, as in the case of the velocity baths, a thin cane blanket is necessary to prevent rafting of rocks on the cane. WASHING Washing takes place initially on a drag-slat conveyor coming out of the hydraulic bath. The cascade principle is used with a large volume of water added near the top of a solid deck conveyor cascading downward in a turbulent flooding flow sweeping the soil out through a slotted deck above the water line of the bath. The cascade conveyor commonly has an angle of 40° and is run at a speed averaging 50 m/min (150 ft/min). FIBROUS TRASH REMOVAL The quantity of fibrous trash - tops, leaves and ground trash - is reduced by means of detrashing rolls. There are two types, Olsen and pipe or collared rolls. Olsen rolls are the most effective. These are constructed in banks consisting of several pairs of rolls made up of rectangular plates welded to a shaft. The banks are set at an angle 45° to 50° and the cane slides down over them by gravity. On revolving countercurrent to the cane flow at speeds in a range of 120 to 200 rpm, the corners of the plates enmesh, drawing leafy material down through and discharging onto a belt below, while the cane slides to a conveyor. Water jets help to push the trash between the rolls. Banks of Olsen rolls are often separated by single pipe rolls. Collared rolls may be used both before and after the hydraulic gap. Collared rolls are pipe rolls with welded circular discs which enmesh. They are set in banks as with Olsen rolls, but at a lower angle, usually 30°. As the cane slides over the rolls, water jets force loose material through the rolls while the cane travels over them. The rolls turn with the flow of cane, thus propelling it. Speed is of the order of 60 to 180 rpm. Effective trash removal requires that the cane pass over the detrashing rolls at single stalk depth. Thus, the cane blanket must again be thinned after cascader washing. This is done by means of combing drums which not only spread out the cane but also distribute it to the separate banks of detrashing rolls. Combing drums, 4 to 10 in number, are set on an upward slope of 20° to 30°. They are of relatively large tip diameter up to 1.8 m (6 ft) and are fitted with enmeshing serrated rings of steel plate. Rotating in the direction of the cane flow at progressively increasing tip speeds, they propel the cane with a bouncing action. Spacing is such that short cane will fall between the drums whereas long cane goes over the top of the last drum. By arranging the chutes below, distribution to the sets of
18
detrashing rolls is made. Jets of water are used in additional washing of the cane. OPERATION Cane cleaning is the most difficult of all factory operations to control and maintain. The one major problem is that of keeping a continuous, thin blanket flow of cane. Success with that gives good cleaning. Without it, there is poor cleaning and lost time caused by jams and equipment failure. The initial dumping of large loads of cane on the feeder table is the cause of most of the problems in distribution. The only reasonably satisfactory solution to this is one in which the cane is raked from the feeder table to the first cane conveyor. This is best followed by double carding, one after the first carrier and one ahead of the hydraulic bath. Without a raking system, care must be taken in feeding from the table to the conveyor to give partial distribution. From the carding drum on, the key is continuous steady flow. The necessity of keeping the clean cane conveyor full too often leads to a stop-go operation in which the cane may be flowing only half the time. This means that the cane blanket averages double the intended and the cleaning efficiency suffers accordingly. Controls are set so that any failure in the cleaner automatically stops all equipment behind in order to prevent jams. However, the conveyor after the hydraulic bath should always be kept running if possible, long enough to clear the cane from the bath. Conveyor speeds through the cleaner should be high enough to provide a minimum depth of cane. Water distribution should be concentrated at the points of maximum effectiveness. These are the cascader, combing drums and detrashing rolls. The quantity at each point must be sufficient to give a complete flooding action, otherwise effectiveness is lost. It must also be remembered that high pressure sprays are only effective at the point of impact. CANE SALVAGER In any hydraulic bath there are always some cane stalks, usually short lengths, that reach the bottom and are discharged on the rock and sand conveyor. These are recovered in a cane salvager which is essentially a small-scale velocity bath. Such a device should recover most of the cane which might otherwise be lost. EFFICIENCY OF CLEANING The quantity of extraneous matter delivered to the factory with the cane varies widely depending on harvesting conditions. With dry weather harvesting, a good burn will eliminate most of the dry leafy material leaving only tops and some ground trash. Soil and rock pickup will then be
19
a minimum varying with the terrain. In wet weather harvesting, with little or no burn, most of the fibrous trash will remain and large quantities of soil and rock will be present. Often during poor harvesting conditions the extraneous material can exceed the quantity of cane. Most cleaners are designed to handle cane with an average amount of extraneous matter and will not be effective under extreme conditions. In considering cleaner efficiency, it is only possible to look at the order of magnitude involved and to separate irrigated regions (relatively dry) from unirrigated regions (relatively wet). Even so it is only possible to get a reasonable handle on fibrous trash. 1
Early data by the Experiment Station of the Hawaiian Sugar Planters 1 Association ^ showed that the cane plant after two years of growth would have produced the following: Average makeup of total dry weight produced in two years: % Tops Leaves Roots Stalks
5 24 2 69
Composition at time of harvesting: Moisture % Tops Leaves Roots Stalks
76 34 — 70
Fiber % 20 64 — 13
During the two years of growth, leaves fall off and become ground trash, some of which decomposes. So at the time of harvest the leafy matter is considerably reduced. With a good burn most of the dry leaves burn leaving little fibrous trash except for the tops. The following figures are indicative of various conditions. Fibrous Trash Delivered to Factory % of Gross Cane Burned Tops Leaves Total (av.)
Irrigated Unburned
3.5 0.5
5.0 11.0 12
Unirrigated Burned Unburned 5.0 8.0
7.0 18.0 20
To these figures must be added dead cane, which is considered to be trash, and stools (root structure). The quantity of soil and rocks can range from a low of 1.5% under dry conditions to over 20% for wet conditions.
20
The cleaners in general perform well in removing soil and rock, the efficiency being in the range of 95%. Removal of fibrous trash, however, is rarely over 50% and usually can be considered to average 40%. Essentially no dead cane is removed. Stools, for the most part, are taken out in the rock bath. Removal of soil and rock is the most important function of the cleaner, since those materials place high wear loads on the equipment. Fibrous trash contributes only to increasing the quantity of fiber to be processed. With a high return value placed on fiber for fuel it can be economically sound to return all fibrous trash to the extraction plant. In this case the fiber should be given a secondary treatment, essentially the same as in the rock bath of the cane cleaner, to remove soil and rock fragments. WATER REUSE Clean water consumption can be cut substantially by multiple use. Extremely muddy water cleans equally with clean water except for a small residual. Minimum clean water use could therefore be only on the detrashing rolls, and perhaps a final rinse. That water is then collected and used on the combing drums. Then that from the combing drums on the cascader and finally in the hydraulic bath. Additional steps may also be used. Clean water requirements are considered to be about 4000 liters/ metric ton (1000 gal./short ton) clean cane per hour. This can be reduced with more efficient application and more stages of reuse. WASTE DISPOSAL Rock, gravel and sand are sent to land fill disposal. Fibrous trash may be returned to the fields but is useful to salvage for fuel. Two methods are used. One is to truck it to an open area, spread it out and allow air drying. The dry material is returned to the bagasse storage area, reduced in a shredder and fed to the furnace with bagasse. Another method is to treat the trash in a rock bath to remove rock fragments and sand and return it to the mill for dewatering. Waste water is allowed to settle in ponds or Hydroseparator-type equipment, with the effluent returned to the fields and the settlings pumped to landfill areas. LOSSES Losses in cane cleaning are in two categories - mechanical, loss of cane with rocks and with fibrous trash, and sugar loss by washing out of juice.
21
The mechanical loss can be kept at a level of 1% or below by use of a cane salvager and by proper operation and maintenance of detrashing equipment. The loss of juice (pol) is dependent upon the damage inflicted on the cane in harvesting, loading, hauling and cleaning. A single stalk of cane, sharply cut, can be washed thoroughly with negligible pol loss. A damaged or roughly cut stalk loses pol proportionally to the extent of the damage. Accurate quantitative data on pol losses are extremely difficult to establish because of the wide variation in damage and the extraordinary difficulty in the representative sampling of cane. Furthermore, measurements made on a weight basis are of limited value since the weight of the cane varies in each step of treatment - increasing with accumulation of soil, decreasing with removal of soil, increasing with accumulation of water and decreasing with loss of water and juice. Comprehensive investigations of the magnitude of pol loss in cleaning were made by the Sugar Technology Department of the Hawaiian Sugar 1 Planters Association on the basis of pol per unit fiber measurements. Although there was wide variability in individual cases the average statistically significant loss was 3.3%. Adding mechanical losses to this brings the value to around 4%. In economic appraisals a round figure of 5% would be a reasonable magnitude. Of course this is only the loss in cleaning. In the field operations during harvesting, the losses are of about the same magnitude bringing the total attributable to rake harvesting to the 10% range. REFERENCES 1 2
Borden, Ralph, Hawaii Planters Record, 46 (1942) 191-238. Stewart, Guy R., Assoc. Hawaiian Sugar Tech., Reports (1929) 221-230.
23
Chapter 3
MILLING INTRODUCTION Milling is basically an exercise in materials separation. In the simplest concept, cane consists of a solid - fiber, and a liquid phase - juice, which must be separated before sugar can be produced. This is done in a milling tandem in which juice is expelled from the fiber by successive applications of pressure as the cane passes between pairs of rolls. The efficiency of juice separation is determined by: Number of squeezes Effective pressure Degree of cell rupture Drainage Physical properties of fiber The capacity of a tandem is determined by the ability of the rolls to accept the cane presented and transport it by friction between the rolls. This is called feedability. The capacity is governed by the quantity of fiber since the juice alone puts no load on the rolls. In practice, a milling plant is designed for a nominal capacity at a nominal juice recovery. That is, the tandem should accept the quantity of cane desired in a unit time and expel a targeted percent of the juice. The actual results obtained depend upon how the tandem is set, how it is operated and how it is maintained. By pressure alone, it is impossible to expel much over 90% of the juice from the fiber because at some point the solid and liquid phases essentially coalesce into a mass which extrudes forward. Therefore, in order to recover more of the juice, it is necessary to add water. The water mixes with the juice and a certain percentage of the diulted juice is expelled in the next pair of rolls. By repeating this process, it is possible to recover substantially all of the juice. A conventional milling plant comprises equipment to prepare the cane for milling, which may be rotary knives, shredders, or combinations of knives and shredders; and a series of four to six 3-roll mills. In some installations a 2-roll mill, or crusher, may precede the mills. A crusher is considered both a preparatory machine and a mill. Each mill is normally equipped with a feeding device consisting of one or two rolls which compress the fiber before presentation to the mill. In some cases, these feeding devices become essentially a part of a mill resulting in 4-roll and 5-roll mills.
24
CANE PREPARATION Efficient juice separation in a milling tandem requires good preparation of the cane, whether obtained in preparatory machines, such as shredders, or in the mills alone. Good preparation means release of a high percentage of juice from the cellular structure of the cane without reducing the fiber size to such an extent that it will not feed well when presented to the mill. An approximate measure of the release of juice is given by a factor called Displaceability Index which gives the amount of juice that can be replaced by water in a fixed period of time. Preparation is best performed at the beginning of a milling tandem, and the best equipment is a heavy-duty shredder which essentially strips the fiber from the cane, releasing the juice, but still retaining a relatively long fiber structure. Such preparation also gives the best feedability. A well-designed shredder with heavy hammers (20 kg) running at a tip velocity of 6000 meters (19000 ft) per minute, should give a Displaceability Index of close to 88%. The horsepower necessary is about 13.2 per metric ton (12 per short ton) cane per hour. Running at lower speeds reduces the preparation and the horsepower. At 4500 meters (15000 ft) per minute, the Displaceability Index would fall to around 85% and the horsepower to 9 per metric ton (8 per short ton) cane per hour. In general, increasing the Displaceability Index by 4% would give a 1% increase in extraction in the same milling tandem. This is approximately the equivalent of adding one mill to a tandem. Lighter shredders are more effective if installed after a crusher, which provides protection from rocks. Also, by crushing the fiber and removing some of the juice, the power requirement is reduced. Such a shredder operating at a tip velocity of 4500 meters (15000 ft) per minute should give a Displaceability Index of 75% at a power requirement of 3.3 horsepower per metric ton (3 per short ton) cane per hour. TWO-ROLL CRUSHERS Two-roll crushers, formerly common in milling tandems, are rarely part of new installations. They are nevertheless simple and useful adjuncts. They serve the three functions of cane preparation, juice extraction and improved feeding for the next mill. Crushers are commonly the first unit of the milling train, although heavy shredders are often installed ahead of crushers. A well-operated crusher should give 50% juice extraction. By a combination of cane preparation, blanket compression and dewatering of the fiber, the first mill following (without maceration) should give an additional 27% juice extraction, giving a total of 77% by dry crushing. Crushers with coarse 7.5 cm (3 in.) circumferential grooving have good feedability and good drainage. Hard-faced, Krajewski-type corrugations perform a good job of rock crushing, which is an important function in protecting shredder hammers and subsequent mills.
25
MILLS The three rolls of a conventional mill are arranged in a triangle so that the fiber is squeezed twice between the top roll and the feed roll and the top roll and the discharge roll. The rolls have cast iron, grooved shells mounted on steel shafts. Fiber passing between the top and feed roll is conducted over a turner plate to the discharge roll. The rolls are pinion driven from the top roll which is driven at a speed of 3 to 6 rpm by a turbine through a gear reduction system. The feed and discharge rolls are fixed, while the top roll is free to move up and down by means of an hydraulic pressure system. Cane is moved between mills by means of intermediate conveyors. They are generally rake or drag-slat type which carry the fiber to a fixed chute leading to the next mill. FEEDING DEVICES The capacity of a mill to handle cane is governed largely by the ability of the rolls to accept the feed. Improved feedability can be obtained by the use of devices which compress the fiber blanket presented to the rolls, thus increasing frictional grab. These include: Overfeed roll Underfeed roll Pressure feeder Two-roll feeder Overfeed Roll A roll on top of the blanket just ahead of the mill compresses the fiber and thus improves feeding. These rolls are usually of smaller diameter 75% that of the mill rolls and are fabricated from sheet steel fitted with longitudinal angle iron strips. Sometimes cast iron rolls with grooving are also used. The rolls are driven by sprocket and chain at peripheral speeds 10% or more above that of the mill rolls. Underfeed Roll The underfeed roll is generally preferred to the overfeed roll. It operates beneath the blanket, compressing it against the top roll, which improves feedability. Construction is similar to that of an overfeed roll and drive is usually by sprocket and chain. The diameter is generally smaller - 75% that of the mill rolls and the peripheral speed is the same or somewhat higher than the mill rolls. In some regions the underfeed roll has been increased to the same size as the mill rolls, enmeshes with the feed roll and is driven by a crown wheel off of the top roll. The result is a 4-roll mill which has good feedability but has the disadvantage of poor drainage. At high mill speeds, expressed juice which cannot drain over the feed roll flows over the top roll and must be carried away from the discharged bagasse by means of a scraper trough. With good design these mills perform well at high grinding rates and offer a relatively easy method of increasing the capacity of an existing tandem.
26
Pressure Feeder A pressure feeder has two feed rolls the same size as the mill rolls set at an angle like a crusher. The rolls discharge into a pressure chute leading to the mill. The bagasse is thus kept compressed between the feeder and the mill. The feeder rolls are geared together with pinions and are driven by the mill turbine at a speed averaging 30% more than the mill rolls. The pressure feeder originated in Australia in order to improve the feedability of bagasse from hot maceration baths. Since a feeder extracts up to 40% of the juice from the fiber, it is not a feeding aid in the true sense. In reality the device converts a 3-roll mill into a 5-roll mill. Juice expressed by the top roll of the feeder is collected in a tray with scraper arrangement, while juice from the bottom roll flows both foward and backward into a collection trough. Because of juice extraction, the 3-roll mill is presented with a drier feed permitting the use of lower work ratios in the mill. The capacity of the 5-roll mill is up to 25% greater than that of a 3-roll mill of the same size, but power requirements are correspondingly increased. Two-Roll Feeder The 2-roll feeder is similar to a pressure feeder without a pressure chute. This is a true feeding aid as it does not extract juice. It is chain driven, and the bagasse discharges into a closed chute leading to the mill. The device gives good compression and has the advantage of good drainage. Because of lower power requirement and less cost, the 2-roll feeder is generally preferred over pressure feeders in application to existing installations. GROOVING Mill rolls are grooved to improve feedability and provide drainage. Grooves are of three types: Circumferential Juice (Messchaert) Chevron Circumferential Grooves Cutting grooves around the roll gives a corrugated surface of increased area which has better gripping action because of the compression of the fiber against the walls of the V-shaped grooves. Since the bagasse at the bottom of the groove is not so highly compressed, a drainage channel for juice is formed. The surface area changes with the angle of the groove, the sharper the angle the greater the surface. Larger pitch increases the speed differential between the tip and the channel. This gives a grinding action which will open up cells. Although sharper grooves give more surface, the effective surface is not much affected by very sharp grooves (30-35°)
27
since little pressure reaches the bottom of the channel. This is useful, however, as better drainage results. Sharper grooves are more susceptible to damage by rocks and tramp iron so angles less than 35° are not used. Normally, 45° is most practicable where damage potential is severe. Other conditions being the same, juice extraction decreases with increase in pitch of grooving. This is the composite effect of several factors, including reduced effective pressure, more groove wear caused by higher differential speed and more slippage. As preparation increases down the milling tandem, finer grooving is used. Good standard practice on the last mill is 1.25 cm (1/2 in.) pitch. With a good shredder, such grooving can be used throughout. With poor preparation, 2.5 cm (1 in.) pitch is effective. In the absence of a shredder, 5 cm (2 in.) pitch is common in the beginning mills. Juice (Messchaert) Grooves Juice grooves are narrow channels cut deeper than the circumferential grooves to give better drainage. They are normally not over 0.6 cm (1/4 in.) wide to prevent much fiber being compressed into them, and are of a depth sufficient to carry away the juice. This averages about 2.5 cm (1 in.). The pitch is also a function of the volume of juice and is normally 7.5 cm (3 in.). The juice grooves must be kept clean by scrapers. Wear causes problems by enlarging the grooves and letting excess bagasse enter with the juice. Narrow and shallow grooves are desirable for this reason. Juice grooves are most effective on the feed roll where juice flow is greater. They may also be used on the discharge roll but are not as efficient because they reduce the effective pressure area of the roll and increase the incidence of metal lost by fracturing of the circumferential groove adjacent to the juice groove. Chevron Grooves Chevron-shaped grooves are cut lengthwise through the circumferential grooving as an aid to feeding. They are nominally cut to one half the depth of the grooving and at a pitch of 25 cm (10 in.). Chevron grooves are effective only on the be detrimental on the discharge roll. Since effective pressure surface of the rolls and mixed juice, more efficient operation of chevron grooves by arcing maintenance of the
feed and top rolls and would such grooving decreases the causes excessive bagasse in a mill is obtained without circumferential grooving.
OPERATION Effective operation of a milling station involves proper setting of the individual mills; close control of the operational variables, chief of which are fiber feed rate, speed, hydraulic loading and imbibition; and finally, good maintenance.
28
Mill Settings The setting of a mill requires three measurements - the distance between the top roll and the feed roll, the distance between the top roll and the discharge roll and the distance between the top roll and the turner plate. The weight of fiber passing through the mill per unit time is the basis for calculating the proper setting. The following concepts are used in arriving at the initial setting. In the course of operation, changes from the initial are often necessary because of the many unknown variables involved. Work Opening: The work opening is the average distance between the rolls as measured on the common axial plane. It is calculated from the mean diameter of the two rolls, which is the diameter half way between the tip and bottom of the grooves. A fixed amount is added to the measured opening to allow for the lift of the top roll when floating in operation under the calculated fiber load. Mill Ratio: The mill ratio is the ratio of the feed and discharge settings. The mill ratio adopted is based upon the calculated discharge work opening which is determined by the fiber rate, size of the rolls, speed of the rolls and fiber content of the discharged bagasse. Once the discharge work opening is chosen, the feed work opening is chosen depending upon trash content of the cane, preparation, drainage and imbibition rate. The mill ratio should be kept at the minimum practicable. Nominal values for mill ratios are 2.5 for the first mill and 2.0 for mills thereafter. With 5-roll mills, ratios as low as 1.5 are possible. Escribed Volume: The escribed volume is the product of the roll length, the work opening and the roll peripheral speed. Fiber Index: Fiber index is the weight of fiber per unit escribed volume of bagasse from the discharge roll of the mill. It is a measure of blanket thickness and compression. Although it would appear that a high value of fiber index is desirable, there is an optimum value beyond which little change occurs for a given mill. The value increases from the first mill to 3 the last. Nominal maximum values of fiber index range from 560 k g / m 3 3 3 1 (35 l b / f t ) for a first mill to 880 k g / m (55 l b / f t ) for a fifth mill. Usual operating values are lower. Fractional Fiber Content: The fractional fiber content expresses the fiber on a weight basis, that is, the weight of fiber per unit weight of bagasse discharged. An indication of nominal values is shown by data from Mauritius. 2
29
Fractional fiber content of discharged bagasse % Crusher (2-roll) Mill 1 2 3 4 5
0.25 0.37 0.41 0.46 0.49 0.52 3
3
3
Using values of 1530 k g / m (95.5 l b / f t ) for fiber and 1028 k g / m (64.2 3 l b / f t ) for juice, these figures are considerably lower than the maximum values shown under Fiber Index. Reabsorption Factor; The reabsorption factor is a measure of the extrusion that occurs in a mill. Extrusion begins when the speed of the mills becomes high enough that the maximum pressure on the blanket of fiber begins to move ahead of the line of minimum distance between the rolls. Material is then propelled forward at a speed exceeding that of the rolls. A better term to describe the phenomenon is forward slip. When this occurs, the volume of the bagasse exceeds the calculated escribed volume. The increase in volume is caused largely by an increase in the amount of juice, since it moves forward preferentially to the fiber. Juice extraction decreases, therefore, as the amount of forward slip increases. The reabsorption factor is: No-void bagasse volume Escribed volume The factor increases with roll speed, increases with fiber rate and increases with a smaller discharge work opening. It decreases with the fineness of the fiber. In setting a mill this factor must be kept as low as possible. Because of the varying fiber load, the actual value of the factor can only be approximated. The optimum range for the factor, however, is 1.3 to 1.4. Procedures: Procedures adopted for setting mills, although based primarily on fiber throughput, must take into consideration many other factors - most important of which are: Mill strength Hydraulic loading Roll grooving - size and condition Cane quality - trash and soil content Cane preparation - Displaceability Index Drainage Imbibition - quantity of water and temperature The first step, however, is to calculate a tentative work opening for
30
the discharge roll. This is based on the fiber rate using a nominal Fiber Index as discussed earlier. A mill speed is selected which will keep the reabsorption factor at a reasonable level (below 1.5). The work opening for the feed roll is then chosen keeping the mill ratio as low as possible. For mills after the first, a nominal 2.0 mill ratio is standard. With a 5-roll mill, this can be lowered to 1.5. Better cane preparation will permit a lower mill ratio, as will more efficient feeding devices. Since feedability is impaired by hot maceration, higher mill ratios are necessary. In general, also, higher hydraulic loading, poor drainage and fine grooving require higher mill ratios. Turner Plate Settings: Turner plate settings are the ratio between the feed work opening and the opening at the toe of the plate. The opening at the toe of the plate is equal to the set opening plus one-half depth of the top roll groove plus hydraulic lift. There is only limited experimental evidence on which to base settings, so general experience is used. If the plate is set too high, the top roll will be lifted by fiber passing over the plate. This will increase power, increase wear and lower extraction. Also, the mill capacity will be decreased. If the plate is too low, there will be too high an angle of contact with the discharge roll which can cause choking. Experience indicates that an average fiber rate of 160 kg/m^ (10 lb/ft^) on preliminary mills and 240 kg/m^ (15 lb/ft^) on the last mill gives good performance. The ratio of toe of the plate to the feed opening is normally 1.8 for finely divided bagasse. With coarse bagasse, this ratio would decrease to 1.2. Maceration In the conventional compound maceration system, water is distributed laterally to the bagasse on the intermediate carrier feeding the last mill. Last expressed juice is then returned to the carrier to the next to last mill and so on up to the second mill. Juice from the second mill mixed with the dry crushing juice from the crusher and first mill constitute mixed juice. Although bagasse fiber will absorb about 650% its weight of liquid, it is not advantageous to use enough water to bring the liquid content to this point on the last mill. A t about 250% water on fiber, the effect on extracation levels off, so that steam and evaporator capacity requirements dictate little economic advantage of using more water. The water should be applied at as high a temperature possible without causing a feedability problem. The fiber becomes plastic at higher temperatures so is more easily compressed giving higher juice expression. Higher temperature also makes the fiber slicker so it does not feed as well. Mixing of liquid with the bagasse takes place mainly at the entrance to the mill where the increasing pressure forces the liquid backward through the fiber. Less lateral flow occurs, so the important consideration in applying the maceration is to insure uniformity across the width of the bagasse blanket.
31
Hydraulic Loading Juice extraction increases with increase in pressure applied to the top roll. The pressure that can be applied, however, is limited by the mechanical strength of the mill. Also, feedability decreases at higher pressures and power necessary increases substantially. The optimum pressure, therefore, is that which permits the top roll to float at the necessary feedability. The actual loading used on a given mill can only be determined by experience. General standards call for a nominal pressure of 14 mt/m (50 t / f t ) on a 2-roll crusher to 21 mt/m (75 t / f t ) on the last mill. Speed Other conditions, such as preparation and pressure, being the same, juice extraction decreases with increased mill peripheral speed. The reason for this is that extrusion (reabsorption, forward slip), meaning that juice is expelled with the fiber, is a function of mill speed. It is desirable, therefore, to operate at as low a speed possible, commensurate with the necessary fiber rate. The finer the preparation, the slower the optimum speed. This is because of better feedability. Also, there is the factor of cell rupture in the mill which, because of closer mill settings, increases with speed. It is standard practice to run all mills in a tandem at close to the same peripheral speed with a slight increase toward the last mill to improve feedability. General Effects The general effects of operating conditions on milling efficiency may be summarized as follows: Pressure Increase Increases juice expelled Increases power required Decreases feedability Roll Speed Increase Decreases juice expelled Increases feedability Cell Rupture Increase Increases juice expelled Increases feedability Imbibition Increase Increases juice recovery Decreases feedability Imbibition Temperature Increase Increases juice recovery Decreases feedability
32
Mill and Turner Plate Settings Decrease Requires mill speed increase to maintain throughput Increases cell rupture Increases power required CONTROL The primary control figures on milling are pol, moisture and fiber in final bagasse. These, of course, give only the total result and it is important to know the performance of individual mills. This requires bagasse analyses from each mill. But constant variation in the fiber load, because of trash, renders such figures of little value unless taken on a statistically significant basis. Therefore, they are commonly not justified from the cost standpoint. As a result, mill settings are estimated and then adjusted on the basis of experience. Juice Density Curves Useful guides to the behavior of each mill are juice density curves. These are made by an analysis of the refractometer solids in samples of juice from the feed roll, discharge roll and total from each mill. Discharge roll figures are plotted giving a curve which for a properly functioning tandem would appear as in Figure 3-1. Typically, the crusher value is lower than the first mill because of the extraneous water on the cane from the cleaner. The figures then fall in a smooth curve to the last mill. Since the refractometer solids content of the juice from the discharge roll of the last mill is about equal to that of the juice remaining in the bagasse, the lower the figure, the better the extraction. The actual value will depend upon the cane preparation and maceration efficiency as well as the performance of the individual mill. The individual values on the curve indicate the total performance up to that point. A deviation from a smooth curve is evidence of malperformance in that area. Usually the deviation is toward making the curve more horizontal and means that the preceding mill is to blame, either by total performance of the mill or poor extraction at the feed roll permitting lower density juice from the discharge roll. Rarely it might also be caused by very high extraction of the feed roll of the succeeding mill. A break, making the curve more vertical, is sometimes encountered. This could be caused by malfunctioning of the mill preceding the one where the break occurs either in the total extraction or poor extraction at the feed roll allowing low density juice from the discharge roll. It could also be the result of low extraction from the feed roll of the mill succeeding the break. Juice density curves are only significant on a routine basis to show deviations from previous curves. An individual curve, standing alone, is of little value, mainly because of the idiosyncrasies of individual mills. Once a curve with a break shows up, the test should be repeated to make sure that a real condition exists.
33
Fig. 3-1. Mill juice curve. The relative extraction of the feed roll and discharge roll can also be calculated from the juice density figures, as an example: Refractometer solids Feedroll Discharge roll Total juice
2.0 4.0 2.0
Calculating by the rectangle method: 2.0
1.5 2.5
4.0
0.5 2.0
34
1.5/2.0 χ 100 = 75% of the juice extracted by the feed roll and 25% by the discharge roll. Such figures would be indicative of good mill performance. REFERENCES 1 2
Van Hengel, A . and Douwes Dekker, K., Some Notes on the Setting and Operation of Mills, Proc. So. African Sugar Tech. Assn., 32nd Congress, 1958, pp. 57-67. Van Hengel, Α . , Some Additional Notes on Mill Settings, So. African Sugar J., 1958, pp. 855-861.
35
Chapter 4
DIFFUSION* INTRODUCTION In the separation of juice from the fiber in sugar cane by the process called diffusion, the juice is displaced from disintegrated cane by the countercurrent flow of water rather than being expelled by pressure as in milling. The unit operations involved are three: Cane Preparation Juice Displacement (Diffusion) Bagasse Dewatering CANE PREPARATION Cane preparation for effective juice displacement requires size reduction to give a compact permeable bed and rupture of close to 94% of the juice storage cells. This should be brought about with a minimum of grinding of the fiber and retaining a fiber bundle length of 10 to 15 cm (4-6 in.). The cell tissue should be stripped from the fiber bundles producing a mixture of shreds and pith tissue. Such preparation will yield a compact yet permeable bed for countercurrent extraction. The equipment most suited for this work is swing hammer shredders. All experimental investigations have shown that good preparation is best achieved in two steps. Shredders developed specifically for this purpose are the Silver Buster and the Silver Fiberizer. The Buster has 20 kg hammers rotating at about HOOrpm, working over an open grid screen grate with anvils between the hammers. Feed control to the machine is provided by a set of variable speed feed rolls. As distinct from conventional shredders, where the hammers are working over a solid ridged plate, cane is extruded through the screen so that no unshredded stalk can pass. Size reduction to a length of 5 to 10 cm (2-4 in.) is obtained with cell rupture up to 87%. Secondary preparation takes place in the Fiberizer which is more like a conventional shredder in that the hammers are working over a ridged 90° arc segment. Here, fiber length is reduced only to a limited extent and cell rupture is increased to 94%.
This discussion applies specifically to the Silver Ring Diffuser. However, the general principles involved are the same for any type of diffusion equipment.
36
JUICE DISPLACEMENT Displacement of juice by water in the prepared cane is attained by a countercurrent liquid flow in an advancing front similar to the phenomenon of saturated flow in porous media. The advancing liquid operates like pistons through the ruptured cell modules replacing the juice. Any macro breaks in the bed lead to piping, channelling and by-pass of liquid from the displacement process. Any mixing in the system reduces the effective countercurrent flow. Any pressing or squeezing will expel juice concurrently as well as counter, causing mixing. The ideal system, therefore, is an undisturbed bed of fiber and juice and plug type liquid flow. In the Silver Ring Diffuser, this mechanism is obtained with a rotating annular bed of prepared cane through which the liquid is pumped in 18 stages (Fig. 4-1). Cane is made to form a continuous bed about 1.5 m (5 ft) in depth, which is maintained for some 335 angular degrees. At this point, the bed is discharged vertically by means of multiple screws. Water is added from a distributor above the bed and several degrees ahead of the discharge screws. The water flows through the moving bed and into a compartment below and behind the distributor. From there, the liquid is pumped forward to the third distributor, passes through the cane bed and collects in the next juice compartment. Juice from bagasse dewatering enters the distributor just forward of the water distributor and flows into the second collecting tank. Flow continues forward to the point at which the cane enters. Here the juice is cycled ahead as first pass juice. On passing through the newly deposited cane, it picks up some suspended matter which may be removed by circulating through the bed. This second pass is called recycle, and pumping is in the direction of cane travel. The effluent recycle juice is analogous to mixed juice from a miling tandem and is sent to the boiling house. With large amounts of soil entering with the cane, this recycling can cause plugging by depositing soil on top of the bed, so should not be used. Efficiency of juice displacement is reduced by some channelling in the bed and by mixing at the overlapping boundaries of the stages in the bed. This is compensated for, to a large degree, by the use of more than the theoretical number of stages, and by application of more water. A typical extraction pattern obtains however. This is shown in Figure 4-2 illustrating the gradients occurring in routine diffuser operation. A rapid drop in pol takes place initially as juice is displaced. Then the decrease becomes more gradual and, finally, becomes asymptotic near the discharge point, being "supported" by the juice returned from bagasse dewatering. Factors which govern the forms of the gradient are: Cane Quality Cell Rupture Bed Depth Speed of Rotation Bed Permeability Quantity of Water Quantity and Quality of Juice Return From Dewatering
Fig.
4-1. Diffuser flow diagram.
37
Fig.
4-2. Diffuser gradients.
38
39
Operational conditions can be adjusted to give the optimum results consistent with the objectives. Pol extraction increases with increase in cell rupture and quantity of water. There is an optimum bed depth and speed of rotation, depending upon the cane rate and bed permeability. It is basic that the capacity of the diffuser is governed by the rate of the gravity flow of liquid through the bed. Cane quality, type of preparation, bed depth and dewatering juice quality are the variables affecting this flow rate. BAGASSE DEWATERING Following juice displacement, the bagasse must be dewatered. This step also is the final stage in the extraction process as some pol in unbroken cells and in the residual liquid is recovered and returned to the diffuser. Dewatering requires some sort of expelling mechanism, and most of the common types of machines are adaptable to such use. First thought usually starts with conventional 3-roll cane mills which are in general use for this purpose. In operating a mill after a diffuser, consideration must be given to the difference in properties between mill bagasse and diffuser bagasse. Firstly, it has a higher liquid content, averaging 650% on fiber. This compares with about half this for bagasse entering the last mill of a milling train at 200% imbibition on fiber. In fact, diffuser bagasse has about the same relation between liquid and fiber as whole cane. Special attention must, therefore, be given to drainage. Secondly, the bagasse is at a higher temperature than normal for mills. It is, therefore, more plastic and slicker, giving a lower coefficient of friction with the rolls. This results in poor feedability. Greater attention must thus be given to feeding devices and maintenance of mill grooving. Thirdly, diffuser bagasse is in a relatively fine state of subdivision. Meeting these conditions effectively is a 3-roll mill fitted with a heavy-duty pressure feeder. The settings should be more like that of a first mill rather than that of a last mill because of the high liquid-fiber ratio. The characteristics of diffuser bagasse are particularly favorable to dewatering in expeller type machines such as the screw press. In these, high liquid content and high temperature favor feeding and reduced friction in contrast to a roller mill. On the unfavorable side are the relatively high power requirements and high maintenance costs. PRESS JUICE TREATMENT Juice from the dewatering equipment is returned to the diffuser in order to recover some of the pol. Insoluble material in this juice may plug the bed and cause diffuser flooding, so treatment is necessary. Investigational work has shown that plugging is caused mainly by particles 40 microns or less in size. These fine particles form a layer of low
40
permeability by filling the interstices near the top of the bed. Juice percolation through the bed slows and juice accumulates on the top of the bed, a condition called flooding. Juice below the impervious layer drains away, leaving a porous bed which results in poor juice contact. A temporary flow can be induced by mechanically breaking up the layer, but a short time later, the layer will form again farther down in the bed (Fig. 4-3). It has been established that the particles that will cause plugging settle rapidly in the low density juice, so that simple gravity settling produces a relatively non-plugging juice. A conventional clarifier will normally give a good juice without chemical treatment with a retention time of about half that of lime defecation.
Fig. 4-3.
MINUTES Press juice percolation through prepared cane bed.
41
The underflow from a press juice clarifier is difficult to handle on a bottom feed rotary vacuum filter. The cake tends to form irregularly and slough off on the feed side. Top feed filters perform satisfactorily. Horizontal filters perform well and specially designed bottom feed rotary machines with increased hydraulic capacity are also functional if bagacillo is added. OPERATION At rated capacity, the Silver diffuser is designed for a target bagasse pol of 1.0 at a draft of 100 (juice % cane). A cell rupture in prepared cane of close to 94% is necessary. Based upon an average percolation rate of 2 2 250 liters/m (6.1 g a l . / f t ) at an optimum bed depth of 1.5 m (4.9 ft), the diffuser speed would be 1.5 revolutions per hour. Under these conditions, the fiber retention time in the diffuser is 37 minutes. The juice, however, 1 would have an average retention time of only half this (Fig. 4-4). Permeability, Bed Depth and Diffuser Speed When operated under design conditions (see flow diagram, Fig. 4-1), juice from a distributor percolates through the moving bed and drains into the next aft collection tank below. Deviation from standard conditions in permeability, bed depth and diffuser speed would have the following effects: 1.
If the bed permeability is lower, some juice is carried beyond the next tank. This mixes the countercurrent flow, moving richer juice toward the discharge end and reducing extraction. Corrective action may be: a. b.
reduce bed depth reduce diffuser speed
Both result in lower throughput. 2.
If the bed permeability is greater, some juice will recirculate and countercurrent flow will be disturbed in the forward direction. This, of course, is a good situation and can be corrected by increasing bed depth or diffuser speed, both of which result in increased throughput.
3.
If the bed depth is greater, the effect is similar to No. 1, low permeability.
4.
If the bed depth is lower, the effect is similar to No. 2.
5.
If the diffuser speed is higher, the effect is similar to No. 1.
6.
If the diffuser speed is lower, the effect is similar to No. 2.
In order to compensate for variation in these conditions, the diffuser has enough extra stages in its design to allow considerable intermixing of flows. Changes of the order of 15%, therefore, have little effect on performance. Changes of greater magnitude will show up in reduced efficiency.
42
REF SOL. FIBER INSOL SOLIDS WATER
REF SOL 2 . 5 4 FIBER 46.3 INSOL. SOL. 4 . 0 MOIST 47.2 POL 1.24 PURITY 48.4
D I F F U S E R FLOW BALANCE PREPARED CANE T/hr. 100 DIFFUSER JUICE % CANE 1 0 0 EXTRACTION % 97
PRESS R E T U R N CAKE 2 . 0 T/hr
Fig.
4-4. Diffuser flow balance.
43
Cell Rupture Cell rupture (Displaceability Index) below 94% gives higher bagasse pol because of the slow rate of diffusion of sugar out of unopened cells. Rich juice will, therefore, persist until the dewatering process, giving higher pol in the returned juice, thus raising the pol level at the discharge end of the diffuser. Good cell rupture is the single most important factor in reaching high extraction. This is obtained by careful maintenance of the shredder hammers and grids. As the surfaces of these wear and clearance increases, the horsepower required to drive the machines increases. Thus, horsepower is a good indicator of the type of preparation being obtained. Periodic analysis for Displaceability Index is, of course, necessary. Draft and Dilution Draft is extracted juice plus dilution water as a percentage on cane, so the 100 draft figure corresponds to a dilution of about 17% on absolute juice for average cane. With a good countercurrent flow pattern and high cell rupture, bagasse pol of 1.0 can be obtained with dilution rates below 17%. Under less favorable operating conditions, it is well to increase the dilution and hence increase draft to the extent that the evaporator station will handle the load. The inventory of liquid in the diffuser must be maintained at such a level that the bed is saturated. That means that the liquid level should be the same as the bed level. If the liquid level is too high, above-surface flow will result in unwanted mixing. If the level falls below the bed surface, plug-type flow will no longer be possible and air filled channels will result in poor liquid to fiber contact and lowered efficiency of extraction. The bed should never be allowed to drain completely because air pockets which form will remain, even if the bed is flooded again. In order to maintain a saturated bed, the quantity of liquid must be of 1 the order of 1200% on fiber (Fig. 4-5). This amount of liquid is maintained by sending to the diffuser all of the return from the dewatering unit plus fresh water to make up the total draft. Return press juice from the dewatering unit at 100 draft will be about 75%, requiring about 25% fresh water. The flow of these streams must be steady and as uniform as possible. When the diffuser is stopped, juice flow is automatically diverted so that it is pumped forward in complete countercurrent flow only at the discharge end but recirculates on itself with only a small forward flow at the feed end. This is to maintain a saturated bed. Shortly after a diffuser stop, press juice will stop and, with automatic control, fresh water will increase some four-fold to supply the running diffuser needs. The water should be kept on for a short time, but on stops of more than a few minutes, the water flow should be reduced to one-half rate after about 30 minutes. Press Return All juice from dewatering the bagasse is returned to the diffuser, just as last expressed juice is returned to the tandem in milling. The return must be relatively free of suspended solids and must have a minimum
Fig.
4-5. Juice flow as a percent of fiber.
44
45
temperature of 85° C. Quality with respect to bed-plugging material can be tested periodically simply by observing the flow of a sample through a 100-mesh screen. The juice should pass through without any holdup on the screen. The press juice distributor should be inspected periodically to ensure that there is no buildup of mud on the bottom. Although treatment, other than settling, is normally not necessary, it is advisable to add sufficient lime to bring the pH above 6 in order to reduce corrosion of the equipment. Settlings from the press juice clarifier need only rough filtration in order to return the juice to the press juice heating tank. Since the pol is low at this point, losses in cake are very small. Temperature Diffuser operation should be carried out under conditions in which microorganisms will not grow. A safe temperature is 70° C. This is held by keeping the press return and fresh water temperature at above 80° C and heating the recycled juice at the front of the diffuser to the same level. Growth of microorganisms in the diffuser will not only destroy sugar, but gum formation can seriously impair the permeability of the bed and cause flooding. It is well not to maintain too high a temperature as higher sugar losses through inversion will occur. A t 70° C, the average inversion loss through the diffuser, with unlimed juice (pH 5.2-5.5), is 0.14%. Also, lower temperatures give less color development and less solubilization of fiber components. This latter factor is responsible for the increase in nonsugar dissolved solids in diffusion with a consequent decrease in juice purity. Bagasse Dewatering The dewatering unit must deliver a constant flow of return juice to the diffuser and reduce the liquid content of the diffuser discharge from about 85% to less than 50%. The principal consideration is the maintenance of good feedability. In the case of a mill, the grooving must be kept in condition by frequent arcing. DIFFUSER FLOODING Failure of juice to percolate rapidly enough through the bed, resulting in accumulation of liquid on top of the bed and, particularly, in front of the discharge screws, is the principal problem encountered in operation. As indicated previously, it is usually caused by plugging with finely divided particles of soil. The first indication of flooding is often seen in the irregular discharge of bagasse from the screws. Liquid accumulating at the base of the screw by poor draining of the bagasse prevents pickup by the screws. Bagasse then, builds up until it reaches a high enough level that drainage has made
46
it dry enough to be picked up. Large quantities are then picked up until the bagasse level falls and pick up ceases. In cases of serious flooding, the bagasse may become so high that enough is picked up to overload the drive motors on the screws. The most common cause of flooding is return of press juice with substantial quantities of fine suspended particles. The remedy is proper control of overflow from the press return clarifier. Usually, the heated press juice settles rapidly and presents no problem. Juice from sour and deteriorated cane, however, may settle poorly. Moderate chemical treatment such as lime or poly electrolytes has not proven to be generally effective. Of course, extensive treatment such as high liming, followed by phosphatation, which give a clear juice, will be effective but expensive. The options available, therefore, are to slow down the diffuser or to discard some of the press return replacing it with clear water. Another source of flooding is at the feed end, from recycling the juice as previously indicated. This situation can be helped by stopping the recycle and sending first pass juice to the boiling house. The juice will not be as clear, but should not cause any difficulty in clarification. A third cause may be just too much soil coming in with the cane, so the whole bed is of low permeability. The obvious solution to this is better cane cleaning. If this is not possible, the only thing to do is slow down the cane rate. An uncommon cause of flooding is growth of microorganisms in the diffuser. This can only occur when temperatures fall below 70° C, permitting multiplication. Microorganism growth can give gum formation which can almost gelatinize the cane bed and block it completely. Devices which rake up the diffuser bed periodically have been used to aid in percolation. Such action does enable juice to go through the bed but poor extraction is the consequence. CONTROL Basic control parameters for a diffuser are: 1. 2. 3. 4.
Cane feed rate Displaceability index Diffuser speed - bed depth Dilution water rate
No sensing device has been developed for the governing element which determines how these parameters, with the exception of displaceability index, will be set, namely the bed permeability. This condition is judged solely by observation and operating set points plugged in accordingly. Cane feed is weighed on a belt scale, the feed to which is controlled by the chest pressure on the turbine driving the Buster. Feed to the Buster is controlled by the Buster feed roll. Cane in the chute to the Buster is
47
sensed by a level arm riding on the cane which actuates a switch to the carrier drive motor. Speed of the diffuser is set by the cane feed rate so as to maintain a minimum bed depth of 1 m and a maximum of 1.5 m. Displaceability index can only be determined by periodic sampling and laboratory analysis. However, a good idea of the degree of preparation can be had by visual observation. Also when the hammers become worn the horsepower of the drive increases. This means poor preparation and thus a low displaceability measurement. Rate of dilution water under standard conditions is that necessary for 100 draft. This means that the total of juice returned from dewatering bagasse (press juice) plus dilution water is equal to the cane rate. Since the return juice normally is around 75% of the cane, the water added is about 25% of the cane. It is often the case that the evaporator capacity is limiting. Under this condition, the quantity of water is that which the evaporator will handle, then the draft may be less than 100. Likewise, with ample evaporator capacity, additional water can be used to lower the bagasse pol. Water temperature should be 80° C. REFERENCES 1
Payne, J. H., Cane Diffusion - The Displacement Process in Principle and Practice, Proc. Intern. Soc. Sugar Cane Tech., 1968, pp. 103-121.
49
Chapter 5
BAGASSE MOISTURE Although the moisture content of final bagasse is governed primarily by the performance of the last mill, the composition of the material entering this mill has a significant effect. This is determined by the quality of the entering cane and the extraction in the preceding mills. Assume cane entering the tandem at 15% fiber and 15% dissolved solids, the entering moisture is 70%. Juice extracted by dry crushing in a 2-roll crusher and a 3-roll mill should be 77%. Then, the residual juice in first mill bagasse would be: (100 - 77) χ 0.85 χ 100
_
(100 - 77) χ 0.85+ 15
" ™ ™
Λ
and the moisture would be: 70 χ 0.23 χ 100 15 + (15 χ 0.23) + (70 χ 0.23)
,
i 4 Rf O 4
b
,
b
*
leaving fiber of
43.4%
moisture plus fiber
90.0%
and dissolved solids of 10.0% In the following mills of the tandem, the application of the countercurrent maceration system results in the replacement of most of the juice with water. If approximately the same relationship of solid to liquid (fiber to juice) is maintained in subsequent mills as in the first, then bagasse leaving the last mill with a juice of 3.0 dissolved solids would have: dissolved solids = 56.5 χ 0.03 = 1.7% and moisture = 56.6 - 1.7 = 54.9% Thus, the moisture content would increase as the dissolved solids decrease. In order to reduce the moisture, it is necessary to increase the fiber content progressively down the tandem. The mill settings must, therefore, be reduced, increasing the fiber, the last mill setting determining the moisture content of the final bagasse. With no mill slippage in a properly set mill, the water content of bagasse leaving a mill is independent of the water content entering. This is true because the load on the mill is determined by the fiber and not fiber plus liquid. However, with slippage, the water content of the expelled bagasse increases since the fiber entering the mill is less and space is available for more water to pass. Maintenance of feedability is, therefore, of prime importance on the
50 last mill. This requires constant attention to building up of roll grooving and effective operation of the feeding aid. Keeping in mind that the higher the dissolved solids in the residual juices (poorer extraction), the lower is the bagasse moisture at the same fiber content; the fiber content of bagasse is a more useful measure of last mill performance than moisture. Effect on Extraction Other conditions being the same, a reduction in the moisture in bagasse results in higher extraction. The actual increase is dependent upon the fiber, moisture, pol, and refractometer solids involved. By using average figures for these a general relationship as shown in Figure 5-1 is obtained. From this it will be seen that a decrease in bagasse moisture of 1% results in an increase in extraction of approximately 0.1%.
53
52
51
50
49
48
47
46
45
44
43
42
MOISTURE % BAGASSE
Fig. 5-1. Effect of bagasse moisture on extraction.
41
40
51
The calculations assume little change in the pol of the juice extracted and are illustrated in the following example: Assume pol in cane Bagasse fiber pol refractometer solids moisture
13.5 48.0 1.44 2.32 49.7
Then extraction =
97.0
Now increase fiber by one unit Bagasse fiber pol refractometer solids moisture
49.0 1.41 2.28 48.7
Then extraction =
97.1
If the bagasse fiber is coarse, the increase in pol extraction may not be as great because the pol of the juice within the fiber may be higher than that of the external juice. The applied imbibition water will not have reached equilibrium with the internal juice. This effect may be counter-balanced to some extent if the lower moisture level has been obtained by higher pressure on the mill or a greater preparation effect in the mill, both of which would tend to give closer to equilibrium conditions. Effect on Fuel Value The higher fuel value of bagasse at lower moisture content can be reasonably calculated but the practical value obtained varies considerably with the boiler and how it is operated. For rough estimations, however, a change of 1% in moisture changes the fuel value also by 1%. An illustrative example follows: At 50% Moisture Analysis
Per 100 Tons Bagasse
Fiber Moisture Soluble solids
%
Tons
48 50 2
48 50 2
Total
100
Approximate Net Calorific Value or
1853.7 1853.7 3336 χ 7760 χ
kg-cal/kg (3336 Btu/lb) (7760 kJ/kg) χ 1000 χ 100 = 185,370,000 kg-cal/100 metric tons 2000 χ 100 = 667,200,000 Btu/100 short tons 1000 χ 100 = 776,000,000 kJ/100 metric tons
52
At 49% Moisture Analysis
Per 100 Tons 50% Moisture Bagasse
Fiber Moisture Soluble solids
%
Tons
49 49 2
48 48 __2
Total
98
The weight of bagasse is now 98 tons. Approximate Net Calorific Value 1903.1 or 1903.1 3425 χ 7967 χ
kg-cal/kg (3425 Btu/lb) (7967 kJ/kg) χ 1000 χ 98 = 186,507,000 kg-cal/100 metric tons 2000 χ 98 = 671,300,000 Btu/100 short tons 1000 χ 98 = 780,766,000 kJ/100 metric tons
The increase in Net Calorific Value is therefore 186,507,000 - 185,370,000 = 1,137,000 kg-cal/100 metric tons 671,300,000 - 667,200,000 = 4,100,000 Btu/100 short tons 780,766,000 - 776,000,000 = 4,766,000 kJ/100 metric tons or
1,137,000 χ 100 = 0.61% 185,370,000
The actual value obtained in a boiler would be higher, approaching 1%. In calculating the fuel value of bagasse, it is customary to include the energy value of the sugar as well as the fiber. It should be kept in mind that this is valid only if the bagasse goes directly from the mill to the boiler. In storage, the sugar content is rapidly destroyed so that after several hours it can be considered to be zero.
53
Chapter 6
THE IMPACT OF EXTRANEOUS MATTER ON MILLING AND DIFFUSION Extraneous matter is the non-cane material in cane delivered to the factory. This is often referred to as trash, although the meaning of this term varies in cane sugar producing areas. In its original use, it referred to leaf material accompanying the cane stalk as harvested. This concept still obtains in many areas. A t present, the broadest definition is that officially used by the Hawaiian Sugar Technologists, which includes all extraneous material delivered with the cane stalk, from root to growing point, and covers not only cane leaves, but also the growing tip and roots, weeds, soil, rocks, casual matter such as dump refuse, structural material, machinery parts, livestock and even superficial water - an important item in washed cane. Purely quantitative figures on total extraneous matter do not provide adequate information for predicting the effect on operations. It is necessary to know the physical characteristics of the foreign materials. Five main categories, with subclassifications, cover most cases and give useful characterization. 1.
Fibrous material Dry leaves Tops Ground trash (partially decayed material) Roots Dead cane Weeds
2.
Soil Clay Loam Sand
3.
Rocks Gravel Stones
4.
Metal (tramp iron)
5.
Water
A consideration of the general effect of these on milling and diffusion follows.
54
MILLING In engineering calculations, the capacity of a milling tandem is customarily considered to be a function of the fiber throughput per unit time. Actually, the quality of the fiber also has a bearing but is difficult to quantify. Therefore, the calculated quantity of cane crushed in a mill is judged by the fiber content of the cane plus the fiber content of the extraneous material entering the tandem with the cane. In a given area, the fiber content of sound mature cane stalk varies within narrow limits (Hawaiian varieties 12-14%). The amount of fiber entering with the cane as extraneous matter is subject to wide variation, however, dependent upon the effect of burning, field and harvesting conditions, as noted in Chapter 2. The amount of fibrous trash entering with the cane ranges from 4% for well-burned cane on an irrigated plantation to 25% for unburned cane on an unirrigated plantation. The fiber content of the trash also varies from a low level of 10% for ground trash harvested under wet conditions, through 25% for green unburned tops to 80% for dry leaves. Using the estimated figure of 40% fibrous trash removal in the cane cleaner, the range of fibrous trash entering the mill is from 2.5 to 15% based on the entering moisture content. After the cleaner, however, the material is completely saturated with water regardless of the entering moisture. Fiber levels then are: Green tops Leaves Ground trash
25% 35% 10%
A rule-of-thumb value for the fiber content of fibrous trash entering the mill from the cleaner would be twice the fiber content of the cane, or 24 to 28%. Thus, in milling calculations, one ton of fibrous trash is approximately equal to two tons of stalk cane. If the fibrous trash is clean, the major effect of it would be to increase the fiber milled. Addition of 10% trash to clean cane would necessitate a 20% increase in capacity of the mill to handle the same amount of cane. Should the mill have sufficient capacity to handle the higher fiber rate, the setting can be adjusted so that the pol and moisture in bagasse can remain unchanged. Only the quantity of bagasse will increase, which means a greater loss in pol and lower extraction. Imbibition water would have to be increased proportionate to the fiber. The approximate magnitude of the loss can be illustrated example. Assume Cane
100 tons 12.5% pol 12.5% fiber
in an
55
Bagasse
2 5 tons 2.5% pol 50.0% fiber
.". Pol lost in bagasse = 25 χ 0.025 = 0.625 tons and Extraction = 12.5 - 0 625 x 100 1
ù
=
9 5
0 %
. 0
Now add 10% fibrous trash of 25% fiber Cane
110 tons 12.5x^100
Bagasse
=
1
.
I
3
6
%
12.5 + 2 . 5 x 100
=
1
3
p
>
ol
6 % 4
f i b re
30 tons 2.5% pol 50.0% fiber
. ' . Pol lost in bagasse = 30 χ 0.025 = 0.75 tons and Extraction = 12.5 - 0.75 x 100 12.5
94%
Therefore the combined effect of adding 10% fibrous trash would be to decrease the mill capacity by 20% and reduce extraction by 1%. A compensating factor is that the quantity of bagasse would be increased by 20%. The value of this as fuel for generating power would more than compensate for the loss in extraction. There are other effects associated with trash which may be of greater and even governing importance. Even with normal cleaning, fibrous trash, particularly ground trash, will carry into the mill soil, sand and rocks. These materials cause heavy wear on the equipment - knives, hammers, mill rolls, pipes and pumps. The result is a decrease in cane preparation and a lowering of the juice expression by the mills. Polishing of rolls will also decrease feedability and, thus, lower the capacity. The net effect can thus be a decrease in capacity of the mill substantially greater than that caused by the fiber alone. Extraction will also be lower than that caused by the fiber alone. Another factor is that soil carried into the mill decreases the permeability of the bagasse blanket holding the juice and lowers juice drainage, giving decreased extraction.
56
Trash is the cause of excessive water entering the mill from the cleaning plant since clean cane stalks can carry no more than about 2% superficial water. Green leaves, likewise, carry little water due to their waxy surface. Dry leaves and ground trash, however, can carry a large amount of extraneous water. This superficial water adds to the dilution of the juice without aiding extraction. In day-to-day operation of a mill, the effects discussed above will only be observed but cannot be measured. Experience alone will enable the operating staff to make the necessary judgment in appraising the daily results. DIFFUSION The discussion of the effects of extraneous matter on milling applies also for diffusion. The different technology presents a somewhat different picture, however. The capacity of a diffuser is determined primarily by the permeability of the bed of prepared cane, that is, the rate at which liquid can flow by gravity through the bed. The physical character of the fiber and the effect of the extraneous matter on this, therefore, are governing. The capacity then is the volume of juice that can pass through a quantity of cane per unit time. The size of the diffusion vessel, finally, is determined by the volume the bed occupies at an optimum bed depth. In diffuser design, a target extraction, say 98%, is set; then a practicable cell rupture value (Displaceability Index), say 94%, selected and the number of stages necessary calculated. A practicable bed depth has been found by experience to be about 1.5 m (5 f t ) . Bed permeability then determines the length of the bed at a given speed. To compensate for variability in permeability, more stages than calculated are normally designed. Some flexibility in operation is achieved by variation of bed depth and speed. Capacity of the diffuser then becomes a function of the third dimension, the width. Preparation of the cane was noted as the most important factor in diffusion efficiency. Maximum cell rupture, to free the juice, is necessary at the same time giving a fiber which is highly permeable. This was shown to be best obtained by a shredding machine which produces relatively long shreds resulting in minimum compaction when formed into a bed by gravity. Well-prepared cane will have a permeability averaging 250 liters/m^ (6.1 gal./ft2). - Fibrous trash, leaves and tops, generally make shredding of cane more difficult and greatly increases power required. Extraneous matter can have a controlling effect on diffuser performance mainly by the influence on bed permeability. Fibrous trash will increase the volume of the bed and so the capacity of the diffuser will be less in proportion to the volume occupied by the trash fiber. The effect is variable, depending upon the physical character of the trash. In general, shredded trash fiber occupies more volume per unit weight than shredded cane so that the capacity influence is greater than indicated by the weight of the trash.
57
The effect of clean fibrous trash on permeability is not great. Shredded dry leaves and green leaves can improve permeability. This is particularly true of green leaves, being waxy, improving liquid flow. Ground trash, being spongy, can have an opposite effect. However, unless the material is extremely fine (less than 300 mesh), fibrous trash does not have a major influence. Soil particles are the most serious suppressors of permeability. Clay type soils can almost completely seal a fiber bed and essentially stop diffusion if present in sufficient quantity. This leads to flooding - the most serious diffuser difficulty of all. With material finer than 300 mesh, the reduction in the permeability is proportional to the amount of the fine material up to a point - after which the amount is immaterial. This is because when an almost impermeable layer is formed, little liquid will pass through and the remainder will flow away on the surface of the bed. Other materials which can effect bed permeability are precipitated substances such as clarifier settlings. These, if returned to the diffuser, must be distributed through the bed and not placed on top. Sand, grit and rocks, of course, are without effect. In conclusion, as to the role of extraneous matter on diffusion, the fibrous trash has the effect of reducing capacity in proportion to its volume. Any other extraneous matter, soil in particular, of a fine particle size that decreases the bed permeability, can have a governing effect on the capacity of a diffuser. The impact on the preparatory and dewatering equipment is much the same as on milling. Because bagasse pol is lower in diffusion the effect of trash on pol loss is substantially less than in milling.
59
Chapter 7
A COMPARISON OF DIFFUSION AND MILLING WITH RESPECT TO RECOVERY AND LOSSES A question often raised, and obviously not subject to definitive answer, is how do diffusion and milling compare from the standpoint of recovery and losses. As with all equipment, it depends upon the capability of the installations and how they are used. There are, however, some basic principles to consider in looking at the technologies involved, and hence arriving at reasonable relative conclusions. Since the conventional extraction figure incorporates cane quality as well as extraction efficiency, it is more useful to look at bagasse pol as a measure of processing results. Basic diffuser design sets a target of 1.0% pol in bagasse. With good quality cane, this would correspond to extraction at the 98% level. Although in common practice mills are not designed to a bagasse pol target, the range for a good operation is 1.5 to 2.0 pol. Thus design consideration are based on losses in bagasse by diffusion 33% to 50% less than those for a mill. The extraction level for a mill is therefore 96% compared to 98% for diffusion. Diffusion takes place in closed equipment growth range for microorganisms. Loss by essentially nil. This contrasts sharply with perature system optimum for microorganism 1% have been reported.
at a temperature above the their activity is, therefore, milling - an open, low temactivity where losses above
Chemical inversion of sucrose, a product of temperature and pH, is negligible in a diffuser if a near neutral pH is maintained, since the average juice residence time is less than 15 minutes. With unlimed juice (pH 5.2), the calculated loss by inversion is 0.14%. With a mill, it would be less than 0.05%. It is true that the purity of diffusion juice is lower than the purity of absolute juice, as determined by direct disintegrator analysis of cane. However, it has been established that the purity change is not caused by a loss of sucrose but by an increase in dissolved solids (Brix). This is attributed to a solubilization of some high molecular weight components of the cane in the diffuser. No losses are involved therefore. It may be concluded from the above considerations that recovery by diffusion should be of the order of 2% greater than that by milling, in properly designed equipment operated at rated capacity. A good practical comparison of milling and diffusion is given by the results in Hawaii obtained by Hawaiian Commercial and Sugar Company in its two factories, Puunene and Paia. The factories are adjacent to one another and operate under as similar conditions as it is possible to find.
60
The Puunene factory had two milling tandems, each with four 198 cm (78 in.) mills preceded by a shredder and a 2-roll crusher. The Paia factory had a Silver 840 diffuser (2134 cm, 840 in. diameter) preceded by a Buster and Fiberizer, and followed by a 214 cm (84 in.) 5-roll pressure feeder dewatering mill. The results for the 1975 operating season were as follows: Extraction Bagasse pol Bagasse moisture Mixed juice purity Boiling house recovery Total recovery
Puunene
Paia
95.1 2.6 45.5 85.6 89.2 84.8
97.3 1.4 45.7 86.4 89.3 86.7
61
Chapter 8
CLARIFICATION The main objective at the clarification station is to increase the pH of the juice to a level where losses of sucrose by inversion are kept to a minimum in the subsequent sugar recovery process. Important, but secondary objectives, are the removal of insoluble material and some undesirable dissolved substances. It is fortuitous that adjustment of pH to the optimum level with the cheapest alkalizing agent, lime, gives satisfactory removal of much of the undesirable material in the juice and provides a suitable base for sugar recovery. Magnesium oxide behaves similarly to lime and is often used to reduce evaporator scaling. Target for pH adjustment of the juice is that which will give a syrup pH of 6.5. This is about the optimum for carrying out the crystallization steps which follow, by giving easy boiling massecuites, minimum development of undesirable compounds and color, small decomposition of reducing sugars and little loss of sucrose by inversion. A t higher pH levels, there is greater development of viscosity and color and substantial losses of reducing sugars, particularly fructose. A t lower pH levels sucrose inversion increases rapidly. Processing a syrup at 6.5 pH will usually give a final molasses pH of around 5.8, so the crystallization is conducted in the range of 6.5 to 5.8. The mixed juice will have to be raised to a pH of 7.5 in order to obtain a syrup of 6.5 because of the drop which occurs in the juice heaters, clarifiers and evaporators. This increase in acidity is caused by the relatively slow reaction with lime and particularly magnesium oxide in the cold, by formation of organic acids and by loss of ammonia from the decomposition of amino acids. The exact pH of juice liming required varies with the composition of the juice, so routine adjustment of set point is essential. Usually, with good quality cane, good clarification also occurs under this type of control. That is, there is good flocculation of suspended matter, rapid settling and overflow of limpid juice. With poor quality or deteriorated cane it is often impossible to get a clear juice and rapid settling. Milky-appearing juice is a sign of sour cane. It is caused by dextrans which, by protective colloid action, prevent good flocculation. In such cases, higher liming is sometimes helpful even though the effects on sugar crystallization are less favorable, so a compromise is necessary. In most cases, however, there is little that can be done to improve the situation. POLYELECTROLYTES Polyelectrolytes are the one class of additives, besides the alkalizing agent, that is so generally effective that its use in clarification is routine. In very small quantities, one to two parts per million of juice,
62
polyelectrolytes improve flocculation, give more rapid smaller volume of settlings. They have no effect on pH.
settling
and
Standard polyelectrolytes are partially hydrolyzed Polyacrylamides. The pioneer agents, used under the trade name of Separans, remain as standard. Preparation of the solution and point of application are important. High purity water, such as condensate, should be used to prepare the solution. Water with high dissolved solids or suspended solids will give a solution with inferior floe forming characteristics. The best point of application is after the juice is heated and a good place is in the flash tank of the clarifier. A metering pump is necessary to control the addition. The useful quantity of polyelectrolyte varies with the quality of the juice and the type of soil present. Usually, one to two parts per million of juice is sufficient. U. S. government regulations limit the quantity to not more than 5 parts per million. MAGNESIUM OXIDE Magnesium oxide (Magox) is an effective alkalizing agent for replacing lime becuse it leads to less scaling in the evaporators. Evaporator scale is composed mostly of calcium sulfate and silica. Since magnesium sulfate is soluble, less calcium sulfate is deposited on the tubes than when lime is used. Magnesium oxide has 40% more basicity per unit weight than quick lime, but even so is more expensive to use. When scaling is a problem, it can be used to replace enough of the lime to control the scaling. The amount must be determined by trial. Often, a 50-50 ratio is adequate. As the solubility of magnesium oxide is low, its rate of reaction with juice components is slow, making addition control by pH unsatisfactory unless a retention time of some 15 minutes is allowed. Although there is lag time in the heaters, better control is obtained from heated juice entering the clarifier. The quality of the magnesium oxide is extremely important as, depending upon conditions of calcination, the product can be highly absorptive (low temperature) or practically nonabsorptive (high temperature). For clarification use, the specifications are loss on ignition 2.5% and iodine number 20. PHOSPHATATION Clarification of juices deficient in natural phosphate is sometimes helped by increasing the phosphate content. In general, juices containing less than 0.03% phosphate are considered deficient. Adding phosphate to this level gives more calcium phosphate floe and often better clarity. Care must be exercised, however, because increased volume of settlings results and, usually, slower rate of settling.
63
Juices with excess natural phosphate (around 0.09%) are slow settling and give large volumes of settlings. Clarification of these is sometimes helped by reducing the liming pH. The best source of phosphate is phosphoric acid, a liquid which is expensive and presents a problem in handling. Normal practice is to use the cheapest and most available form which is fertilizer grade ammonium phosphate. SULFIT ATION Sulfur dioxide is widely used, both in cane and beet sugar processing, to reduce sugar color. It also has a secondary effect in improving the boiling characteristics of massecuites. The action of sulfur dioxide on color is complex, involving converting colored compounds to colorless, preventing color formation by oxidation and inhibiting color development from reducing sugars and amino acids by the browning reaction. Addition of the sulfur dioxide gas may be made before or after liming, with little apparent difference in effectiveness. The simplest application is to control the quantity entering the mixed juice relative to juice flow, then carry out the customary liming by pH control. In order to avoid the corrosiveness of sulfured mixed juice, partial preliminary liming may be used. To simply improve the quality of raw sugar, it is not necessary to have substantial absorption or retention equipment. Sulfur dioxide from a standard sulfur burner can be introduced proportional to flow into prelimed mixed juice by means of a venturi unit. A certain percentage of juice recycling will give uniformity. Liquid sulfur dioxide may be introduced directly into the mixed juice line, controlled by means of a flow meter. The quantity of sulfur dioxide used in improving the color of raw sugar is in the range of 100 to 500 ppm. This compares with around 1000 ppm used in the manufacture of a direct consumption plantation white sugar. JUICE HEATERS The objectives of juice heating are elimination of microorganisms by sterilization, completion of chemical reactions with the alkalizing agent, flocculation and removal of gases. Effective gas elimination is obtained by flashing as the juice enters the clarifier. The juice temperature must, therefore, be raised above the boiling point at atmospheric pressure which, at sea level, means a minimum of 103°C (217°F). If flashing does not occur, gas bubbles adhering to the floes decrease the rate of settling. Good steam economy is obtained by heating with vapors from the evaporator. Usually this is done in two steps with primary heating by vapors bled from the second or third effect and secondary heating with vapors from the first effect. The efficiency of heating is primarily a function of the heat transfer between the vapor on the outside of the
64
tubes and the liquid within. The optimum juice velocity is 2 m (7 f t ) per second which gives good heat transfer without too high pressure drop. The governing factor, however, is scaling of the tubes so periodic cleaning is required. Heater scale is relatively soft and can often be removed by cracking with steam followed by flushing with water. If this is not effective, flushing with caustic soda solution will usually remove the scale. Good heat transfer also requires removal of noncondensible gases and condensate discharge from the heaters. ROTARY VACUUM FILTERS Settlings from the clarifier, which contain 5 to 10% insoluble solids, are handled on a rotary vacuum filter to remove most of the insoluble material and wash the juice from it. The juice from the settlings, along with the washings, are returned to the incoming juice. A diagram of a vacuum filter station is shown in Figure 8-1. The filter is a rotating drum, the lower part of which is immersed in a tank of settlings. The drum is made of independent filter sections covered with a screen, usually of stainless steel with 0.6 mm (0.023 in.) perforations. The filter sections are connected by piping to a manifold at the end of the drum. The manifold opens to a rotary port valve which controls the connection to a vacuum system. See Operating Action Diagram, Figure 8-2. The filter sections immersed in the tank containing clarifier settlings mixed with bagasse fines are opened to a low vacuum of 18 cm (7 in.) of mercury called pickup. In this position, liquid flows through the screen and a cake is formed on the screen. After the formation of the cake, the bagasse fibers begin to retain the insoluble particles and the liquid passing through becomes relatively clear, as distinct from the first influx which is unfiltered. As the section continues to rotate through the settlings, the cake becomes thicker until it finally emerges from the settlings. Here, the valving connects with a high vacuum system of up to 50 cm (20 in.) of mercury and water is applied to the surface by sprays. The water passes through the cake, washing out the juice. A t the top of the drum, water is added by means of drip pipes which keeps the cake wet and the vacuum sealed until the vacuum is released and the cake scraped from the drum. The juice and washings are pumped to mixed juice and the cake is discharged to waste. Bagasse fines are used as a filter medium for retaining the suspended solids. These are obtained by screening bagasse and are mixed with the settlings before going to the filter tank. Bagasse Fines Fines for the filter are commonly obtained by means of a screen fixed in a bagasse chute. A suitable screen is flanged-lip type with 0.3 χ 0.5 χ 2.0 cm (1/8 χ 3/16 χ 3/4 in.) slot dimensions at head, tail and length.
Fig.
8-1 Vacuum filter station.
65
66
A. Β. C. D.
Fig.
PICK UP CLOSE WASH DISCHARGE
8-2. Vacuum filter operation action diagram.
67 2
2
Minimum area required is 500 c m (0,5 f t ) per ton prepared cane per hour capacity. The fines are pneumatically conveyed to a cyclone above the settlings mixer. A t the collection point, the stream of the fines is adjusted to fall past the intake in a manner so that sand drops out of the stream by air separation. Good quality screening should provide fines, 90% of which will pass through a 14-mesh screen. The material should contain fibers as well as pith in order to have good permeability. Mixing Fines with Settlings Fines are mixed with the settlings in a ribbon screw mixer. The fluidity of the mix must be maintained in order to prevent piling up of the bagasse on the surface. Since fluidity depends upon the consistency of the settlings, as well as the quantity of added fines, the ratio of fiber (fines) to insoluble solids that can be maintained is variable. A minimum ratio or 0.35 is considered standard and up to 0.5 is desirable. To maintain a 0.5 ratio, a quantity of bagasse fines equal to that of the insoluble solids must be added because the fines are 50% moisture. Thus, 100 tons of mixed juice at 1% insoluble solids would need 1 ton of fines. Pickup Vacuum At the filter pickup point, the vacuum applied should be just enough to hold the cake, but not enough to compress the cake that will reduce permeability; 18 cm (7 in.) vacuum is optimum. If a high vacuum is applied here, the cake will be compressed and the subsequent wash water will not penetrate in sufficient quantity to displace the juice. Wash Vacuum On the wash cycle, the vacuum must be high enough to provide juice displacement flow through the cake. A minimum vacuum of 38 cm (15 in.) is standard and 50 cm (20 in.) is the upper limit so that excessive flashing will not occur. Washing The cake must be uniformly covered with water during the wash cycle. It should not be allowed to dry out. Otherwise, the cake will crack, vacuum will be lost and cake penetration cease. The cake should always have a shiny, wet surface. Initial washing is with sprays set to give enough water to keep the cake wet without running down and eroding the surface. For these sprays to operate properly, clean water must be supplied at constant pressure. A pressure control valve is used on this line and a screen is essential. The water should be at a minimum temperature of 70° C. With a cake of good physical characteristics the quantity of wash water necessary can be as low as 0.5% on mixed juice and should not exceed 1.0%. Speed Standard filter speed is 10 revolutions per hour. The speed may be increased under some conditions and, in particular, with high ratios of
68
fiber to insoluble solids. In general, it is desirable to operate at the slowest speed compatible with a given composition of the settlings. Cake Pol Standard target for cake pol is 1.0%. conditions, the pol should be lower than this.
Under good
clean
cane
Cake Quantity A rule-of-thumb indicator of the quantity of cake is: Filter cake = 4 χ insoluble solids in mixed juice + 0.01 gross mixed juice
CLARIFIE RS A clarifier alkalizing step sterile product, give a syrup of of: 1. 2. 3. 4. 5. 6.
should provide the means of withdrawing juice from the in a condition suitable for sugar recovery. This means a relatively free from insoluble matter and at a pH level to 6.5. The equipment must, therefore, supply the functions
Gas removal Settling Scum removal Clear juice drawof f Settlings thickening Settlings removal
Present practice makes use of equipment in which the treated juice enters continuously with simultaneous withdrawal of clear juice, settlings and scums. The best design is that giving the minimum flow velocity at the entry and drawoff points, minimizing interfering currents. Units with multiple feed and drawoff points are less amenable to control. Flashing Clarifiers are equipped with a flash tank where the juice first enters and boils at atmospheric pressure, eliminating gases. The effectiveness is more a function of the area rather than the volume. Commonly, flash tanks have too small an area. The La Musse * formula is useful: Area for flash in m2 = 4.186 W ( T l - Tg) L Where W = Weight of juice in metric tons/hr T\ = Temperature entering, ° C T2 = Temperature at flash, ° C L = Latent heat of steam at T 2 , kcal/kg For a juice rate of 221 tons/hr and temperatures of 103 and 100°C, 2 respectively, the formula calls for 5.15 m area.
69
Settling Space The capacity of the clarifier is determined by the retention time necessary to allow the settling of a filterable mud. The most efficient design in this regard is the single mud compartment, multiple launder drawoff, of which the SRI (Australia) is a version. A retention time of 30 minutes or less is possible in these types. This compares with the multitray and multimud compartment units with retention time in the range of 2 to 2-1/2 hours. A disadvantage of short retention time clarifiers is that the juice flow must be relatively constant, for there is little surge capacity. The area involved in settling is a major factor in efficiency. The greater the area per unit volume, the more rapid the settling and the smaller the volume of settlings. Multicompartment clarifiers suffer more from uneven flow distribution and disturbing currents, even though total area may be adequate. Operation of Multicompartment Clarifiers Control of a multicompartment clarifier remains a manual operation. Drawoff of clarified juice is controlled by adjusting the drawoff valves to maintain the desired clarity, reducing the flow from those giving cloudy juice. This can be a time-consuming job since reducing the flow from one increases the flow from others in unequal amounts. Settlings drawoff is usually set to maintain a minimum consistency of 5% insoluble solids so that good operation of the filters is possible. With clarifiers having one mud compartment, this is relatively easy. For those with two mud compartments, it is difficult and for those with four, it is more difficult. Usually, more mud enters the first and last mud compartments than those in-between, so adjustments must be made accordingly. However, changing the amount of drawoff from one changes all the others and all the clarified juice drawoffs as well. As the quality of the incoming juice changes, so must the operation of the clarifier. The common solution to this problem is to provide excess clarifier capacity. This is not only capital expensive, but also increases losses because of longer retention time. The clarifier should be kept full at all times and not be used as a surge tank by allowing the level to drop. Holdover Juice Normal sucrose losses in clarification, exclusive of filtration, are of the order of 0.2%. The figure includes sucrose inversion, destruction and handling losses. Losses where juice is held in the clarifier for longer periods, such as weekend shutdowns, is subject to continued losses principally by inversion. These depend upon the temperature and the pH of the juice. In order to keep losses to a minimum, the temperature must be kept above 70°C to prevent the growth of microorganisms. The pH tends to
70
drop with storage, so addition of soda ash is useful to prevent a fall below 6.0. Normally, juice should not be held over 24 hours because of the difficulty of maintaining the temperature. The growth of organisms cannot be tolerated, as not only loss of sucrose occurs but subsequent sugar boiling operations are affected. Clarified Juice Screening Clarified juice usually contains a small amount of fine bagasse. This should be removed by flowing through a hillside screen of 80 mesh. REFERENCES 1
LaMusse, J. P., South African Sugar J. 61, (1977), 103.
71
Chapter 9
EVAPORATION The evaporation station performs the first step in the process of recovering sugar from juice - the evaporative removal of water. Standard practice is to concentrate the clarified juice to about 65 refractometer solids which requires removal of approximately 75% of the water. Steam economy dictates the application of the multiple effect principle. An appropriate installation makes use of a quadruple or quintuple effect arrangement with sufficient capacity to evaporate the water and is so designed to provide vapor for juice heating and vacuum pan operation. The station also furnishes boiler feed water from condensate. MULTIPLE EFFECTS In multiple effect evaporation the vapor from a vessel of boiling juice is used as the source of heat for a subsequent vessel. This can be done by reducing the pressure in the second vessel so that the boiling point is lower. In a series arrangement, or multiple effect, the Rillieux Principle states that one unit of steam will evaporate as many units of water as there are vessels, or effects. Thus, in a generally used four-unit series, or quadruple effect, one unit of steam would evaporate four units of water. The quantity is not absolutely correct but is close enough for routine estimations. Factors which cause deviations are: 1. Heat necessary to raise the temperature of the juice to the boiling point in the first effect. 2.
Heat losses by radiation and removal of incondensible gases.
3.
Increase in latent heat of the vapor as the temperature decreases.
4.
Decrease in specific heat of the juice as it is concentrated.
5.
Flash evaporation of the juice as it enters a lower pressure effect.
6.
Condensate flash.
The first three of these tend to decrease the total evaporation while the last three tend to increase the evaporation. The overall effect found in practice is that the total evaporation is somewhat less than the principle states. Calculations based upon estimations of the values of the above six factors show that the evaporation rates of the first three effects of a quintuple are less than shown by the Rillieux Principle whereas those of the last two, and in particular the last effect, are larger. However, because of scaling this is not found in practice. In order to have effective heat transfer, there must be an adequate temperature drop across each heating surface in order to transfer heat
72
through the tube to the juice, so the number of effects practicable is limited. The temperature in the calandria of the first effect is determined by the pressure of the exhaust steam available. Although it is advantageous to use higher pressure steam, general practice in cane sugar 2 2 operations is to utilize exhaust steam at not over 1.05 k g / c m (15 l b / f t ) (103 kPa) to enable better power production from the turbines. If the steam is saturated, the temperature would be 121° C. Actually, because of loss in the mains, the exhaust steam would enter the calandria of the first effect at around 116°C. Usually, the steam is superheated, and an average of 30° C superheat is not objectionable. Since it is not advisable to operate the last effect of the evaporator station at less than 55° C because of the poor heat transfer with high viscosity syrups encountered at lower temperatures, the total temperature drop across the station with saturated steam would thus be 61°C (116°C 55° C). With a quintuple effect, this corresponds to an average drop of 12°C across each body, a value that is about the minimum for satisfactory heat transfer, so six effects would not be desirable. The heat transfer would be better in a quadruple effect with an average temperature difference of 15° C, which is about optimum. The total effect of temperature increase in the juice caused by hydrostatic head also increases with number of effects making the average temperature difference smaller. The sum of all these factors leads to the generalization that converting a quadruple effect to a quintuple effect by addition of a body of the same heating surface does not raise the capacity of the evaporator unless the initial steam pressure is increased. A fifth body added at the front of the evaporator, and having a larger heating surface, would increase the capacity if the vapor produced by the added area is used for juice heating or pan boiling. The increase in capacity would be about equal to the quantity so used. In addition to temperature difference, heat transfer rate is proportional to the area of the tubes (heating surface), to the heat transfer coefficient and thickness of the tubes. The latter two have little import because with use scaling becomes the governing factor. The capacity of the evaporators can be increased by installing a juice preheater as an auxiliary ahead of the first effect. The heater, operating on exhaust steam, should raise the juice temperature close to the flash point. VAPOR BLEEDING As vacuum pans are single effect evaporating bodies, better steam efficiency can be obtained by heating them with vapor from one of the evaporator effects. The steam saving achieved increases with the downstream position of the effect from which vapor is bled, as represented in the formula: Steam saving =
73
Where m is the position of the effect and η is the number of effects. Thus, bleeding from the number one effect of"a quadruple would result in a saving of one fourth of the .weight of vapor withdrawn. It is generally desirable to maintain a positive pressure on the calandria of vacuum pans, so normal practice is to heat pans with vapor from the first effect. The same is true of secondary juice heaters where juice must be heated above the atmospheric flash point. Primary juice heating may be supplied by vapor from the second effect or, in some cases, the third effect. Vapor bleeding also reduces the quantity of condenser water required. CAPACITY The capacity of an evaporator station to remove water is established by the evaporation rate per unit area of heating surface, the number of effects and the location and amount of vapor bleeding. With no vapor bleeding, the capacity is determined by the performance of the least productive effect. The system is self-balancing. If a downstream effect cannot use all of the vapor developed by the preceding effect, the pressure in the preceding effect builds up and evaporation slows down until equilibrium is established. With effects of uniform heating surface, the productivity of the last effect is often the lowest on account of heavier scaling and high syrup viscosity. With vapor bleeding, the effect bled must have increased heating surface to provide for the vapor withdrawn, as well as the amount required for the downstream effects. Furthermore, all the bled vapor must be withdrawn, otherwise, as explained above, the system will drop to the evaporation level of the least productive effect. This means that the bled vapor, if not used in the pans or heaters, must be vented if the evaporation rate is to be maintained. This point must be carefully considered in controlling the station. If too much vapor is used by the pans and heaters, the pressure in the vapor will drop and there will be insufficient vapor for them and for the evaporators. Pressure can be maintained for the pans and heaters (vapor loop) by throttling vapor to the downstream effects. This, however, will reduce the total evaporation and in order to maintain evaporation, makeup steam must be added to the loop. OPERATION In operating the station, the exhaust steam supply to the first cell is controlled to give the total evaporation required to maintain the product syrup at the set range of 65 to 70 refractometer solids. Uniform juice feed is essential to good evaporator performance, especially with vapor bleeding. Regulation can be maintained by level sensing in the evaporator supply tank. Standard practice is to reduce the steam supply when juice level drops in the evaporator supply tank. With a demand remaining for first cell vapor, the vapor to the downstream cells is throttled. If the supply tank level falls below a set minimum, water can be automatically
74
opened to maintain evaporator operation. If the first cell vapor pressure drops too low, makeup with exhaust steam will be necessary in the vapor loop. The following factors evaporator station.
are
important
in
maintaining
an
efficient
Automatic Control Efficiency is enhanced by automatic essential elements controlled are:
control instrumentation.
The
Absolute pressure Syrup solids Liquid level Feed supply Absolute pressure is controlled by regulating the quantity of water to the condenser thereby maintaining a standard temperature of syrup in the last cell of about 55° C. The absolute pressure set point will depend upon the refractometer solids of the syrup. In the range of 65 to 70, the absolute pressure will be around 10 cm of mercury (4.0 in.). Syrup solids is controlled by regulation of the syrup discharge valve from the last cell. A minimum of 65 refractometer solids is standard. Concentrating above 70 should be avoided so that there is no danger of crystallization occurring. Liquid level in the tubes is an important factor in the heat transfer rate. If the level is too low, all the tube heating surface is not used and the tubes may dry out at the top. If the level is too high, the lower part of the tubes will be flooded with juice moving at too low a velocity to give maximum evaporation. The optimum level is that at which the liquid is just being carried by vapor bubbles to the top of the tubes with only a small flow over the tube sheet. It varies with the size of the tubes, temperature, heat transfer rate, scaling and viscosity of the juice. Under good conditions, the best level is around 25% of the tube length. Rarely is it over 40%. Since no flashing of feed occurs in the first cell, the situation differs from the others. If the entering juice temperature is much below the temperature of evaporation, the juice must be some distance up the tube before it starts to boil, resulting in higher levels. In the last effect, although the heat transfer rate is lower, the vapor expansion is much greater because of the lower vacuum giving a higher explosive force up the tubes. Feed supply should be kept uniform using the supply tank control. Dead band control is such that above a set level the supply is signalled to cut back. Below a set level, the steam reduced, and at a minimum level, a water valve is opened to evaporator going.
for surge source of supply is keep the
Condenser and Vacuum System With
satisfactory
condenser
design
and
adequate
vacuum
pump
75 capacity, the points of concern in operation are the quantity of water, water temperature and air leaks. Good condenser design will give the rated capacity with a difference of 3°C between the condenser discharge water and the entering vapor. The quantity of water required depends on its temperature - the higher the temperature, the more water necessary. Air leakage is usually the principal cause of evaporator malfunction. Individual vessels and the piping system must be checked periodically for leaks. Another common difficulty is air coming in with the juice feed. This would not be detected in testing the equipment for leaks. Condensate Removal Inadequate condensate removal can cause partial flooding of the tubes on the vapor side of the calandria with reduction of the effective heating surface. Since the first effect is always under pressure, condensate is readily removed by means of a steam trap. Condensates from the other cells, which are under vacuum, are removed by pumping. Condensates from individual cells are kept separate so that in case of contamination only the water from the leaking unit is discarded. Also condensate from the second cell contains volatile organic matter from the juice in the first cell. The principal substance present is ethanol but other alcohols, esters and acids are present and these are undesirable as feedwater for high pressure boilers. This condensate is better used for the house hot water supply than in the boilers. Noncondensible Gas A considerable quantity of noncondensible gas (air and carbon dioxide) may enter the calandria with the heating vapor. Air enters also through leaks in the vacuum vessels and carbon dioxide is generated in the juice. If not removed, these will collect and interfere with the condensation of steam on the tube surface. Calandrias under pressure can be vented to atmosphere. Those under vacuum must be vented to the vacuum system. Although it is easy to vent to the body of the same effect, the gas is then transferred to the next calandria and so on until removed in the condenser. Better practice is to vent the effects individually to the vacuum system. This requires careful valving to prevent appreciable vapor losses. A good guide is to monitor the temperature of the vent line. The closer it approaches the calandria temperature, the more vapor is lost. Calandria venting should be from both the top and bottom of the calandria and at points opposite the vapor entrance. Scaling Cane juice becomes saturated with respect to calcium sulfate and silica before the concentration of dissolved solids reaches the desired syrup level of 65 refractometer solids. The precipitation of these two compounds, together with small amounts of other substances, causes a build up of hard scale, principally in the last cell. Heat transfer is then greatly impaired. The quantity of scale deposited depends on the total concentration of
76
precipitable components in the juice, but the major constituent is calcium sulfate. Its concentration can be reduced by using magnesium oxide instead of lime in clarification, since magnesium sulfate is soluble. A l though scaling cannot be prevented completely, control to workable levels is possible by substituting magnesium oxide for part or all of the lime. Scaling is greatest near the vapor entrance and at the bottom of the tubes. Scale builds up with use and evaporation rate diminishes so that cleaning becomes necessary. The most effective cleaning agent is caustic soda which can be used directly as a 50% commercial solution. The efficiency of the cleaning solution decreases rapidly with dilution so care must be exercised to keep dilution to a minimum. Standard practice is to spray the caustic soda over the tube sheet with sufficient steam in the calandria to heat it. Treatment time averages two hours. The caustic solution may be reused but becomes less effective because of dilution and concentration of dissolved material. After caustic soda treatment, a powdery film will remain on the tubes. This is easily removed by washing with water followed by dilute sulfamic acid (0.25%) rinse and water wash. Entrainment Entrainment, or the carrying over of liquid with the vapor from one effect to the calandria of the next or to the condenser of the final effect, not only results in the loss of sugar but also causes contamination of condensate for boiler feedwater and pollution of the condenser water discharge from the factory. Juice is expelled from the top of the tubes with a velocity sufficient to atomize the liquid and project droplets to a considerable distance vertically. The velocity increases from cell to cell reaching a maximum in the final body where calculated velocities for 5 cm (2 in.) diameter tubes may reach 18 m (60 ft) per second. This means that some droplets have a velocity sufficient to project them 18 m. This is a hundred times that in the first effect. The problem is, therefore, most serious in the last effect and an efficient separator is essential. Separation, for the most part, works by effecting the impingement of the droplets on a surface from which the liquid can be returned to the cell. The main problem encountered with separators, if their design is adequate, is keeping them clean, as scaling takes place rapidly. Saddle or ring-packed separators, approximately 30 cm (12 in.) in depth extending across the top of the cell, are most effective provided they are also cleaned when the evaporator is cleaned. This involves spraying the caustic soda solution through the separator. Similarly, packed separators in the vapor pipe need also to be cleaned with the evaporator. They are less effective because of the high vapor velocity in the small diameter pipe. They are better installed in the enlarged sections. Stainless steel mesh separators extending across the body are effective if kept clean. Zig-zag plate units, also placed across the body, are effective but more expensive. With sufficient height above the calandria, simple flow reversal baffle-type separators in the dome serve well. Condensates should be monitored by conductivity routinely for entrainment. Condenser water should be tested for sugars on a scheduled basis.
77
Malfunctions Evaporator problems arise from many causes. The principal ones are: Low steam pressure Air leaks in system Condenser water supply Vacuum pump Noncondensible gas removal Condensate removal Scaling Vapor bleeding Difficulties with the steam supply, vacuum and water system are easily detected. Problems in individual cells with gas and condensate removal and scaling are best detected by observation of the temperature drop across the cells. Temperature and pressure measurements in each should be recorded on a regular basis. A malfunction can be traced by changes in these measurements. For example, if the temperature gradient across a cell increases, while the total drop across the set remains the same, then the drop across the other cells will decrease. This means a problem in the abnormal cell that requires investigation. It could be caused by failure of condensate removal or noncondensible gas removal. Of course, it could be caused by scaling, but this would occur gradually and would only occur to a great extent in the last cell. A sudden decrease in evaporation over the whole set might be caused by limited withdrawal of vapor to heaters and pans. Unless vapor is vented, pressure will build up in the vapor, reducing evaporation rate. This can be detected by pressure readings. EVAPORATOR CALCULATIONS In the design of a new evaporator station, many assumptions are made in order to arrive at the area of heating surface required in the multiple cells. Often, the designer is given only the one figure - tons cane to be handled per hour. He must, therefore, estimate the following: Weight of juice to be evaporated per hour. Refractometer solids content of juice. Temperature of juice. Exhaust steam pressure. Exhaust steam temperatures. Quantity of vapor to be bled. Heat transfer rate at the heating surface of each cell. Boiling point elevation of juice in each cell. Refractometer solids of syrup. Condenser water quantity and temperature. Heat losses. For an operator on a day-to-day basis, such estimations are of little value. He is faced with conditions in which the quantity and composition of the juice varies continuously, as does the exhaust steam pressure, the requirements for bled vapor, heat transfer rate, and the quantity and temperature of the condenser water. He can, however, make use of
78
approximations to arrive at problem areas in the evaporator set. Knowing the exhaust steam and juice flow to the evaporator by measuring temperatures, refractometer solids and pressure in the cells, calculations may be made on the basis of assumed heat transfer values. The Dessin-Coutanceau tion is: E
= v
1
method is useful for this purpose. The equa-
(100 - B ) ( t v - tj)(tj - t c ) + ( t v - tj) T670ÜÖ
Where
Ev Β tv tj tc
= = = = =
2
rate of evaporation in lb/ft /hr average brix of juice in the cell temperature of the vapor in ° F average temperature of juice in ° F temperature at vacuum in last cell in ° F
The constant of 16,000 applies for evaporators with tubes 1-3/4 χ 72 2 in., 5 lb/in. exhaust pressure and 26 in. Hg vacuum in the last effect. The values from the last cell are set into the Dessin-Coutanceau equation using measured values of solids, temperature of syrup, vapor to condenser and calculated temperature of vapor to the cell. Calculations for the other cells follow in order. The following example is from an actual case of an evaporator showing limited capacity. Basis To evaporate 157,400 kg (347,000 lb) water per hour from 196,860 kg (434,000 lb) juice at 13.0 solids. Quadruple effect with vapor bleeding from first cell. Cell number 1 2
Heating surface, m 2 ft
1,394 15,000 2
Evaporation rate, kg/m /hr 2 lb/ft /hr Vapor bled, kg/hr lb/hr
39.0 8.0
2
3
1,324 14,250
1,231 13,250
25.4 5.2
4
27.3 5.6
975 10,500 34.2 7.0
22,680 50,000
Water evaporated, (without bleeding) kg/hr 33,566 lb/hr 74,000
33,566 74,000
33,566 74,000
33,566 74,000
Total water evaporated, kg/hr lb/hr
33,566 74,000
33,566 74,000
33,566 74,000
56,246 124,000
Solids out
18.3
24.1
35.3
65.0
Solids average
15.7
21.2
29.7
50.1
0.7 1.2
1.1 2.0
2.0 3.6
4.3 7.8
Boiling point elevation, ° C °F
79
Vacuum °F
mm Hg
Pressure
Cell 4
°C
in. Hg
Vapor t ( t c ) Juice t4 ( t j 4 ) tc - tj4 At (drop)
54.0 58.3 4.3 20.0
129.0 136.8 7.8 36.0
64.8
25.5
Vapor t4 ( t V 4 ) 58.3 + 20 136.8 + 36.0
78.3
172.8
42.9
16.9
Juice t 3 (tj3> 78.3 + 2.0 172.8 + 3.6
80.3
176.4
tc - t j 3 80.3 - 54.0 176.4 - 129.0 At
26.3 8.3
47.4 15.0
Vapor tß ( t V 3 ) 80.3 + 8.3 176.4 + 15.0
88.6
191.4
26.2
10.3
Juice t2 (tj2> 88.6 + 1.1 191.4 + 2.0
89.7
193.4
t c - tj2 89.7 - 54.0 193.4 - 129.0 At
35.7 5.8
64.4 10.5
Vapor t2 ( t V 2 ) 89.7 + 5.8 193.4 + 10.5
95.5
203.9
11.4
4.5
Juice t\ ( t j i ) 95.5 + 0.7 203.9 + 1.2
96.2
205.1
tc - tjl 96.2 - 54.0 205.1 - 129.0 At
42.2 6.9
76.1 12.5
Vapor t i ( t v i ) 96.2 + 6.9 205.1 + 12.5
103.1
217.6
kg/cm
2
lb/in.
Cell 3
Cell 2
Cell 1
0.12
1.7
2
80
These calculations indicate that the evaporator should operate if the 2 2 exhaust pressure were 0.12 k g / c m (1.7 lb/in. ) and the exhaust steam were supplied at the rate of 56,246 kg/hr (124,000 lb/hr). However, a very high rate of evaporation was used for the fourth cell because of its small 2 2 heating surface - 34.2 kg/m /hr (7.0 lb/ft /hr). This is higher than normally encountered by about 50%. The conclusion would be that the evaporator cannot handle this quantity of juice. The limiting cell is number 4, and it should be replaced with one having more heating surface. A traditional heat balance can be developed at this point. However, it must be kept in mind that conventional heat balances are based on so many assumptions that they are of only limited value to the operator. The proper approach is to measure the conditions in the evaporator. Because of the numerous conditions which cause an evaporator to perform below calculated capacity, it is advisable to design with ample provision to take care of all possible situations ranging from low steam pressure to heavy scaling. REFERENCES 1
Staub, S. and Paturau, M. Principles of Sugar Technology. P. Honig (ed.), Elsevier, Amsterdam, 1963, Vol. Ill, p. 58.
81
Chapter 10
COMMERCIAL RAW SUGAR CRYSTALLIZATION Raw sugar of commercial standards (Hawaii 1980) requires crystals of reasonably uniform size 0.8 to 1.0 mm in length and 98.8 to 99.3 pol. Of other quality factors, color, ash and filterability, only color remains of major importance to the refiner. Marketable sugar meeting these specifications can be produced efficiently by crystallization from syrup in two steps. These are necessary because of the physical limits imposed by the centrifugal separation of the sugar crystals from the mother liquor or molasses. In the first step, A-massecuites utilizing syrup in the purity range 83 to 88, when more than 60% of the sucrose is crystallized, the massecuite becomes close to a solid mass which cannot be centrifuged. Crystallization is therefore kept below this point, and the molasses is returned for the second step or B-massecuites. In this crystallization, the limit for the percentage of sucrose in crystal form is lower because of the higher viscosity of the mother liquor, so that the maximum limit is not over 50%. The molasses from the B-massecuite is also returned for further crystallization, but the product cannot be used as marketable sugar. This C-massecuite sugar, or low grade, is too small in crystal size and too low in pol and must, therefore, be dissolved and used in the feedstock for the B-massecuite or, in a practice which is disappearing, as a magma of seed crystal for the B-massecuite. VACUUM PAN CRYSTALLIZATION In generating sugar crystals of relatively uniform commercial size, use is made of the fortuitous property of sucrose in forming extraordinarily stable supersaturated solutions. This reflects the reluctance of concentrated sucrose solutions to develop crystal nuclei, a property making it possible to establish a fixed degree of supersaturation in a footing of sugar liquor. Then a predetermined number of nuclei, as seed crystals, can be added, and by maintaining the supersaturation relatively constant with simultaneous feeding of sugar liquor and by evaporating, crystals of the average desired size can be grown. Pan operations are normally conducted at a standard constant vacuum of 125 mm Hg (5 in. absolute pressure). A-Massecuite A-massecuites are boiled on an initial footing of evaporator syrup. Syrup purities will vary from time to time, the usual range being from 80 to 90 (refractometer pol purity). In order to have relatively uniform conditions for seeding, it is generally advisable to establish a standard purity of the graining charge - one that can be maintained most of the
82
time. A reasonable value in most locations is 85. In case the syrup is of higher purity, Α-molasses is added to the footing syrup. Of course, if the syrup purity is below 85, nothing can be done but adjust the standard graining procedure. The footing is then concentrated to the degree of supersaturation optimum for establishing the grain. The degree of supersaturation can only be sensed by indirect means. Those commonly used are: Refractometer solids Boiling point elevation Electrical conductivity Consistency Refractometer Solids, as measured by a pan refractometer, is a simple method of estimating the supersaturation. Data are available in the literature showing the effect of temperature and purity on the 1 refractometer readings at various supersaturations (Figs. 10-1 and 10-2).
62
64
66 68 70 72 TEMPERATURE °C
74
76
78
Fig. 10-1. Purity-refractometer solids relative to temperature at saturation.
80
83
TEMPERATURE ° C
Fig. 10-2.
Purity-refractometer solids relative to temperature at 1.2 supersaturation.
Since the refractive index is highly temperature temperature correction or compensation is essential.
sensitive,
good
Boiling Point Elevation (ΒΡΕ) is difficult to measure and for that reason is the least used method. Proper instrumentation requires a sensing of the vapor temperature at the surface of the boiling massecuite compared with the temperature of boiling water at the same absolute pressure. Literature data relating ΒΡΕ to temperature and purity are available, however, and make this a useful technique (Fig. 10-3). Electrical Conductivity is simple and is reasonably self-compensating with regard to temperature. There is a problem in locating the electrodes and keeping them free from scale. Its sensitivity to ash content makes it difficult to use in areas where there are wide variations in the inorganic salt content of the incoming juice.
84
Fig. 10-3.
Nomograph of boiling point, refractometer solids and purity relationships at 1.3 supersaturation.
Consistency, up to the time of seeding, is a measurement of viscosity. Like conductivity, it is somewhat temperature self-compensating. Since the sensing rotor of a consistency probe is affected by the quantity of vapor bubbles if placed above the tube sheet and by the movement produced by a mechanical circulator if placed below the tube sheet, a suitable location is difficult to find. A serious disadvantage of the method is that viscosity is increased greatly by small quantities of dextrans and gums, and a false indication of supersaturation is given when deteriorated cane is being processed. In choosing pan instrumentation for seeding, it would be desirable to make a selection that could be used also after the grain is established. The possible choices are electrical conductivity and consistency, since refractometer solids and boiling point elevation both measure only the molasses and do not make necessary allowance for the effect of the crystals which is governing near the end of the strike. Both electrical
85
conductivity and consistency are sensitive to the crystal content, but they suffer the drawbacks described. The conclusion is that it is better to use one type of instrumentation for seeding and another type for the remainder of the strike. An effective system is the use of refractometer solids for seeding and consistency for the remainder of the boiling. Regardless of the type of instrumentation adopted, empirical relationships between the apparent supersaturation and the instrument readout must be established under the conditions obtaining. Considerable trial and error experience is thus essential. In addition to sensing and control instruments for supersaturation, it is important to have automatic absolute pressure control in order to maintain stable temperature conditions. Massecuite level sensing is useful for programming the changes inherent as crystallization continues. These include purity decrease of the mother liquor, which means an increase in viscosity, increase in crystal content and effect of the massecuite pressure head. Automatic steam pressure control is advisable also. The supersaturation at seeding point for an 85 purity footing should not exceed 1.15 to be in a zone safe from spontaneous nucleation. This corresponds to about 80 refractometer solids. A t this reading, the seeding slurry is added. This is prepared by grinding good quality sugar in isopropyl alcohol in a ball mill to a uniform size (see Chap. 12). The amount of crystals necessary for the completed strike can be calculated as shown in the example at the end of this chapter. Experience has shown, however, that 2 to 3 times the calculated amount must be used to give the right number of crystals. The reasons for this are multiple. The average size of the seed particles made by the method shown is 4.5 microns. Particles so small tend to conglomerate on standing. Another factor seriously reducing the effective number of particles is the flash that occurs when the alcohol slurry is introduced well above its boiling point. In the flash, some of the particles cake together; others are lost by sticking to the tubes and to the sides of the pan. Finally, some are dissolved when they reach superheated zones near the tube walls. The correct amount of seed must therefore be determined experimentally. It will vary with the type of pan and the manner in which the pan is operated. Once the correct quantity is determined, however, the procedure can be standardized. After seeding, the steam to the calandria should be shut off while the grain xbecomes established. This takes about 5 minutes, which allows the introduced crystals to reach a size where they have enough surface to absorb the sucrose introduced when the syrup feed is started. After the waiting period, steam flow is resumed but only at a reduced rate. A t this point water may be introduced for a short time to insure against reaching too high a supersaturation. When the crystals have grown to a size that they appear close together, automatic syrup feed can be started. This is a critical time and feeding must not be forced at a rate faster than the crystals can accept the sucrose, or new crystals may develop. Feeding is continued until the pan is full (1.5 m above the tube sheet). Massecuite is then cut to the finishing pans or a massecuite holding vessel.
86
The final sugar crystals will never be of uniform size, not only because the original ground seed particles lack uniformity, but also because apparently uniform crystals grow at different rates. This is caused by non-uniform conditions in the pan and also by the fact that crystals with molecular dislocations grow at a faster rate than those without. Fast growing crystals tend to be more uniform in size than slow growing ones for this reason. Also, regardless of how carefully conditions are controlled, there will always be some new nuclei formed in the boiling at all stages and stray crystals will appear. The best that can be done is to keep the quantity of these to a minimum. Good circulation is the major factor in preventing a large number of stray crystals. Normally, the volume of the massecuite at striking time must be a minimum of 8 times the volume at seeding to give crystals of average standard size. When the crystal size is suitable, feed is stopped and evaporation is permitted to continue until the correct solids content for centrifuging is reached. The refractometer solids at this point is approximately 92.5. The massecuite should have a crystal content above 50%. With a corresponding pan drop of close to 20 points, the A-molasses from the centrifugals will be in the range of 65 purity. Since remelt (dissolved C-sugar) is a mixture of relatively pure sucrose crystals and final molasses, it should not be used in the Α-strikes but should be sent to the B-strikes. In this manner about 50% of the total commercial sugar is not subject to crystallization in the presence of final molasses. Quality, in particular color, should therefore be better. B-Massecuite The purity at which B-massecuites are boiled depends on the purity of the Α-molasses and remelt. With an Α-molasses purity near 65 and remelt in the 80 to 85 range, the B-massecuite purity can be at the 75 level. A suitable footing is usually remelt. Pan procedures are similar to those for Α-strikes. Viscosity is greater and rate of crystallization is slower so more time is required. Because of the lower purity, crystal content will be less and a target of 40% is reasonable. Refractometer solids at the end of the strike should be at the 93 level. Crystal Content and Pan Drops It is important to obtain the highest yield practicable at each stage of the crystallization process. The yield is the percent of the sucrose in the massecuite which is in crystal form, as shown by the formula: massecuite purity - molasses purity 100 - molasses purity
m
a
s
s
e
c
u ei
f r a c to m e t e r solids
tr e
Massecuite purity minus molasses purity is usually called pan drop. This figure is the one commonly observed in comparing pan work, but the crystal content is a more useful value because of the sensitivity of yield to purity levels. For example, compare the crystal content at the same pan drop (20) for massecuites of 80 and 85 purity.
87
Crystal content 80 Purity = 85 Purity =
?n — — χ 92.5 = 100 - 60 — — χ 92.5 = 100-65
46.3 52.9
So the higher the purity the less the pan drop that is necessary to obtain the same yield. As a rough guide, it is not necessary to multiply by the refractometer solids so relative yields may be noted from the ratio of the pan drop to 100 minus molasses purity. Vacuum Pan Design For efficient operation, the vacuum pan must be designed for good circulation, essential to both rapid and uniform crystallization. In calandria pans, natural circulation by expulsion of the massecuite upward through the tubes by the action of expanding bubbles of vapor, and drainage down through the center well give effective distribution in a well-designed vessel. Circulation can be improved, nevertheless, by means of a mechanical circulator at the bottom of the center well. The critical points in design are short tubes, low head of massecuite above the tube sheet at time of striking, minimum volume between the bottom and the tube sheet, belt diameter the same as the calandria and proper heating surface in the tubes. Optimum length for standard 10 cm (4 in.) diameter tubes is 75 cm (30 in.). In longer tubes, the percolation effect is diminished. Shorter tubes do not provide enough heating surface. The tubes should be arranged around a center well of approximately 40% the diameter of the pan. Circulation diminishes as the level of massecuite above the tube increases, so the level at the time of striking should not be over 1.5 m (5 ft). This volume capacity should not be obtained by flaring the pan to a greater diameter above the tube sheet (so-called low head design) because the inventory of massecuite beyond the tube sheet suffers from poor circulation. The belt, therefore, should be the same diameter as the calandria. The pan bottom should be streamlined so that the massecuite flow is directed from the center well outward with a minimum volume of massecuite below the tubes. The necessary heating surface of the tubes depends to some extent on the pressure of the heating vapor used. In pans boiled on vapor from the 2 2 first evaporator effect, the pressure is a normal 0.5 k g / c m (7 lb/in. ). A good design would, in this case, provide a ratio of heating surface to 2 3 2 3 volume of 5 m / m (1.5 f t / f t ) . The location for a mechanical circulator is on the center line of the bottom tube sheet. The design should not be that of a ship's propeller to give forward thrust but to force the massecuite laterally under the tubes.
88
The speed depends upon the diameter but must be left low, 40 rpm is 3 optimum for a 600 hi (2000 f t ) pan. Procedures Instrumentation - Semiautomatic control - Graining by refractometer Boiling by consistency or conductivity 1. Automatic absolute condenser. Recorder. 2.
pressure
control
by
flow
of
water
to
Pan refractometer. Recorder.
3. Automatic feed control, sensing by means of consistency or electrical conductivity to maintain maximum supersaturation. Recorder. 4.
Automatic level control. Recorder.
5.
Pan vapor thermometer. Indicator.
6.
Pan microscope.
7.
Pressure gauge calandria. Indicator.
Operation - typical example 1. Draw in footing of just sufficient volume to cover the tube sheet when evaporated to the seeding point. Footing should be syrup if its purity is 85 or below. If syrup purity is above 85, add enough A-molasses to lower the footing purity to 85. 2. sure.
Concentrate to 80 refractometer solids, at 125 mm absolute pres-
3.
Shut steam valve. Keep mechanical circulator running.
4.
Add 30 ml slurry per 283 hi (1000 f t ) final massecuite.
3
5. When grain is established (5 min) open steam valve half normal and start syrup feed. 6. After 10 minutes, open steam valve to normal and maintain supersaturation of 1.2 until massecuite is 1.5 m above the tube sheet. 7.
Cut seed to striking pans or seed storage.
8. Continue boiling in striking pans until crystal reaches average size of 0.8 mm feeding Α-strikes with syrup and B-strikes with Α-molasses and remelt. If crystals have not reached 0.8 mm when pan volume reaches 1.5 m above the tube sheet, another cut must be made. 9. Drop strikes at 92.5 and 93.0 refractometer solids, respectively, for the A - and B-strikes. Calculation 3
Seed slurry required for 283 hi (1000 f t ) massecuite. Massecuite: 3
283 hi (1000 f t ) 93 refractometer solids
89
Crystal content 55% refractometer solids Crystal size 0.8 mm Weight of massecuite at 1.4 kg/1 = 39,620 kg Weight of crystal in massecuite = 20,265 kg 39,620 χ 0.93 χ 0.55 Seed required: 20,265 χ (M045)3 \ 0.8 ' 20,265 χ 0.0000001778 = 0.0036 kg or 3.6 g Slurry has 1000 g in 2 liters alcohol Total volume = 2630 ml 1000 ·+ 2000 = 630 + 2000 1.587 lmlhas
3
S - S =0
38
6
' = 9.5 ml 0.38 Estimated practical quantity is 3 times this or 28.5 ml (round number 30). Quantity of Massecuites The quantity of massecuites boiled is a minimum in a straight-forward boiling system with the only return being C-sugar to the B-massecuites. Any inboiling, that is, return of low purity molasses to higher purity massecuites, increases the total. The quantity also becomes greater as the syrup purity decreases. The following calculations illustrate this effect in examples of 85 and 80 purity syrups. For simplicity, the sugar has been calculated at 100 pol and no losses in processing are included. Straight A , B, C boiling is used with C-sugar returned to B-massecuites. Basic assumptions include:
85 Purity syrup 80 Purity syrup
A-Massecuite crystal yield
C-Massecuite purity
50 45
58 56
85 Purity syrup Tons Syrup Refractometer solids Pol Nonsugars
100.00 65.00 55.25 9.75
Final molasses purity 35 36
90
A-Massecuite Crystal yield 50% assumed Sugar crystal, tons = 27.63 0.5 χ 55.25 Refractometer solids to B-massecuite, tons = 37.37 65.00 - 27.63 Α-molasses purity = 70 85 - purity u 100 - purity '° C-Massecuite Assume:
58 purity massecuite 35 purity final molasses Crystal yield = 35% ό
100 - 35 "' * Nonsugars, tons =9.75 Pol, tons = 13.46 Pol - π c o 5 8 P o l + 9.75 " ° Refractometer solids, tons = 23.21 9.75 + 13.46 Sugar crystal, tons = 4.71 0.35 χ 13.46 Sugar (solids) at 80 purity returned to B-massecuite, tons = 6.81 4.71 0.692 100 \ 35
/
80
45 / \ 20
x
1
00 =
6
20 χ
1
00 =
3
ët
9
·
2 %
0
%
c r
Y
m
o
s t al
l
a
s
solids
ss e
s o l i sd
D D
Molasses (solids) returned to B-massecuite, tons = 2.10 6.81 - 4.71 Pol returned to B-massecuite, tons = 5.45 4.71 + (0.35 χ 2.10) Final molasses Nonsugars, tons = 9.75 Pol, tons = 5.25 _P£l = 0 35 U e DJ Pol + 9.75 Refractometer solids, tons = 15.00 9.75 + 5.25
91
B-Massecuite Tons Refractometer solids from A-massecuite Refractometer solids from C-massecuite Total Pol from A-massecuite Pol from C-massecuite
37.37 6.81 44.18 27.63 5.45
Total
33.08
Purity = 74.9 33.08 χ 100 44.18 40.2% Crystal yield 74.9 - 58 0.402 100 - 5 8 Quantity of massecuite Tons refractometer solids A-massecuite B-massecuite C-massecuite
65.00 44.18 23.21 Total
132.39
80 Purity syrup Tons Syrup Refractometer solids Pol Nonsugars
100.00 65.00 52.00 13.00
A-Massecuite Crystal yield 45% assumed Sugar crystal, tons = 23.40 0.45 χ 52.00 Refractometer solids to B-massecuite, tons = 41.60 65.00 - 23.40 Pol to B-massecuite, tons = 28.60 52.00 - 23.40 Α-molasses purity = 63.6 80 - purity 0.45 100 - purity C-Massecuite Assume
56 purity massecuite 36 purity final molasses
Crystal yield = 31.3% 56 - 36 = 0.313 100 - 36
92
Nonsugars, tons = 13.00 Pol, tons = 16.55 Pol ηcß 5 6 Pol+13.00 - ° Refractometer solids, tons = 29.55 13.00 + 16.55 Sugar crystal, tons = 5.18 0.313 χ 16.55 Sugar (solids) at 80 purity returned to B-massecuite, tons = 7.53 5.18 0.688 100 Ν
36
44 Ü
44 80
x
1
00 =
6
8
·
8 %
c r
y
s t al
s o l l sd
/
~
| ^ χ 100 = 31.2% molasses solids
Molasses (solids) returned to B-massecuite, tons = 2.35 7.53 - 5.18 Pol returned to B-massecuite, tons = 6.03 5.18 + (0.36 χ 2.35) Final molasses Nonsugars, tons = 13.00 Pol, tons = 7.31 =
3 6
Pol+13.00 °" Refractometer solids, tons = 20.31 13.00 + 7.31 B-Massecuite Tons Refractometer solids from A-massecuite Refractometer solids from C-massecuite Total Pol from A-massecuite Pol from C-massecuite
Crystal yield i l d = 33.0% 70.5 - 56 = 0.330 100 - 5 6
49.13 28.60 6.03
Total Purity = 70.5 34.63 χ 100 49.13
41.60 7.53
34.63
93
Quantity of Massecuite Tons refractometer solids A-massecuite B-massecuite Omassecuite
65.00 49.13 29.55 Total
143.68
These calculations show that with 85 purity syrup the quantity of solids handled is about double that incoming with the syrup. A t 80 purity it is of the order of 10% additional. Most of the increase is on the C-massecuite where the solids handled is greater by 27% and the final molasses is greater by 35%. REFERENCES 1
Honig, P. (ed.), Principles of Sugar Technology, Vol. 2, Elsevier, Amsterdam, 1959, pp. 358-359 (from Thieme).
95
Chapter 11
LOW GRADE SUGAR CRYSTALLIZATION Traditionally, the final step in sugar recovery is allowed to take place by cooling in crystallizers rather than during evaporation in vacuum pans. The reason for this is that the rate of crystallization becomes progressively slower as the molasses purity falls. A more cost effective procedure, therefore, is to boil low grade massecuite in a vacuum pan for a limited period of time, then discharge it into atmospheric crystallizers where sugar can be allowed to crystallize at length without evaporation. CRYSTALLIZATION BY COOLING The technology of crystallization by cooling differs markedly from that of crystallization by evaporation, and although crystallizers appear to require little attention, success in recovery depends to a great degree on the details of their operation. Two major factors are important in the technique. 1. A t a temperature range of 50 to 60° C and below, the rate of diffusion of sucrose molecules to the surface of a crystal exceeds the rate of deposition on the crystal; whereas at higher temperatures, the rate of diffusion is less than the rate of deposition (Dedek). 2. In this same temperature range (50-60° C), saturated molasses (and 1 supersaturated molasses also) has a minimum viscosity (Fig. 11-1 J. Application of these following conclusions:
facts
to crystallizer operation leads to
the
1. Mixing in the crystallizer is necessary only to prevent the crystals from settling and to aid in heat transfer. 2. Maximum crystal surface sucrose deposition. 3.
is essential
to a favorable rate of
The optimum holding temperature is in the range of 50 to 60° C.
These points must be superimposed upon the one basic fact that water is the principal component that keeps sucrose in solution. A t a given temperature, the saturated molasses purity varies directly with the water content. In order to obtain a lower purity molasses water must be removed, increasing the saturation temperature. But as water is removed molasses viscosity rises sharply, giving proportionately higher massecuite consistency. As the crystal content is relatively constant, molasses viscosity becomes the limiting factor. Since consistency must be kept at a level that permits massecuite pumping and separation of crystals from the molasses by centrifuging, molasses viscosity must be carefully controlled.
96
Fig. 11-1.
Viscosity-temperature relationships for saturated molasses at varying water content.
With the optimum viscosity-saturation relationship occurring in the 50 to 60° C temperature range, the water content of the massecuite should be targeted to give the maximum consistency of massecuite practicable. Such a practice will give the lowest attainable molasses purity for a given material composition in a fixed period of time. The control procedure for the crystallizers, therefore, is determined by the composition and physical character of the massecuite discharged from the pans. Crystallizer Design Design criteria for a crystallizer are directed toward heat transfer from and to the viscous mixture of sugar crystal and molasses. The most
97
effective design is that of rotating pipe coils through which water is circulated. These give enough movement to prevent crystals from settling and provide reasonable opportunity for heat transfer. They are more effective for heating than for cooling. When the coils are hotter than the massecuite, the molasses viscosity is less and the massecuite slides off the surface permitting access of cooler material. When the coils are cooler than the massecuite, the cooled material at the surface of the coil becomes more viscous and clings to the coil insulating it from the body of the massecuite. For this reason plate-type crystallizers are not effective for cooling heavy massecuite. The coils must be far enough apart that massecuite is not carried as a body between them. The coils must move through the massecuite giving time for exposing new material to them. The coils must be heavy and well supported. Since little agitation is required, a speed of 12 rph is sufficient. Massecuite level should be above that of the coils to prevent incorporation of air which causes an increase in the viscosity of the massecuite. Operation Procedure Instrumentation. 1. 2.
Thermometers. Recorder. Drive motor ammeter. Alarm.
Batch crystallizers. 1.
Cool the massecuite to the range 50-55° C. In an efficient water-cooled crystallizer, this will take some 15 hours. With cooling water at ambient temperature, it is practically impossible to cool a properly boiled massecuite rapidly enough to cause spontaneous new grain.
2.
Hold the massecuite at the chosen temperature by circulation of water at a slightly higher temperature for a minimum of 15 more hours. After this period, the supersaturation and molasses purity have become so low that the crystallization rate is exceedingly slow. It is, therefore, of little practical value to hold longer, although it can be done if capacity is available.
3.
Start reheating in the crystallizer by circulation of about 60°C water. This can be done for a variable period of time depending upon the need for the crystallizer for subsequent strikes and the availability of massecuite heaters. The object of reheating is to reduce the supersaturation to close to zero before centrifuging. With the use of massecuite heaters, reheating in the crystallizer can be kept to a minimum sufficient to improve the flow of massecuite from the crystallizer. Otherwise, several hours reheating will be necessary. The temperature to which massecuite is reheated, either in the crystallizer or heater, will depend upon the
98
saturation temperature of the molasses. Since it is not practical to determine the saturation temperature on each strike of massecuite, a general temperature is chosen, based upon periodic measurements on typical samples and experience. Continuous crystallizers The same principles hold for the operation of continuous crystallizers with necessary variations in procedure required by the individual installation. For continuous operation, the massecuite must be discharged from the pan into a strike receiver which acts as a reservoir to feed the bank of crystallizers. The preferred number of units in a bank is 7, permitting 2 or 3 for cooling, 2 or 3 for holding and 2 for reheating. In order to make coil repair possible without emptying two units, flow is from the top at one end, out the top of the other. Surface flow is prevented by a horizontal baffle plate located one third of the way from the entry point and extending not more than 50% of the depth of the massecuite. Troughs between the units should be wide and shallow to give a channel rather than a tube-like flow. A possible water flow diagram is shown in Figure 11-2. Cooling water, at the chosen temperature, flows through the second crystallizer and then through the first. The reason for counter current flow is to avoid, to some extent, too rapid initial cooling causing solidification around the coils and bypassing of massecuite. Heating water enters the last body and flows through the next to last, then out to a tank where it mixes with cooling water and is temperature controlled for holding water which goes through the third, fourth and fifth bodies in series. Temperature control is required only at two points - the holding water and the heating water. Quantity of water must be controlled relative to the flow of massecuite. In the operation of continuous crystallizers it is important to keep in mind that the retention time is a function of the rate of flow, which is set by the quantity of massecuite that must be handled per unit of time. The quantity of low grade massecuite is governed largely by the purity of the syrup. As an example of the wide variation, the quantity produced from 80 purity syrup would be about double that at 88 purity per unit of pol in the syrup. Therefore, unless the rate is controlled, the retention time with massecuite from 80 purity syrup would be only half that from 88. For optimum operation, therefore, the flow through the continuous system must be monitored and controlled. VACUUM PAN CRYSTALLIZATION Low grade boiling procedures are set by the crystallizer operation discussed above, the objective being to provide a massecuite of a consistency and purity which will give maximum recovery in the crystallizers.
99
Ο m LÜ
U
Ο LÜ
3
Lü
Hl
< 5 ο χ
CO
(Τ Ο
ΙΟ
"Mr (Τ LÜ *< ο
Ο Ο
If)
C M
rÜJ
ϋ Ο LÜ
ω <
^
Γ
λα
ζ *
— ο ζ _ι ο ο ο
Fig. 11-2. Continuous crystallizer bank.
ΓΟ
tr
LU Η <
100
Consistency is governed by the viscosity of the molasses, crystal content and the size of the crystals. Because of the slow rate of growth, the size of the crystals is usually limited to the range of 0.20 to 0.30 mm, with workable average being 0.25 mm. The crystal content, which is the percent of the total solids in crystal form, has a practicable range of 25 to 40, depending upon the purity of the massecuite. The viscosity of the molasses, at saturation, depends upon the quantity and nature of the nonsucrose matter and the temperature. Temperature, therefore, is the only controllable variable. Since the minimum viscosity is in the neighborhood of 55° C, the molasses should be concentrated in the pan to a water content that will make it saturated at 55° to 58° C. Because the composition of the raw material (mixed juice) may vary from hour to hour, field to field, and week to week, it is not possible to maintain a constant quality massecuite by control of pan work. Experience alone will enable a pan operator to judge at striking time, 70°C and unknown supersaturation, what the consistency of a massecuite will be many hours later at 50° C and minimal supersaturation. He will, therefore, try to have the solids content at an average as high as possible to give a massecuite that can be handled in the crystallizers, pumps, heater and centrifugals. Low grade vacuum pan work should thus be directed to giving an average massecuite at the time of centrifuging of the following composition: Purity Crystal content (refractometer solids) Crystal size Molasses, saturated at
58 30-35% 0.25 mm 55°-58° C
The purity at which the massecuite is boiled is dependent to a major extent on the purity of the B-molasses. Starting with syrup purity of 85, with standard pan drops on A and Β strikes, the B-molasses purity should not be higher than 56. This permits a massecuite purity of around 58. In order to give the seed a fast start, the footing should be a few points higher than the final massecuite. In a fully counter current system, this should be made up of B-molasses and remelt. As it is inconvenient to make this mixture, Α-molasses is often used because it is about the right purity. The footing should be concentrated to a supersaturation of 1.2, then the estimated quantity of seed slurry is added (see Table 11-1). With a mechanical circulator to maintain good movement in the pan, the steam should be shut off while the seed is becoming established. Otherwise there is danger of concentrating the footing so much that spontaneous nuclei may be formed. After a waiting period of about 10 minutes, steam can be opened. A t this point water may be introduced until vigorous boiling results before starting B-molasses feed. Automatic feed control can be programmed to allow the supersaturation to be held as high as possible, governed by the consistency
101 Table 11-1 MOLASSES Refractometer Solids Supersaturation 1.20 Refractometer Pol Purity Tempérât ure U C
60
62
64
66
144
62.3
83.5
83.2
82.8
82.5
146
63.4
83.8
83.5
83.2
82.5
148
64.5
83.9
83.6
83.3
83.0
150
65.5
84.1
83.8
83.5
83.1
152
66.8
84.2
83.9
83.6
83.3
154
67.9
84.3
84.0
83.8
83.5
156
69.0
84.5
84.2
83.9
83.6
158
70.1
84.6
84.4
84.1
83.8
160
71.2
84.8
84.5
84.2
83.9
162
72.3
84.9
84.7
84.4
84.1
164
73.5
85.1
84.8
84.5
84.3
166
74.6
85.3
85.0
84.7
84.4
168
75.7
85.6
85.3
85.0
84.7
170
76.8
85.7
85.4
85.2
84.9
which should never be allowed to rise above the point of good circulation and vigorous boiling. Particularly toward the end, when the massecuite level is high, circulation becomes sluggish. Water addition is often advisable to improve the circulation by ebullition. During all boiling, the absolute pressure should be kept constant at about 125 mm Hg. (5 in. absolute pressure). Higher temperature improves circulation and speeds crystallization. Temperatures above 70° C are not advisable, however, because of excessive inversion and propagation of reactions of reducing sugars and amino acids which cause excessive frothing in crystallizers and final molasses. When pan capacity is reached, which should not give a massecuite level in excess of 1.5 m above the tube sheet, the massecuite is cut to finishing pans. There boiling is continued until the crystals are of sufficient size and the pan drop target is reached. A t striking time, a molasses purity of 42 would mean a pan drop of 16 points. This would give a crystal content of 26%. The average size of the crystals should be larger than 0.2 mm.
102
Vacuum Pan Design Pans for low grade boiling are basically the same as for A and strikes. The only difference is that because of the much slower rate crystallization the heating surface does not have to be as large. 2 3 2 3 nominal ratio of heating surface to volume of 4 m / m (1.2 f t / f t ) standard.
Β of A is
Mechanical circulators should run at a lower speed because of the higher viscosity massecuites. Operation Procedures Instrumentation (semi-automatic control) 1. 2. 3. 4. 5. 6.
Automatic absolute pressure control. Recorder. Pan refractometer. Recorder. Automatic feed control, sensing by means of consistency or electrical conductivity. Recorder. Automatic level control. Recorder. Pan vapor thermometer. Indicator. Pan microscope.
7.
Pressure gauge, calandria. Indicator.
Typical Example 1. Draw in footing (B-molasses plus remelt or A-molasses) sufficient in volume to just cover the tube sheet when evaporated to the seeding point. Purity of footing 65. 2. Concentrate rapidly to 84 refractometer solids at 125 mm absolute pressure. 3. Shut steam valve. Keep mechanical circulator running. 3 4. Add 600 ml seed slurry per 283 hi (1000 f t ) final massecuite. 5. When grain is established (10 min.), open steam valve and start B-molasses feed. (B-molasses purity is estimated to be 56.) 6. Maintain maximum supersaturation consistent with vigorous boiling (1.3) until massecuite is 1.5 m above tube sheet. If boiling is not vigorous near the end, feed sufficient water to maintain it. 7. Cut seed to striking pans or seed storage. 8. Continue boiling in striking pans until crystal reaches the minimum size of 0.2 mm. 9. Drop strikes at a refractometer solids that will give a saturated molasses at 55°-58°C (95 massecuite refractometer solids normally). Calculations 3
Seed Slurry Required for 283 hi (1000 f t ) Massecuite Massecuite 3
283 hi (1000 f t ) 95 refractometer solids
103
Crystal content 35% of refractometer solids Crystal size 0.25 mm Weight of massecuite at 1.5 kg/1 = 42,450 kg Weight of crystal in massecuite = 42,450 χ 0.95 χ 0.35 = 14,115 kg Seed Required (theoretical) 0
4 5
3
0.0000058 = 0.082 kg or 82.0 g 14,115 χ ( ° ό ' 2 ° 5 ) = 14,115 χ Ο.ι Slurry has 1000 g in 2 liters alcohol. Density of sugar = 1.587 g/ml Total volume = 1 ml has .'. Require
1000 + 2000 = 630 + 2000 = 2630 ml 1.587
1000 = 0.38 g 2630 82.0 = 216 ml 0.38
Practical estimate = 3 χ 216 = 648 ml Nominal value = 600 ml REFERENCES 1
Micheli, L. I . A . and DeGyulay, O. S., Proc. Intern. Soc. Sugar Cane Tech. 6, 1938, p. 1094.
105
Chapter 12
GENERALIZATIONS AND DATA ON SUCROSE CRYSTALLIZATION GENERALIZATIONS Temperature A 10°C increase in temperature increases rate of crystallization 3.5 times. Supersaturation At 50° C, an increase in supersaturation from 1.05 to 1.15 increases the rate of crystallization 5 times. A change in refractometer solids of 1.0 changes supersaturation by 0.06. Temperature and Supersaturation Same rate of crystallization * Temperature
Supersaturation
°C 70 60 50 40
1.25 1.30 1.35 1.40
Viscosity A 5°C decrease in temperature doubles the viscosity of final molasses. The viscosity of massecuites is 3 to 6 times that of the suspending 2 molasses. Also, the following are approximate estimates of the increment required of a variable to cause a 20% decrease in viscosity: 0.4% decrease in solids 2° C increase in temperature 3% decrease in crystal content Threefold decrease in mean crystal size 10% increase in purity 100% increase in coefficient of variation of crystal size
106
Purity Crystallization rate relative to pure sucrose at 1.09 supersaturation Purity
Relative rate
100 92 80 70
100 59 14 2
3
Recirculation of Final Molasses with Return of Low Grade Sugar Sugar purity
Relative increase over 90 purity %
86 82 79
60 140 245
Low Grade Seeding ml of seed slurry = Seed Slurry
f
t 3
"assecuite
3
Preparation 1000 g sugar 2 liters isopropyl alcohol Grind 24 hr in a jar mill of 4-liter capacity with 4 kg of 20 χ 20 mm cylinders. Average particles Number: Size:
4.8 χ 10^ particles/ml slurry 4.5 microns
Crystal Yield Crystal yield % = ref. sol. mass, χ Pur. mass. - pur mol. J J 100 - pur. mol. Exhaustion Exhaustion is parts of sucrose in crystal per 100 parts sucrose in massecuite. Λ The crystal yield formula must be multiplied by 100 % pol in mass.
107
. „ . , ref. sol. mass. pur. mass. - pur, mol. c . . Exhaustion -r-, Ί . χ *•— : % pol in mass. 100 - pur. mol. 2
lOO pur. mass.
x
pur, mass. - pur, mol. 100 - pur. mol.
Perk^ says this figure should be over 60 irrespective of the type of massecuite. DATA Single Crystal Measurements
5
Screen opening mm
Weight mg
Surface mm^
0.5 0.6 0.7 0.8 0.9 1.0
0.139 0.240 0.381 0.568 0.809 1.110
1.22 1.75 2.39 3.12 3.94 4.87
Crystal Growth Rate, mg/m^/min.G
\
Super-\ saturation\ 1.005 1.010 1.015 1.020 1.025 1.030 1.035 1.040 1.045
Temperature ° C 0
20
30
40
50
60
5
40 80 120
75 150 225
145 285 490
240 490 800
340 720 1340
150 190 230
380 495 625
675 855 1060
1200 1800 2300
2210 3100
275 320 360
755 910 1115
1300 1540 1800
2870 3510 4060
420 480 525
1320
2085 2580
9 14 19
1.050 1.055 1.060
32
1.065 1.070 1.080
38 45
1.090 1.100 1.110
52 62 76
25
575 620
108
Fig. 12-1. Effect of purity on crystallization rate.
3
109
MASSECUITE
TOTAL-SOLIDS
SUCROSE
CRYSTAL CONTENT
NON-SUCROSE
VISCOSITY
CIRCULATION
MOBILITY
SUPERSATURATION
TIME
CRYSTALLIZATION RATE
MOLASSES EXHAUSTION
Fig.
TEMPERATURE
12-2. Variables governing molasses exhaustion.
CRYSTAL SURFACE A R E A
110
SYRUP 85
Β REMELT 90
C
MASSECUITE 76
MASSECUITE 88
CO
-Jin
- J 1^o
ο Σ m
5
COMMERCIAL SUGAR 9 9 . 3 POL
MASSECUITE 60 Id CO CO < CO
to
CENTRIFUGAL
REMELT 85
Β CENTRIFUGAL
CENTRIFUGAL
\ΛΛΑΛΑΑΛΑΛΛ
vyvWvWWT ο
F I G U R E S ARE REFRACTOMETER POL P U R I T I E S
Β
C REMELT 85
REMELT 90 FINAL MOLASSES 35
Fig.
12-3. 3-Massecuite, 1-high pol sugar boiling system.
Ill REFERENCES 1 2 3 4 5 6 7
Saint, John, Int. Sugar J., 35 (1933) 311. Awang, M. and White, E. T., Proc. Queensland Soc. Sugar Cane Tech., 43 (1976) 263. 1 unpublished data, Experiment Station of the Hawaiian Sugar Planters Association. Perk, C. G. M., The Manufacture of Sugar from Sugar Cane, Sugar Milling Research Institute, Durban, 1973, p. 203. Wieninger, J. Α . , Jahresbericht Zuckerforschungs Institute, 1974-75, 78. Kucharenko, J. Α . , Planter and Sugar Manufacturer, 30 (1928) 485. McGinnis, R. Α . , Sugar J., 39 (1976) 7.
113
Chapter 13
CENTRIFUGATION Although the mechanics of the process run counter to apparently sound principles, sugar crystals are universally separated from molasses by centrifugation. Under centrifugal force the denser sugar is restricted from flying away from the lighter liquid molasses by means of a perforated screen. The molasses then finds its way through the spaces between the crystals and out through the perforations. The two types of machines in use are batch and continuous. BATCH MACHINES Commercial sugar Standard machines for commercial sugar have 1220 mm χ 915 mm (48 in. χ 36 in.) baskets driven by single-winding, pole-changing, two-speed, 600/1200 rpm, alternating current induction motors. A t full speed and at the center of the sugar cake, the nominal gravity factor is 900. A separate reverse drive small motor is used for low speed discharge. Fully automatic operation requires only the time settings for the successive steps of initial acceleration, charging, low speed acceleration, washing, high speed acceleration, running, regenerative braking, mechanical braking and discharge. These settings are determined by the characteristics of the massecuite and the sugar quality desired. A complete cycle normally takes about 3 minutes. Higher capacity machines, 1372 mm χ 1016 mm (54 in. χ 40 in.), are also available. It is generally accepted that the separation of molasses takes place in three stages. These are: 1. Removal of molasses in excess of that required to fill the space between the crystals. 2.
Further expelling of molasses leaving voids between the crystals.
3.
Reducing the molasses film around the crystals.
The rate of removal of molasses is rapid at first and very slow later on, governed by the size of the crystals and the viscosity of the molasses. 2 With crystals of uniform size the rate is proportional to m /n where m is the length of the crystals and η the viscosity of the molasses. Rate of molasses removal is less in sugars of nonuniform crystal size because of a packing effect. In extreme cases small crystals can almost completely block the passage of molasses, leaving a layer on the inside of the solid mass of sugar.
114
Water washing is necessary to reduce the molasses film in order to give the desired sugar pol. The effectiveness of the wash depends upon the method of application, quality of the sugar with respect to size distribution and form, quantity of water, temperature and timing of application. The water should be distributed uniformly over the surface of the sugar and should be hot (75° C or above). It should be applied at the moment the excess molasses is gone from the crystal but the space between is still filled with molasses (before the machine reaches top speed). This can be observed as the point at which the face of the sugar in the basket starts to lighten as the molasses leaves. The water then will give a plug type flow and uniform washing. If applied after voids are present between the crystals, then the water will seek the path of least resistance and leave some sugar over washed and some underwashed. The quantity of water and spinning time are adjusted to give the pol desired. In producing high pol sugar (99.3-99.5), close to complete washing is necessary. Two stages of application are effective in this case, the first as described, and a second just when the machine reaches top speed. The second wash quantity is adjusted to give the sugar pol required. Double washing also makes possible the separation of the wash liquors. Low Grade Sugar The small-sized crystals and high viscosity of the molasses make washing in batch centrifugals impracticable. Separation of molasses therefore requires use of higher gravity factors and longer spinning times than used for commercial sugars. Standard machines are 1016 mm χ 762 mm (40 in. χ 30 in.) giving a nominal gravity factor of 1800 at high speed. Larger machines - 1220 mm χ 762 mm (48 in. χ 30 in.) - are also available. The two-speed main drive motor is a nominal 900/1800 rpm induction type. Because of the sticky character of the massecuite, the discharge mechanism employs a "double-cut" system. Spinning time varies depending upon the viscosity of the molasses and the size and uniformity of the crystals. An average cycle time for Hawaiian massecuites is 30 minutes. Heavy massecuites may require double this. In regions with less viscous material, the cycle time may be 15 minutes or less. Uniformity of crystal size is extremely important in obtaining good molasses removal in batch machines. Even a small amount of fine crystals will form an almost impervious layer within the sugar and will prevent the molasses from reaching the screen. The molasses will then collect in a layer, called a mirror, on the inside of the sugar wall and will run down into the bottom of the basket when the machine is stopped. The same condition can occur when finely divided extraneous matter is present. A sometime cause is the accumulation of calcium magnesium aconitate in the form of diamond-shaped crystals which, when they reach a size of over 0.05 mm, can completely plug the sugar wall. Machine curb top should be provided with a cover to reduce windage and retain heat over long operating cycle of the machine.
115
CONTINUOUS MACHINES Commercial Sugar Continuous centrifugals have not proven to be satisfactory for commercial sugar because of crystal breakage and difficulty of producing a sugar of uniform high quality. It is possible to use continuous machines on intermediate sugars which are remelted. Low Grade Sugar Continuous centrifugals have largely replaced batch machines in most low grade installations. Continuous machines are much lower in first cost, have higher capacity and lower power consumption. They have the major disadvantage of giving a molasses purity increase across the machines. This averages between 1 and 2% in most equipment, but unless carefully controlled, can be substantially higher. Crystal breakage causes fines which pass through the screen with the molasses. The crystal breakage also renders the sugar unsuitable for the preparation of seed magma. Continuous machines also require more operator attention than batch machines as adjustments must be made in throughput rate and washing on the basis of visual monitoring. Continuous centrifugal baskets usually have a conical angle of about 30° from the vertical and are run at about 2200 rpm. Centrifugal screens are usually chrome plated and have perforations of two types - circular and slotted. Standard circular perforations are 0.125 mm in diameter. Slotted perforations are usually 0.06 mm wide and up to 1.6 mm in length. Although both types of perforations increase in size with wear, slotted perforations are subject to deformation into elliptical shape resulting in considerably greater enlargement. Rate of massecuite feed is usually controlled by the load on the V-belted drive motor. This is not a fully-automatic system as constant load does not mean constant quality of product. However, it does give a simple control and leaves quality in the hands of the operator. The machines usually have distributors for application of water and steam to the internal face of the cone. Both of these are difficult to keep in optimum flow so without close attention the performance - that is molasses removal without loss of sugar - varies widely. The most satisfactory results are obtained by the use of steam alone introduced into the massecuite feed pipe with none impinging on the screen. Optimum performance is found with massecuites heated to the maximum temperature permissible without solution of sugar. Operation Batch machines - Commercial sugar Timing of the steps in the cycle must be adjusted to characteristics of the massecuite which often vary with Particularly important is the time of application and amount mentioned before. Also, the final drying time must be set to of the proper deterioration factor for storage.
the physical each strike. of water as give a sugar
116
Batch machines - Low grade Only two controls are possible - time of spinning and temperature of massecuite. The time is often governed by that available to handle the volume of massecuite, so the cycle is adjusted accordingly. The massecuites should enter the machines at close to the saturation temperature. The machines should always be closed to diminish air flow and maintain temperature. Although steam is sometimes introduced outside the basket, it is not advisable because of possible machinery damage. Continuous machines - Low grade Constant attention is the key to good performance. The capacity of a machine is that obtained with a given massecuite when operated to give good quality sugar without excessive increase in molasses purity in the machine. Screens are very thin and so are subject to damage. Great care is required in mounting the screens to insure that they fit tightly against the backing so that no flexing occurs in operation. The massecuite should be screened to remove lumps of sugar and extraneous material such as bolts and nuts. A target for maximum molasses purity increase in the machines of 1.0, although rarely obtainable, is workable. Since the actual value is only available to the operator at wide intervals from laboratory analysis, observation of the sugar is the customary basis of control. With reasonably good quality crystal, at the size level of 0.20 mm, the sugar purity should be around 85. Decisions for adjustment of steam and throughput are usually made on the basis of color. A better guide is consistency. The sugar should be neither a dry powder nor have the consistency of massecuite. It should have the consistency of a soft brown sugar. An amusing but reliable measure is taste. If the sugar tastes sweet it is overwashed (steamed). If it is bitter it is underwashed. A brackish taste is about right.
117
Chapter 14
FINAL MOLASSES The principal loss of sucrose in the boiling house is that retained in the final molasses. This is not really a loss since molasses is a standard commodity and is sold at a price which at times exceeds that of the value of the sucrose content. The objective of the processor therefore is to keep the sucrose content of the molasses at near the economic optimum based upon the prevailing prices of sugar and molasses. As sucrose crystallization from final molasses proceeds, the rate of crystallization becomes slower and slower until finally a point is reached at which no more sucrose will crystallize at a given temperature. The molasses then is termed exhausted. The purity of exhausted molasses depends principally on the water content. At the same water content, however, the nature and quantity of the nonsucrose constituents, having their origin in the cane, govern the purity. Although all constituents exert an influence, only those present in large quantities have a major effect. Of these, reducing sugars and ash are controlling. These two ingredients have opposite effects upon sucrose retention. Reducing sugars, in essence, take the place of sucrose almost weight for weight, so the more reducing sugars present, the less sucrose will remain. Therefore, the higher the percent reducing sugars, the lower the exhausted molasses purity at a given water content. On the other hand, ash in general, and potassium chloride (the main ash constituent) in particular, tend to increase the solubility of sucrose giving a higher exhausted molasses purity. A statistically significant relationship exists between the reducing sugars-ash ratio and the purity of molasses at saturation, i.e., molasses from which no more sugar can be crystallized at a given temperature. This relationship is the basis of the system established by the Experiment Station of the Hawaiian Sugar Planters Association and used by factories in Hawaii as a basis for judging sugar recovery potential from final 1 These are 50° C molasses. Standard physical conditions are set. temperature and 600 poises viscosity. A standard viscosity is essential because of its controlling effect in the handling of massecuites, in the crystallizers and in the centrifugal separation of sugar. The following equations give the statistical relationship between the reducing substances and ash composition and the expected purity of molasses at 50° C and 600 poises. It is only necessary to analyze the molasses for reducing sugar and ash and then insert the figures into the equation to estimate the theoretical minimum purity - called "Expected Purity" - that it is possible to obtain from the molasses. The difference between this value and the purity actually obtained is called "Points Above Expected."
118
Expected refractometer sucrose purity = 33.160 -
4.028(
reducin
gy
ances
)+
0.161 (refractometer sucrose purity)
or 33.253 - 5-351 ( ^ n S Î i t y Μ ? ) 8
+
°'
1 34
<
r
e
f
r
a
c
t
o
m
e rt
e s
u
c
r
o es
P
u r i t
y>
In actual practice, Expected Purity, or the theoretical minimum purity of exhausted molasses, will not be reached in a factory because complete exhaustion may require as much as 30 days. For a given factory, how close it approaches expected purity depends upon many factors. The most important are the composition of the cane juice, the equipment such as vacuum pans, crystallizers and centrifugals available, and the technology used. In general, also, the higher the ash content, the slower the crystallization rate, so with a given capacity of equipment a factory processing high ash juice cannot achieve as low a Points Above Expected as one with low ash. The lower the purity of the incoming juice, the more molasses will be produced per unit sucrose and the greater quantity of low grade material that must be processed. For example, 80 purity juice will theoretically produce 2.4 times as much final molasses as 90 purity juice. So basically, the capacity of the low-grade station should be at least two times greater for the lower purity juice. With equipment limitations, high purity juice should make it possible to reach lower purity molasses more readily. Juice purities must therefore always be kept in mind when comparing losses in molasses as the total sugar lost depends not only on the purity of final molasses, but on its quantity. The question, then, is how close to the expected purity should be the target for factory operations. This depends upon the factors described above and on the economics of the entire operation. The relative prices of sugar and molasses must enter into the decision. As mentioned there are times when the value of a unit of sucrose in molasses exceeds that in sugar. This does not mean that it would pay to permit more sugar to go into molasses. Molasses is sold by the ton, therefore sugar must be accompanied by the normal amount of nonsucrose material to bring this return. However, it does mean that the cost of recovering sugar beyond a certain point may not be warranted. Experience leads to the conclusion then, that a factory should have the capability of processing final molasses to a level of 5 points above the expected purity with incoming juices of average composition. Poor quality juices will cause the molasses purity to increase because of the effects on molasses quantity, rate of crystallization and viscosity factors over which the factory has no control. Operational procedures being the same, the rate of crystallization of sugar from final molasses is slowed down sharply by high ash content. The
119
component largely responsible is potassium chloride. Aside from ion exchange treatment there is no effective means of removing this compound. A solution would be to install more equipment which would permit a longer crytallization time. However, in areas subject to a normal condition of high ash, it is advisable to maintain a routine daily measurement of the electrical conductivity of mixed juice so that the operating staff will be able to predict the crystallization rate. It has been found that there is a good correlation between the conductivity of mixed juice and the points above expected of final molasses obtained from that juice. Another rate-retarding factor is elongated crystal growth. In cane sugar processing this behavior is a characteristic of juices from deteriorated cane. This is distinct from beet sugar processing where the phenomonen is caused by raffinose. Small quantities of substances produced in the growth of microorganisms in the cane, reportedly oligosaccharides, accumulate preferentially on certain faces of the crystal and inhibit the deposition of sucrose. Crystal growth then becomes limited to the other faces and elongated crystals are formed. Often the ratio of length to diameter is in the range of 3-5 to 1. In extreme cases "needle grain" occurs in which the ratio is even higher. The rate of growth of elongated crystals is much lower than that of normal crystals. No practicable way of removing the interfering substances has been found. When such crystals appear, the purity above expected of the molasses increases markedly. A good indication to the operating staff of problems ahead is the pH of first expressed juice. When this drops below 5.0 deteriorated cane is indicated and elongated crystals can be anticipated. The viscosity of final molasses governs the density to which the massecuite can be boiled so that it can be cooled in the crystallizers and handled mechanically. The higher the viscosity, the more the amount of water that must be left in the molasses, so the purity will be higher. Again the viscosity is related to the composition of the incoming juice so the factory has no control. In general, however, high molasses viscosity is found when juice from deteriorated, over-ripened, drought-stricken or immature cane is processed. Advance knowledge of this incoming material will be helpful to the operating staff in controlling the processing. In conclusion, the expected purity values are useful operational guides. In practice however, actual results must be appraised in light of other factors also. Other systems of estimation of expected purity are in use in other regions. The Sugar Research Institute in Australia has developed a formula based upon logic similar to that used in Hawaii. The formula is: Expected true purity = 40.67 - 17.80 log
r e d u c
^jj[
s
u
?
a
r s
In addition to estimations based upon a fundamental approach, numerous empirical formulae resulting from a statistical analysis of actual factory results in given locales have been developed and used.
120
Chief among these are: Douwes-Dekker (Java) Expected purity = 35.886 - 0.08088 R + 0.26047 A Where R = reducing sugars % nonsugars A = sulfated ash % nonsugars Hugot (Reunion) Expected purity = 40 - 4 χ
r e d l
1
1
^ ^ g sugars
These are useful, practical guides under applicable conditions. REFERENCES 1
Moritsugu, T., Somera, B. J., Sloane, G. E., Proc. Intern. Soc. Sugar Cane Tech., 1974, pp. 1236-1245.
121
Chapter 15
RECOVERY FACTORS The quantity of sucrose that it is possible to recover in sugar, relative to the quantity present in the juice, is governed by the purity of the juice and the purity of the final molasses. The relationship is shown by the SJM formula: Theoretical Recovery % = j | g _ ^ j Where
S = Sugar purity J = Juice purity M = Final molasses purity
The formula is valid as long as the same units are used in expressing all the purities. Standard is refractometer-pol purity. The actual recovery will be less than the theoretical by the known loss in filter cake and by unmeasured losses reported as Undetermined. Included in the latter are mechanical losses and chemical decomposition of sucrose in the processing. Normal undetermined losses are approximately 1.7%, made up of the following: Clarification Sugar Crystallization Mechanical
0.2 1.0 0.5
Factory balance figures for a weekly or monthly period may vary ± 1.0% from the actual value because of the unreliability of estimations of material in stock. No concern is needed, therefore, in the range of 0.5 to 2.5%. Figures greater than this should trigger checks of calculations, measurements and sources of real losses. The effect of juice purity is much greater than the effect of molasses purity on recovery. This is shown in Table 15-1. A one point change in juice purity from 85.0 changes the recovery 1% at 40 molasses purity, whereas a change of one point in molasses purity changes recovery only 0.45%. A t lower juice purities, the figures are 1.1% and 0.66%, respectively, showing relatively higher importance of molasses purity. The reason for this is the rapid increase in the quantity of molasses with lower juice purity. This is shown in Table 15-2, where molasses is calculated as a percent of the pol in syrup. Molasses quantity increases from 19.8% at 90 juice purity to 47.6% at 80 juice purity. Thus, purity is molasses recovery
it is important to keep in mind that obtaining low molasses of far greater value with low juice purity than high. Reducing purity from 40 to 25 at 80 juice purity gives an additional of 8.5%, compared with only 3.5% at 90 juice purity.
122
TABLE 15-1 96 DA SUGAR RECOVERY % POL IN SYRUP* y Molasses \ Purity
40.0
35.0
30.0
25.0
93.0
101.5
102.3
103.1
103.7
92.0
100.6
101.7
102.5
103.3
91.0
99.8
101.0
102.0
102.9
90.0
98.9
100.3
101.4
102.4
89.0
98.0
99.6
100.9
102.0
88.0
97.1
98.8
100.3
101.6
87.0
96.2
98.1
99.7
101.1
86.0
95.2
97.3
99.1
100.6
85.0
94.3
96.5
98.5
100.1
84.0
93.3
95.7
97.8
99.6
83.0
92.2
94.9
97.2
99.1
82.0
91.2
94.0
96.5
98.6
81.0
90.1
93.2
95.8
98.1
80.0
89.0
92.3
95.1
97.5
79.0
87.9
91.4
94.4
97.0
78.0
86.7
90.5
93.6
96.4
Syrup Purity
Refractometer pol purities. refractometer solids molasses.
Assume
98.8
purity
sugar
and
84.0
123
TABLE 15-2 MOLASSES % POL IN SYRUP* Molasses Purity
40.0
35Λ)
3CL0
25.0
93.0
12.6
11.6
10.8
10.1
92.0
15.0
13.8
12.8
11.9
91.0
17.3
16.0
14.8
13.8
90.0
19.8
18.2
16.9
15.8
89.0
22.3
20.5
19.0
17.8
88.0
24.8
22.9
21.2
19.8
87.0
27.5
25.3
23.5
21.9
86.0
30.1
27.8
25.7
24.0
85.0
32.9
30.3
28.1
26.2
84.0
35.7
32.9
30.5
28.4
83.0
38.5
35.5
32.9
30.7
82.0
41.5
38.2
35.4
33.0
81.0
44.5
41.0
38.0
35.4
80.0
47.6
43.8
40.7
37.9
79.0
50.7
46.8
43.4
40.4
78.0
54.0
49.8
46.1
43.0
Syrup Purity
\
Refractometer pol purities. refractometer solids molasses.
Assume
98.8
purity
sugar
and
84.0
124
Table 15-3 shows the relationship between the quantities of sugar and molasses at various purities. TABLE 15-3 MOLASSES % 96 DA SUGAR* \ \
Molasses Purity
40.0
30
30
34.0
93.0
12.4
12.0
11.6
11.2
92.0
14.9
14.3
13.8
13.3
91.0
17.4
16.7
16.1
15.6
90.0
20.0
19.2
18.5
17.9
89.0
22.7
21.8
21.0
20.3
88.0
25.6
24.6
23.6
22.7
87.0
28.5
27.4
26.3
25.3
86.0
31.6
30.3
29.1
28.0
85.0
34.9
33.4
32.0
30.8
84.0
38.2
36.6
35.1
33.7
83.0
41.8
39.9
38.2
36.7
82.0
45.5
43.4
41.5
39.8
81.0
49.4
47.1
45.0
43.1
80.0
53.4
50.9
48.6
46.5
79.0
57.7
54.9
52.4
50.0
78.0
62.2
59.1
56.3
53.7
Syrup Purity
*
Refractometer pol purities. refractometer solids molasses.
Assume
98.8
purity
sugar
and
84.0
125
Since purities are ratios, the SJM formula is a ratio of ratios. When a ratio changes either the numerator or the denominator or both may have caused the change. For this reason, in order to obtain a true picture in abnormal recovery situations, it is sound practice to run balances based on weights of pol and weights of refractometer solids, comparing the quantity in mixed juice with the total recovered in sugar and molasses. Assuming that the accepted standard procedures in sugar crystallization are followed, the recovery actually obtained from a given equipment installation is affected also by the quality of the juice processed as well as its purity, as indicated in the chapter on Final Molasses. Deleterious substances exert an influence not only on the exhausted purity of the molasses, but also on the rate of crystallization and the workability of the massecuites. Thus, the recovery obtained from day to day may vary considerably as the quality of the juice varies in other ways than purity. The following are the principal adverse factors on recovery. Low reducing substance-ash ratio. Juice from deteriorated cane. High content of alcohol precipitable matter. The rate of sucrose crystallization decreases with the reducing substance-ash ratio. This is caused usually by an increase in the salt content - notably potassium chloride. In general, cane that is grown in low-lying coastal areas gives juices of low ratios and ones that are difficult to process. It is for this reason that there is a good correlation between electrical conductivity of a juice and its processing characteristics. For example, a juice with a specific conductivity of 1000 micromhos processes much more easily than one of 3000 micromhos. Juices with reducing substance-ash ratios above 1.50 are generally easy to process and those below 1.0 difficult. When processing these low ratio materials, there is no alternative to taking a longer time to boil the pans and to allow more time in the crystallizers. Juice from deteriorated (sour) cane will invariably result in elongated crystals. Reported to be caused by some oligosaccharides produced by the growth of souring microorganisms, the substances concentrate on specific faces of the sucrose crystal inhibiting sucrose deposit, so only the unaffected faces grow. A slow-growing crystal face tends to increase in size. A fast-growing crystal face will decrease in size and may disappear giving rise to such sucrose crystal forms as needles and triangles. Since crystal growth in elongated crystals is limited on some faces, the rate of crystallization is diminished, so a longer time is necessary to obtain a desirable crystal yield. Also, the viscosity of the molasses from sour cane is usually higher. The combination of these factors results in difficult boiling massecuites, poor performance of crystallizers and centrifugals, affecting both the processing rate and recovery. There is no satisfactory solution to the problem, so normal practice is to process the poor material at the standard rate, take a lower recovery, and get the material out of the processing stream.
126
High content of alcohol-precipitable matter is normally associated with sour cane but is also characteristic of drought-stricken cane. Syrup and molasses from such cane has high viscosity so recovery is less because the massecuites cannot be handled at high refractometer solids. Again, there is limited action that can be taken, so recovery is lower. It is better to operate pans and crystallizers at higher temperatures which lower viscosity.
127
Chapter 16
SUGAR QUALITY Commercial sugar consists of sucrose crystals with an adhering film of molasses, so its quality depends on the composition and quantity of each component. Established practice is to set standard specifications on the quality of the crystal and on the sugar as a whole. Since the individual crystal is relatively free of impurities, even though comprising most of the sugar, the quality of the whole sugar is governed primarily by the quantity of molasses. The higher the pol of the sugar, the less molasses present, so the closer the analysis of the whole sugar approaches that of the crystal. Efforts to improve the quality of sugar take two courses - minimize the amount of impurities within the crystal and reduce the quantity of molasses outside the crystal. Material included within the crystal may be kept to a minimum by use of boiling techniques which are directed toward the growth of perfect crystals. Under the same conditions of growth, the quantity and quality of the nonsucrose within the crystal will depend upon the composition of the juice from which the mother liquor is derived. The quantity of molasses retained on the crystal is governed by the operation of the centrifugals and the physical character of the crystals. The important raw sugar quality factors are: Pol Crystal size and uniformity Color - crystal and whole sugar Filterability Moisture Ash POL Since the crystal is close to 100% pure sucrose, the pol of the sugar is determined by the amount of molasses surrounding the crystal. For example, a sugar of 99.0 pol and 0.2% moisture would be composed of approximately 1.3% molasses solids and 98.7% crystal solids, assuming the molasses layer is exhausted. If the molasses layer is reduced further by washing to a sugar of 99.5 pol, the quantity of molasses is approximately halved. The limit that can be achieved in good raw sugar operations is about 99.5 pol. With uniform crystal of standard size (0.8 to 1.0 mm) and free of conglomerates, the pol is controlled at the centrifugals by timing of the spinning, wash and drying cycles and quantity of water used. With irregular or conglomerate crystal, it is difficult to wash molasses from the
128
interstices without the use of excessive amounts of water which dissolve sugar, lowering the yield and decreasing the crystal size. CRYSTAL SIZE AND UNIFORMITY Crystal size is controlled in the vacuum pans. The final size is determined by the quantity of seed used relative to the final volume of massecuite, making certain that additional nuclei do not develop during the boiling. As discussed in an earlier chapter, this requires careful control of the pan conditions and good circulation. Even under the best controlled conditions, however, there is always the possibility of stray nuclei forming and random crystals coming in. Also, even under the same conditions, crystals of the same general size do not grow at the same rate. This is because growth occurs most rapidly at dislocations (irregularities) on the crystal surface, so the more dislocations, the faster the crystal grows. The probability of dislocations increases with the rate of growth also, so a self-perpetuating growth rate system develops. For this reason, crystals grown under conditions of high crystallization rate tend to be more uniform in size distribution, although they will be less perfect in form. Since crystallization rate increases with temperature, more uniform crystal distribution can be expected from massecuites boiled at high temperatures. Controlled washing in the centrifugals is necessary to avoid dissolving sugar. COLOR Crystal The color of the crystal from which the superficial layer of molasses has been washed away is caused by absorption of colored components from the mother liquor or entrapment of mother liquor within the crystal. If conditions of crystallization are the same, then the color is determined by the composition of the liquor. This means that the quality of the cane is governing. Minimum color is present when clean, sound, mature cane stalks are the raw material. Color increases with greenness of the cane, deterioration of the cane and quantity of tops and leaf trash entering the extraction plant. Some color develops in the processing of the juice to sugar. In general, processing conditions are relatively constant with respect to temperature, pH and time, so this factor is not much of a variable. Little or no color development occurs in evaporation. In sugar boiling, it is a function of the product of time multiplied by temperature. Colored compounds are produced from the reaction between amino acids and reducing sugars, decomposition of reducing sugars and many condensation type reactions. But, as indicated, these take place under approximately the same conditions. The operator has really only one way to keep color in the crystal to a minimum, and that is to prevent inclusions in the crystal. The more
129
nearly perfect the crystals, the less the color. This means operating the pans under controlled conditions so that the crystals grow uniformly requiring, as a prime consideration, good circulation. In general, the slower the rate of crystallization, the more perfect the crystals. Crystals of sucrose that sparkle like diamonds can be produced under extremely slow growth rates. Such are impractical on a commerical scale, of course, so uniform fast rates are to be employed. Formation of inclusions must be strictly avoided. These develop especially when dissolution of the crystals is allowed to occur because of conditions of undersaturation. As a crystal dissolves, which it does at about five times the rate of crystallization, the corners and edges become rounded first. Channels then develop on the surface. When such crystals start to grow again, the edges of the channels, having more contact with the supersaturated mother liquor, grow more rapidly and the channels become overgrown, leaving mother liquor entrapped. Such crystals also will likely have irregular surfaces and the entrapment phenomenon tends to continue. This reality points up the importance of having near perfect crystals for seed. If imperfect crystals are used, or crystals that have been partly dissolved, inclusion will start at the beginning and probably continue throughout the growth of the crystal. Also not to be overlooked is the common, but unacceptable, practice of "washing out" false grain. This cannot be done without lowering the quality of the remaining, partially dissolved crystals. If some dissolution has taken place, the best procedure is to resume crystallization at a very slow rate so that damage repairs can take place before rapid growth is started. The alternative is to completely dissolve all crystals and start over. Since it is more difficult to maintain uniform crystallization conditions over the surface of a large crystal than a small one, inclusions increase with crystal size. For this reason, when grown under the same conditon, the smaller the crystal, the less color. Optimum commerical crystals are of uniform minimum size averaging 0.8 mm. Whole Sugar The color of the whole sugar is the sum of the color of the crystals and the molasses - the major one being the quantity of the molasses. As shown earlier for a 99.0 pol sugar, the composition is about 98.7% crystal solids and 1.3% molasses solids. The average value of whole raw color is about three times the crystal color, so two thirds of the color comes from 1.3% of the material. Important factors in keeping the whole color minimal are uniformity of crystal size, freedom from false grain and absence of conglomerates. Cycle timing and washing efficiency in the centrifugals must be carefully controlled. Two-stage washing is standard practice.
130 FILTERABILITY Sugar filtration rate measurements are made on the washed crystals. Therefore, quality considerations are the same as those for crystal color, namely, that it is primarily the quantity and the type of material included within the crystal that affects the filtration rate. It has been established that the controlling element is the quantity of insoluble particles at the size level of 1 micron. It has also been shown that the composition of the particles is not important - only the size. Furthermore, almost any type of material can be grown inside a sugar crystal. Likewise, the size can be anything from micro to macro; for example, from colloids to such things as grains of sand, string, or even rubber bands. Although there is some absorption, which may be preferential, the bulk of the filtration impeding insolubles enters the crystal by inclusion. Therefore, the final condition is determined by the quantity of 1 micron level insoluble material present in the mother liquor and the extent of crystal inclusions. Clarification which reduces the amount of micron material to a minimum, of course, should be given first consideration. However, as already noted, usually the effectiveness of clarification is more dependent upon the quality of the juice rather than the clarification procedure. So the most important factor that something can be done about is the sugar boiling technique. Good circulation, prevention of inclusions and uniformity of crystal growth will lead to maximum filtration rates. Again, other conditions being equal, the filterability is governed by the quality of the incoming juice. MOISTURE There is water both on the surface of the sugar crystal and within the crystal. That on the surface is related to the amount of the molasses layer and the relative humidity of the atmosphere surrounding the sugar. Moisture within the crystal, again, is governed by the extent of inclusions of mother liquor. As moisture is of primary importance in maintaining the keeping quality of the sugar relative to the growth of microorganisms, it is the superficial moisture that is important. This can be controlled in the operation of the centrifugals with proper maintenance of the washing and drying cycles. The Deterioration Factor or Safety Factor is used as an indicator of the keeping quality of sugar with respect to moisture: Deterioration Factor =
f^
1Stur
?
100 - pol This is generally applicable to sugars in the pol range of 97.5 to 98.5, where a factor of 0.25 is considered safe. For high pol sugars, 99.0 and above, 0.20 is a safer limit, although moisture is less of a factor for such sugars.
131
With all sugars, safe moisture levels can be obtained in the centrifugal without the use of sugar dryers as is common practice in some sugar regions. Conditioning of the sugar, that is allowing it to come to equilibrium with respect to internal moisture, external moisture and the atmosphere, is an important step in safe storage and prevention of caking. Conditioning can well be accomplished in a "dryer" operated essentially as a cooler. ASH Ash is no longer a factor of much importance in a raw sugar because of higher pols. It was traditionally run on the whole sugar and was really a measure of the quantity of molasses present. Ash level of the sugar crystal is a function of the amount of inclusions, as in the case of color and filterability.
133
Chapter 17
SUGAR AND MOLASSES HANDLING SUGAR The important factors in the storage of bulk raw sugar are the pol-moisture relationship, the temperature of the sugar, and the relative humidity of the air in the storage enclosure. The initial pol-moisture relationship determines the susceptibility to action by microorganisims. Temperature governs the rate of chemical decomposition (particularly of the molasses layer) and caking tendency. The relative humidity affects the maintenance of the initial moisture and the caking or liquifying trend. In the general pol range of 97.8 to 99.3, growth of microorganisms is inhibited if the factor, Moisture 100 - pol is not much above 0.25. This ratio is called the Deterioration Factor or Safety Factor. Principally involved is the moisture level of the molasses layer surrounding the crystals. If the sugar were not washed in the centrifugal, then the keeping quality of the sugar would not be impaired for microorganisms will not grow in a saturated molasses because of the high osmotic pressure. Washing dilutes the molasses and permits growth. Growth of microorganisms apparently causes further dilution and loss of sugar becomes continuous, although slow. First involved in the microorganism activity are reducing sugars, and in particular the fructose. Thus, in the initial stage, the pol of a relatively low pol sugar will actually increase, because of fructose loss, before the pol starts to decrease as sucrose inverts. Care in the washing and drying steps in the centrifugal will give a sugar with a Deterioration Factor in the safe range without much difficulty. The molasses layer has a definite vapor pressure (water vapor) depending upon the temperature and to some extent on composition. There is for every sugar a humidity at which it neither gives up nor absorbs moisture. This is called the Equilibrium Relative Humidity. For raw sugars in the safe range this is about 65% at ambient temperatures. If the atmosphere in which the sugar is stored contains less moisture, then water will evaporate, the molasses concentrate and the sugar become drier. If it contains more moisture, water will be absorbed and the sugar becomes wetter. Sugars stored at a relative humidity much below this will tend to cake because of sucrose crystallization caused by molasses concentration and cementing together of crystals. Sugars stored at a relative humidity much
134
above can reach the dilution level where microorganism activity increases. A t very high humidity the sugar will completely liquify. The point at which this occurs is somewhere above 85% relative humidity. Storage of raw sugar, however, does not require a controlled humidity of 65%. In a relatively closed warehouse only the sugar in the surface area of a large bulk pile is usually subject to the atmospheric effects found in most subtropical and tropical regions. There is a general diurnal pattern in which, during the middle of the day, the humidity above the storage pile drops below 65% and the surface crusts over. Late at night, when the humidity may go above 95%, the surface liquifies. This drying and liquifying action continues only at the surface, seldom extending below 25mm. In cases where rain continues for several days keeping the humidity high in the daytime the surface may remain wet. It is important in such circumstances to keep the storage area completely closed in order to prevent entrance of much outside air. Bulk storage is much better than bag storage as the action described is not possible with bags. In bags a large surface area is exposed and therefore a substantial quantity of sugar is involved in the humidityproduced action. Furthermore, if the bag is jute or cotton, the material acts as a wick transporting the liquified sugar to the outside of the bag, removing it from the sugar surface and ultimately permitting it to flow away as molasses. Sugar loss can become high in areas of constantly high humidity unless completely closed storage is used. Temperature is the second important factor in storage. The molasses layer on the crystals is subject to the same danger of exothermic decomposition as final molasses if the temperature of storage is elevated. The critical temperature for sugar is about the same as molasses - 45° C at which the chemical reaction responsible becomes self-sustaining. When insulated by a large volume of material, as in the center of a pile of sugar, the heat builds up to a point where complete decomposition can occur. This has happened with sugar as well as molasses. Sugar is more susceptible the lower the pol - that is, the more molasses it contains. High pol sugar (99.3 or above) is in little danger. Even if the temperature of the sugar does not reach the critical point, serious darkening of the sugar can occur. Although the action is mostly in the molasses layer, it may extend to the crystal also. Sugars subject to this thermal action have a reddish color. Raw sugar therefore should not enter a bulk storage facility above 40° C. Caking of stored sugar, particularly in bags, can also be caused by storage of hot sugar. On cooling, sugar in the molasses layer crystallizes causing a cementing together of the crystals. Sugars which have been conditioned by cooling while being agitated, either by air or mechanical means, are less subject to caking. This is a standard practice with refined sugar. MOLASSES The standard for final molasses for shipping is 85 brix at 1-to-l dilution. This corresponds to approximately 82 refractometer solids.
135
Final molasses at this concentration is not subject to microorganism activity except superficially, because of its high density, but is thermal sensitive. As noted in the discussion under sugar, complete decomposition can occur if the molasses temperature reaches a threshold value. This is approximately 45° C. Evidence indicates that the triggering chemical reaction is that between reducing sugars and amino acids. This reaction is exothermic and carbon dioxide is released. A t low temperatures, the reaction proceeds slowly with only a small evolution of carbon dioxide. As the temperature increases, evolution of gas is sufficient to cause the molasses to increase in volume and frothing occurs. In the absence of circulation within a large body of molasses, the heat is not dissipated and the reaction can become rapid and violent with complete decomposition, leaving only a char-like residue. Preventive measures are based on temperature control. Molasses, like sugar, should be cooled below 40° C before entering storage. The storage tank should be fitted with circulation or agitation equipment to give sufficient mixing to prevent hot spots from developing. A satisfactory procedure is the use of compressed air piped to the bottom of each of the quadrants of the tank. When frothing begins, air will provide both cooling and circulation. Storage tanks should be fitted with temperatureaccuated systems which automatically start the air pumps at a set temperature. Molasses at 25° C changes very little in composition over a period of years. In storage there may be slight microbial activity on the surface made possible by localized dilution from moisture condensation. Such activity is minor unless the humidity is high. The surface of stored molasses must be protected from casual water.
137
Chapter 18
STEAM GENERATION A cane sugar factory is captive energywise, obtaining its power and heat requirements by burning its own fuel - bagasse. A factory designed for energy efficiency and properly operated will produce surplus bagasse from which electricity can be generated and exported. BOILER DESIGN The design of a bagasse boiler used primarily to supply factory power, process steam needs and plantation power load (such as irrigation) tends to center on one capable of burning bagasse at 48% moisture and producing 2 2 steam at a nominal 32 k g / c m (450 lb/in. ) (3103 kPa) pressure. If export 2 of more power is favorable, then doubling the pressure to 64 k g / c m (900 2 lb/in. ) (6206 kPa) will be justified. The cost of boilers above the 32 kg/cm* range increases substantially, however. Also operation control becomes more critical. In particular, boiler water quality becomes of great importance. With finely prepared bagasse, spreader stoker feed permits burning most of the fuel in suspension. Travelling grates allow burning of the remainder on the grate and give effective removal of ash. Suspension burning also gives a faster response to load changes. High boiler efficiency requires keeping the losses to a minimum. These tend to increase as the boiler pressure increases. The use of air heaters and economizers constitutes the principal means of reducing sensible heat loss. These are heat exchangers in the flow of the exiting gases from the boiler, which transfer heat to the entering air in the case of air heaters and to boiler feed water in the case of economizers. The industry still awaits the development of a system for recovering the latent heat of the water vapor in the flue gases. Effective insulation is essential to reduce losses from the boiler itself. Design specifications must include stack emission control equipment. The most efficient of these are wet scrubbers. Because of corrosion, these must be constructed of corrosion-resistant material. Gas velocity must be kept as low as practicable in order to increase the efficiency of the scrubbers. CONTROL Boiler Efficiency The overall efficiency of a boiler is expressed as the percentage ratio between the heat transferred to the steam and the heat available in the fuel. The formula is:
138
Fff * <* - Heat transferred to steam χ 100 fcinciency <*> - Gross Calorific Value of bagasse A value of 68% is considered to be a reasonable target for an efficient bagasse boiler. Actual measurements of efficiency are rarely conducted because of the difficulty of measuring the weight of bagasse entering the boiler in a given time period. Useful approximations of the efficiency are possible however. A boiler heat balance is composed of the sum of the following items: Transferred to steam (Efficiency) Condensation loss Sensible heat loss Unburnt gas loss Blowdown loss Undetermined Condensation loss is the latent heat of water contained in the bagasse. Sensible heat temperature.
loss
is
sensible
heat
of
flue
gas
above
ambient
Unburnt gas loss is due to incomplete combustion to carbon monoxide instead of carbon dioxide. Blowdown loss is due to the necessary continuous drain of boiler water to maintain dissolved solids in the boiler to safe operating level. Undetermined includes loss to surroundings, unburnt solids and furnace ash. Analysis of the flue gas, temperature measurements and visual observation, together with an estimate of undetermined losses give a good estimate of boiler performance. Gross Calorific Value of bagasse is the total heat of combustion, which for dry, ash-free bagasse fiber is taken as 4643 cal/kg (8350 Btu/lb) (19422 kJ/kg). For bagasse as fired, correction must be made for moisture, ash and soluble combustibles. Assume a typical bagasse of 48% moisture, 2% pol and 2% ash, then Fiber % = 100 - [48 + 2(+) + 2] The soluble combustibles would be somewhat higher than the pol. However, the calorific value of sugar, 3953 cal/kg (7110 Btu/lb) (16538 kJ/kg), is less than that of fiber, so it is close enough to consider pol as having the same fuel value as fiber. Therefore, Fiber % = 100 - 52 = 48% Gross Calorific Value
= 0.48 χ 4643 = 2229 cal/kg 0.48 χ 8350 = 4008 Btu/lb 0.48 χ 19422= 9323 kJ/kg
Net Calorific Value is less than the gross by the amount of heat necessary to vaporize the water in the original bagasse and that from
139
combustion (condensation loss). A standard deduction of 573 cal/kg (1030 Btu/lb) (2396 kJ/kg) of water formed is often used. Condensation loss as a percent of Gross Calorific 1 estimated by the formula: ~ , 4.· ι 100 (562 - 4.82 χ moisture % bagasse) Condensation loss = Gross Calorific Value (Btu)
Value can be
In the above example this is: 100 (562 - 4.82 χ 48) _ 0 O 8 2 4008 (Btu) ·
Q L %
Sensible heat loss is determined by the temperature and composition of the gases leaving the boiler. This loss as a percent of the Gross Calorific 1 Value can be estimated by the Siegert formula:
K
Sensible heat loss % = Ç Q
+\j
The constant Κ depends upon the C O 2 content of the flue gas and the moisture content of the bagasse, tp and t^ are flue gas and ambient air temperatures. C O 2 is volumetric % of carbon dioxide and U is volumetric % of unburned gas converted to carbon dioxide. This formula can be simplified to: Sensible heat loss = Κ (tp - t ^ ) without unburned gas The following temperatures.
table
shows
Κ
values
relative
to
VALUES OF 1,000 Κ FOR SENSIBLE HEAT LOSS
Fahrenheit
1
Moisture in Bagasse % in gas %
CO2
6 8 10 11 12 13 14 15 16 17
\
44
46
48
50
52
68.1 53.4 44.5 41.3 38.6 36.4 34.4 32.7 31.3 30.0
68.4 53.7 44.9 41.7 39.1 36.9 34.9 33.2 31.7 30.5
68.8 54.1 45.3 42.1 39.4 37.2 35.2 33.6 32.1 30.8
69.4 54.6 45.8 42.6 39.9 37.6 35.7 34.1 32.6 31.3
69.9 55.1 46.3 43.1 40.4 38.2 36.2 34.5 33.1 31.7
140
In the example with 48% bagasse moisture, 14% C O 2 and temperature difference of 300°F, the sensible heat loss would be 10.6%.
a
Unburnt gas loss can be measured by analysis of the carbon monoxide 1 content. As a percent of Gross Calorific Value the formula is: Unburnt gas loss % = 42 where U is volume % of unburnt gas converted to C O 2 . Undetermined loss consisting of radiation loss, loss of sensible heat in ash and loss in unburned bagasse particles can vary widely but is usually in the range of 5 to 10%. An average value is estimated for making efficiency calculations. Summarizing the estimations based upon the example previously cited, the total would be: % of gross calorific value Transferred to steam (Efficiency) Condensation loss Sensible heat loss Unburnt gas loss Blowdown loss Undetermined Total
68.0 8.2 10.6 4.5 3.0 5.7 100.0
BOILER WATER Feedwater for steam generation should not cause scaling or corrosion in the boiler and should give steam free from contaminants. The best source of water meeting these requirements is from condensation of the steam itself. Thus, the condensate of exhaust steam is the principal source of supply water. As the quantity is not sufficient because of losses, some makeup is necessary. This is often taken from the following effects of the evaporator, juice heaters and pans when these are heated by vapor from the evaporator. Treated raw water is used only when the supply of condensates is insufficient. Exhaust steam condensate should be essentially distilled water subject to contamination only by carryover from the boiler and leakage in the heat transfer vessels where the exhaust steam is used. The exhaust from reciprocating steam engines will also contain oil. Leakage of sugarbearing material should not occur in evaporator cells and pans heated by exhaust steam since the calandria pressure is higher than the body pressure. Of course when the vessels are shut down there is a possibility of contamination. In juice heaters, however, the juice is at higher pressure and a leak can cause serious leakage into the condensate. Condensates from vapors are called vegetal condensates and contain volatile organic compounds from the cane juice, chief of which is ethanol.
141
Also present are organic acids, higher alcohols, esters and oils. Although such materials are not a problem in low-pressure boilers, they are of 2 2 concern with pressures above 32 k g / c m (450 lb/in. ) (3103 kPa) because they can cause priming and carryover. Treated fresh water should be used for make up in these boilers. Raw water must be treated chemically depending upon its quality and the boiler pressure. This includes removal of insoluble solids, precipitation of scale-forming constitutents, increasing pH, and in some cases removal of soluble constituents by ion exchange. Boiler Water Treatment Standard boiler water treatment requires: 1.
Deaeration to remove oxygen by flashing
2.
Heating
3.
Chemical treatment
4.
a.
to inhibit scaling
b.
to prevent corrosion
c.
to reduce foaming or carryover
Blowdown to reduce accumulation of solids
Deaeration Much of the dissolved oxygen in feedwater can be removed by deaeration. This is accomplished by heating and allowing the feed to flash to lower pressure. Heating All condensates should be maintained without cooling. With an economizer the feedwater is heated by flue gas before entry to the boiler. Chemical treatment To inhibit scaling, the principal elements to immobilize are calcium, magnesium and silica. Normal treatment consists of making alkaline with caustic soda and adding phosphate. This causes precipitation of calcium phosphate, magnesium hydroxide and some silicates. These insoluble materials are removed from the boiler as a sludge in the blowdown. Chelating agents, which sequester positive ions, are part of some treatments. They are effective in removing copper and iron. An increase in alkalinity prevents acidic corrosion. To prevent oxygen corrosion, oxygen scavengers are used. Those commonly used are sodium sulfite and hydrazine. Hydrazine is preferable in high pressure boilers as it does not increase the solids content. Priming and foaming, which cause carryover of the boiler water into the steam, are usually caused by the presence of surface-active substances in the water. Although antifoaming agents are often used to reduce priming, there is no effective treatment for high pressure boilers, so the water should be free of any organic material which might cause
142
priming. Priming is more serious when the load oscillates, so a steady steam flow is always desirable. Blowdown Blowdown prevents the build up of insoluble solids. Since blowdown is wasteful of energy it is important to keep solids in the water to a minimum. This also means keeping water treatment chemicals to a minimum. OPERATION A steady uniform operation is vital to maintain efficiency in a boiler. This means a constant supply of bagasse of uniform quality. If milling stops, stored bagasse must be introduced without interruption. Bagasse Bagasse quality is of prime importance and moisture content is the most important factor. Most boilers are designed to burn bagasse of 50% moisture. Problems in burning can be expected if the moisture goes above 52%. Much of the bagasse does not dry and burn in suspension but accumulates on the grate. In the case of a travelling grate, there may not be much difficulty but on a dumping grate, bagasse may accumulate in piles, combustible gases can be generated which ignite explosively from time to time causing puffing. Bagasse with green leaves dries slowly, even though the content is satisfactory, and will cause burning difficulties.
moisture
The steam generating capacity of the bagasse is a function of the moisture content. As a rule of thumb, a one percent change in moisture means about a one percent change in the fuel value obtained fom a given quantity of fiber entering the mill. Stored bagasse dries out and becomes a more efficient fuel. However it also loses its sugar content rapidly which means a loss in energy. With good mill extraction this amounts to the order of 3% of the total energy available. Thus, in short periods of storage (2 or 3 days) the two effects are somewhat balancing. Bagasse of high sugar content should be burned at once. It should be kept in mind in this connection that with poor mill extraction bagasse may show a low moisture content simply because the pol is high. Air Supply Air supply must be adjusted to the bagasse rate using the minimum amount of air necessary for combustion. In practice it is not possible to get complete combustion without using excessive amounts of air, so a reasonable balance is arrived at leaving some unburnt gases. In the absence of analysis of flue gas for unburnt gas content, control can be by C O 2 or oxygen content. In reality, routine control is normally based upon air pressure measurements in the boiler.
143
Boiler Water Standard boiler water control procedures are used. The basic requirements of which are to maintain alkalinity, a scale treatment additive - phosphate or chelate, a residual oxygen scavenger content and a low level of solids. Condensates are monitored continuously for contamination caused by equipment leaks. Electrical conductivity is commonly used for sensing as the sugar streams carry relatively high salt concentrations. Contaminated condensate should be discarded to the house hot water system. Oil Firing Oil firing is used to supplement bagasse as required. It is, of course, vital to keep its usage at a minimum. This requires careful management of the bagasse supply and steam demands of the factory. Since fuel oil has some five times the calorific content of bagasse it is difficult to burn the two fuels together with good efficiency. Stack Emission Government regulations require careful control of the particulate matter in the stack emission. Even with effective wet scrubbers control of particulate matter is difficult with variable loads, so it is important to maintain steady operation. Bagasse of high moisture content or containing green trash does not burn readily and produces more unburned materials. Much of the fine particulate matter also has its origin in soil present in the bagasse. BAGASSE DRYING The fuel value of bagasse becomes higher as the moisture content decreases mainly because of less heat necessary to vaporize the water. The most energy-efficient means of removing water is to do it mechanically in a press or a mill, but there is a practical lower limit. The traditional standard for a cane mill is to get a bagasse of 48% moisture. Mills have operated in the past with yearly averages below 40%, however. This would be considered uneconomical now because of higher milling rates. About the lowest considered practicable is 45% average moisture. There is sensible heat in flue gas which could be used to evaporate some of the moisture from bagasse. A bagasse dryer utilizing flue gas is essentially a heat trap like an air heater or economizer. If the stack gas has a high enough temperature to evaporate some water and still keep well above the dew point then a bagasse dryer heat trap may have economic value. In a boiler with a high stack temperature of 250° C a bagasse dryer working on 46% moisture bagasse which halved the temperature to 125°C would increase the calculated boiler efficiency by approximately 7.8%. From this must be subtracted the energy used in the drying system such as for driving fans and also radiation losses. In an efficient boiler with air heater and economizer the temperature of the stack gas is already close
144
to 125°C so the sensible heat available above the dew point may be too low for recovery by bagasse drying. It must be kept in mind that all of the water vapor originally in the bagasse remains in the stack gas after a bagasse dryer. FIBROUS TRASH FUEL There is a quantity of potential fuel that can be made available to the factory from fibrous cane trash - that is, tops and leaves. During its growth a 24 months cane plant produces close to twice as much fiber in leaves as is present in the millable stalk at harvest. Most of the leaves drop off and become ground trash before harvest. Some of these decompose but essentially all, as well as attached dry leaves, is burned in a dry weather cane fire. There is left only a small partially dried tops which averages in weight only about 3% of the total cane harvested, and a bit of unburned leaves amounting to about 1/2%. The tops average around 25% fiber and the leaves from 35% to 70% depending on their dryness. Using these figures there is available from completely burned cane per 100 tons stalk cane of 13% fiber: Tons fiber Stalks Tops Leaves (50% moisture)
13.00 0.78 0.26 14.04
This is 8% more than in the stalks alone. Under the same conditions if the cane is not burned the weight of tops would be about 4% but dry leaves would increase to about 11%. The figures per 100 tons stalk cane would then be: Tons fiber Stalks Tops Leaves (50% moisture)
13.00 1.18 6.47 20.65
This is an increase in fiber of 59% over that in the stalks alone. The relatively small amount of fiber from trash in well-burned cane can be processed in a mill or diffuser without much difficulty, provided there is a capacity for the increased load. Pol in bagasse can be held at approximately the same level as without the trash. Extraction, that is recovery, will be lowered by the amount of additional bagasse. A t the 95% extraction level this would correspond to a drop of about 1/2% in extraction. Processing of unburned cane, aside from harvesting and transportation, would necessitate roughly a 50% increase in extraction plant capacity to achieve the same pol in bagasse. Assuming that this capacity were available the extraction would decrease by about 3%. Under such
145
conditions handling of the trash separate from the cane would have to be considered. Two options are available for handling the fibrous trash from the cleaning plant. In one, the trash may be shredded and partially de watered in a mill then returned to the bagasse. The moisture content of the trash would be around 62%, so the moisture of the bagasse would have to be low in order to provide a boiler feed of bagasse and trash of 50% moisture. In another, the trash is trucked to an open storage area where it is allowed to air-dry. It can then be returned to the bagasse supply for feed to the boiler. REFERENCES 1
Laboratory Manual for Queensland Sugar Mills 5th Ed. Bureau of Experiment Stations, Brisbane 1970, pp 169-170.
147
Chapter 19
USE OF STEAM The uses of high pressure steam for power and exhaust steam for processing make a cane sugar factory uniquely efficient in energy utilization. However, there is always difficulty in establishing a balance between the two demands. This arises to a large degree, from the variable process demand of batch vacuum pans. Thus, if sufficient exhaust steam is generated to meet peak demand (when all pans are on stream), then when some pans are down exhaust steam may be lost to atmosphere, or vapor lost to atmosphere (if the pans are operated on vapor.) This gives not only a loss in energy but also a loss in condensate for boiler feed water. To reduce this, therefore, it is common practice to generate less than the full demand of exhaust steam and make up the remainder by passing high pressure steam through a pressure reducing valve. To determine just how much exhaust steam should be generated requires a careful study, not only of the steam requirements of the several stations, but also of the general experience of how the factory runs with respect to cane supply, juice quality, extraction plant time schedule and time efficiency. A general rule of thumb is that about 80% of the peak processing demand should be supplied as exhaust steam. This figure should be used only as a starting point and not as a valid figure for an individual factory. STEAM PRESSURES 2
As discussed earlier, the optimum boiler pressure is around 32 k g / c m , 2 (450 lb/in. ) (3103 kPa). This makes possible the meeting of the power demands of the factory, outside requirements for irrigation pumps and still leaves some available for export of electricity. Of course, with a profitable export electricity market, it is advisable to use the greater 2 power generating efficiency of higher pressure. In such cases 64 k g / c m 2 (900 lb/in. ) (6206 kPa) is reasonable. Exhaust pressures should be the minimum necessary to meet the processing uses because the higher the back pressure on the prime movers, the less their efficiency. The minimum is actually determined by the requirement for heating limed juice which must be flashed as it enters the clarifier. The minimum heating temperature is therefore 102° C. The first effect evaporator vapor used for heating therefore has to be about 108°C in order to give adequate heat transfer. This means a pressure of 2 2 0.35 k g / c m (5 lb/in. ) (34 kPa). In order to obtain this pressure, the exhaust steam should enter the first effect of the evaporator at a 2 2 minimum of 0.75 k g / c m and preferably up to 1.0 k g / c m . The back
148
pressure at the prime movers should therefore be in the range of 1.0 to 1.4 2 2 kg/cm (14-20 lb/in. ) (97-136 kPa). STEAM BALANCE Traditional steam balances for a factory usually involve a series of calculations based on assumptions and approximations. Rarely are actual heat and power requirements measured and individual steam flows recorded. Calculated balances are useful exercises and give the order of magnitude of the requirements, but are less useful in appraising a going operation in order to achieve better efficiency. Actual measurement of steam being used is the best lead to improved economy. The factory loads outside the boiler are as follows: High pressure steam Turbogenerators Prime movers Makeup Exhaust steam Evaporator Juice heaters Vacuum pans Miscellaneous (centrifugals, molasses heating, steaming out, crystallizers, hot water) The total quantity of high pressure steam depends upon the power requirements of the factory equipment. With most of the prime movers electrical, except for large units such as for the mill or diffuser stations, the turbogenerator will be the main consumer in providing electricity for factory loads as well as export. Power necessary for the extracting plant is variable depending largely on how much is used in the cane preparation machines. With a given installation there is not much that can be done in operation to improve efficiency. In the case of exhaust steam, however, efficient operation is most important. Here most of the steam goes to the evaporator, with pans and juice heaters operating on vapor. Flows to these stations should be monitored and recorded. Recording is particularly important in pointing out the load fluctuations. Once the amount of steam being used at each station is known, steps can be taken to make more effective use of the available steam. PROCESS STEAM REQUIREMENTS Evaporation In a quadruple 1 follows:
effect
evaporator typical conditions might be as
149
Pressure kg/cm^ lb/in.2 Exhaust 1st effect vapor 2nd effect vapor 3rd effect vapor 4th effect vapor
Vacuum mm Hg in. Hg
0.91
13
—
—
0.49
7
—
0
0
— —
Temperature
»_c
Latent heat cal/kg Btu/lb J/g
119
246
527
948
2205
—
111
232
532
957
2226
0
0
100
212
539
970
2256
—
381
15
82
179
550
990
2302
—
648
25.5
55
131
568
1022
2377
If no vapor is taken for juice heaters and vacuum pans, one unit of steam will evaporate approximately one unit of water in each individual effect. It will be noted from the table that there is almost enough latent heat from condensation to provide the latent heat of the vapor in each case. There is however some flash evaporation as the juice passes from a higher pressure to a lower in each effect. If this flash is utilized it will more than compensate for the latent heat necessary. In actual practice nevertheless, the steam required is somewhat more than the one-to-one relationship because of losses in operation of the multiple effect system. For approximation purposes the one-to-one ratio may be used. Assuming 100 units of clarified juice of 13% solids to be evaporated to 65% solids, the quantity of water vaporized will be: 100 -(100 x 0 . 1 3 ) ,
OS
Qn 8 0
Since each effect will evaporate the same quantity of water the first body will need: 80 - r = 20 units of exhaust steam. 4 Juice Heating The rule of thumb for juice heating is that one unit of steam will heat 100 units of juice 10°F. This decimal relationship in English units comes about from the fact that the specific heat of juice is 0.96 and one pound of steam has roughly 960 Btu latent heat. For 100 units of clarified juice there would be about 90 units of limed juice to heat to the flash point, say from 85° F to 218°F, or 13 3° F total 133 -j-g- χ 0.9 = 12.0 units steam would be necessary. In addition, the clarified juice will have to be heated from about 202°F to its boiling point in the first effect, 232° F, or 30° F total regardless of whether this is done in a preheater or in the evaporator. For 100 units of juice it will take: 30 • j ^ χ 1.0 = 3.0 units of steam.
150
Sugar Boiling Calculations of the amount of steam necessary for sugar boiling are of limited practical value because the calculated amount is greatly exceeded by addition of water during the boiling. Using average values of solids content for the various massecuites, molasses and limited inboiling, the calculated steam necessary to boil massecuites from 100 units of clarified juice is about 10 units. In practice the actual amount ranges from 11 to 16 units. So a reasonable assigned value might be 14 units. Total Process Steam Required With no vapor bleeding from the evaporator the total steam per 100 units clarified juice would be: Units steam Evaporation Juice heating Limed juice Clarified juice Sugar boiling
20 12 3 U
49 With vapor bleeding, if first vapor is used for heating limed juice and sugar boiling, then 12 + 14 = 26 units of vapor are required. Thus 80 - 26, or 54 units will have to be used for evaporation in the quadruple, or = 13.5 units 4 in each effect. The total units of steam required entering the evaporator, therefore would be 26 + 13.5 = 39.5. Add to this the 3 for clarified juice heating gives a total of 42.5, compared with 49 without bleeding. To these figures should be added about 10% to cover miscellaneous losses. POWER RELATIONS OF STEAM The power available from steam in a turbine is dependent upon the pressure and superheat of the steam and the back pressure of the exhaust. 1 These are illustrated in the following table from Eisner :
151
Heat Steam Steam Back availHeat utilized pressure temp, pressure able 50% Eff. 65% Eff. lb/in.a Btu/lb Btu/lb lb/in> F Btu/lb
Steam required 50% Eff. 65% Eff. lb/hp/hr lb/hp/hr
180 180 180
380(sat.) 500 500
15 15 10
140 160 172
70 80 86
91 104 112
36.5 31.5 29.5
28.0 24.5 22.7
250 250 250
407(sat.) 500 500
15 15 10
166 180 196
83 90 98
108 117 127
30.5 28.0 26.0
23.5 22.0 20.0
OPERATION Steady operation is the critical factor in maintaining steam usage efficiency. A reliable cane supply must be available so that the processing rate, once set, can be held with minimum fluctuation. A stop-and-go situation means excess and shortage of steam. Also for every factory there is a minimum cane rate at which the bagasse produced is sufficient to supply the fuel needs. This minimum should be established and the factory should not attempt to run below that rate except on an emergency basis. If the design steam balance of the factory is such that 80% of the process steam is supplied by exhaust steam, then at full load the makeup valve will be open and no steam is lost. Should the processing load drop below the 80% level because pans are shut down, then exhaust steam will have to be wasted by blowing to atmosphere. This can be prevented visually in large factories with several pans but is difficult in a small factory with a few pans. Steady pan operation is important therefore, a condition which is helped by having sufficient storage capacity for syrup and intermediate molasses. When the factory is operating in the range where makeup is being used, it is vital to keep the makeup to a minimum. This means good control of the quantity of water entering the process streams. First action in reducing water usage is usually to cut down on imbibition water to the mills or diffuser. Unless the imbibition is above the optimum, however, extraction will suffer. It is prudent to fix the imbibition rate at the optimum and not sacrifice in extraction. Often water use around a milling tandem can be lowered. Large quantities of water can get into the mixed juice from washing down the juice pans and allowing bearing cooling water to enter the juice. Syrup should be kept above 65 solids to minimize use of processing steam. Finally, pan operators almost invariably use excessive amounts of water in the vacuum pans. It is difficult to control, but it is possible to
152
boil a strike of sugar without the use of any water. Steamout of pans after a strike of sugar is often extravagent. In all pans, except those used for making seed, steaming is only necessary to wash the massecuite from the dropping valve. REFERENCES 1
Eisner, J., Basic Calculations for the Cane Sugar Factory. Booker Brothers, McConnell London, 1958, 24 pp.
153
Chapter 20
INTRUMENTATION Instruments can be at once the most useful and the most troublesome components of the many unit operations in a factory. Unreliable instruments are worse than none; nonworking instruments are useless. The basic rules for application of an instrument are: 1.
Put in only devices that the staff can maintain on a 24-hour basis and repair or replace immediately.
2.
Keep control instrumentation simple.
The main purpose of instrumentation is to improve efficiency. Reducing manpower may result, but if the factory is already on a minimum manning basis, added manpower will be necessary to maintain the instruments. INDICATING INSTRUMENTS Common instruments used to show conditions in the system are those measuring temperature, pressure, flow and level. More specialized instruments are used to indicate density, consistency, pH, refractive index and electrical conductivity. These may be either visual or recording. Recording instruments are essential for all critical measurements so that a complete record is obtained. In this way, malfunctions can be traced and appropriate corrective action taken. The most important consideration in the use of these instruments is the location of the sensing element. It must be at a point that is representative of the condition being measured. It often occurs that the stem of a thermometer, for example, is too short to reach into the body of the vessel or pipe. False readings result. Temperatures in a large vessel like a vacuum pan or evaporator cell vary substantially from spot to spot. Locating the thermometer, therefore, is all important. Temperature Most commonly used thermometers for the lower ranges are glass stem and bimetallic types. Thermocouples are used for higher temperatures. Glass stem thermometers depend upon the expansion of a liquid. They are simple, reliable and low cost. They have the disadvantage of breaking easily. Also, they are often difficult to read, but one can tell at a glance whether they are broken. Bimetallic thermometers depend upon the differential expansion of two metals bonded together. When formed into a helix, which can be fitted into a stem, a winding or unwinding motion takes place with a change in temperature. The movement can be used to move a pointer on a
154
dial. These instruments have the advantage over glass stem thermometers in their ease of readability, and are less subject to breakage. The accuracy is not as good and they are subject to change in calibration with usage. Since a small change in calibration may not be noticed, the reliability is not as good as with glass stem thermometers. Thermocouples function on the basis of a change in electromotive force caused by heat at the junction of two dissimilar metals. Thermocouples can be used up to very high temperatures and, so, are the chosen instrument in this area. They are reasonably accurate, easily read and rugged. They require frequent calibration. Pressure Pressure gauges are usually bourdon, diaphragm or bellows mechanical types, although manometers are used to some extent in vacuum systems. Since the mechanical types are subject to change with use, they must be checked routinely. The bourdon gauge is the most commonly used in the range of 2 to 2 7000 kg/cm^ (30 to 100000 lb/in. ). Its action depends upon the difference in pressure between the inside and outside radii of a circular, spiral or helical tube, which causes the tube to tend to straighten when pressure is applied. Precision measurements are not possible with this action at low pressure, so they are not useful in systems like exhaust steam lines, evaporators and vacuum pans. Diaphragm gauges depend upon the deflection of a flexible disc under pressure for their action. The most useful range is from a vacuum to 14 2 2 kg/cm (200 lb/in. ), covering that for which the bourdon is unsuited. Bellows gauges depend upon the expansion of a bellows under pressure 2 for their action. Their range is from vacuum up to 140 k g / c m (2000 2 l b / i n . ) . In general, they are not as reliable as the bourdon and diaphragm types are, so they are used less. The main disadvantages of the mechanism are changes in the bellows with use, because of work hardening, and the fact that the bellows are temperature sensitive. Flow Measurement of flow is the least satisfactory of all routine instrumental measurements. In fact, measurement of the flow of some process streams has thus far defied even reasonably accurate measurement. There is, nevertheless, a wide variety of measuring devices available. For general factory use, however, the choice is usually between pressure differential instruments and rotameters. Others that are sometimes used are weirs, magnetic flowmeters and metering pumps. Orifice meters are the most common type of flowmeters. The principle involved is measurement of the differential pressure across an orifice plate placed in a pipe of flowing material. They are suitable for measuring both liquids and gases over a wide range of flow rates, depending on pipe size. Accuracy is highly variable, but rarely is better
155
than 2%. Limitations are many. Of course, they cannot be used for liquids containing insoluble solids as such will collect at the orifice plate. Entrained gases can cause difficulty. Variations in density cause errors. For these reasons, orifice meters are useful in the factory only for measuring water. Rotameters consist of a tapered tube with a float whose position is governed by the flow rate of the liquid or gas. They are useful in a wide range of flow rates from 0.01 ml to 15000 liters per minute. The accuracy again varies widely but cannot be expected to be better than 2% and usually is not that good. The rotameter tends to be self-cleaning so will handle particulate matter of small size. Although small size instruments are sensitive to density and viscosity changes, the larger instruments are less sensitive. In general, the use of rotameters in the factory is limited to measurements of relatively small water flow rates. Weirs are useful for measure of large flows in open systems, such as inflow of supply water and disposal of waste water. They are really headtype flowmeters based on the principle that the flow rate is proportional to the head. They consist of a dam across the flow with an opening through which the liquid flows. The level of liquid in the opening gives a measure of the flow rate. Accuracy can be as good as 2%, but 5% is more common. Magnetic flowmeters are based on the principle that motion at right angles between a conductor and a magnetic field will develop a voltage in the conductor. Thus, the liquid under measurement must have electrical conductivity to act as the conductor. The devices have the advantage that they introduce no obstruction to flow and require no special piping arangement. They also are insensitive to viscosity and consistency changes. They have the disadvantage that the pipe carrying the liquid must always be full and any entrained gas is measured as liquid. It is mainly for this reason that the magnetic flowmeter has not proven useful in the factory. Metering pumps are positive displacement pumps that are customarily used to feed a liquid at a determined rate into a process. They can be used, however, to measure a flow rate with reasonable accuracy. There are several types of pumps available, the most important consideration being that no internal leakage occurs in the pump. Good maintenance, therefore, is important. Level There are a number of types of instruments available for level detection. Aside from float indicators and sight glasses, the most useful instruments are differential pressure-type detectors. In these, the hydraulic head is measured by sensing the pressure difference at the bottom of the vessel with that in the space above the liquid. Commonly used sensing elements are force balance and dry motion bellows types.
156 Proximity gauges are used to some extent. Level detection does not usually present a problem. For that reason, simple devices suffice. Specific Purpose Instruments These include pH, density, consistency, refractive index and electrical conductivity as the most common instruments. pH measurements are essential for juices and syrup. The glass electrode is universally used in conjunction with a calomel reference electrode to measure electromotive force. The glass electrode must be handled with care as it is fragile. Being of high resistance, it is also subject to insulation failure. Another factor in operation is that electrical fields will change the potential between the electrodes, so screening of the electrical field is necessary. Since the electromotive force measured is affected by temperature, the pH meter must be temperature-compensated. Generally, acid solutions have a positive temperature coefficient and alkaline solutions a negative coefficient. Most sugar solutions cause scaling of the electrodes. The only satisfctory solution to this problem is periodic cleaning. Careful washing with hydrochloric acid is effective. Density of syrup and molasses is usually observed by means of hydrometers or hydraulic head devices (similar to those for determining level). The main difficulty with molasses measurements is the effect of gas bubbles which can be a large factor. For this reason, refractive index measurement is a more reliable method. Consistency indicators are useful in sugar boiling operations. The most useful of these are rotating probes driven by constant speed motors from which the torque output is a function of cosistency. Special care is necessary to maintain a constant shaft friction. This is obtained by means of a water seal. Also, flow patterns in the massecuite affect the readings. Location of the probe in the pan should be such that the flow patterns are relatively constant. Electrical conductivity indicators are also used in sugar boiling control. These measure the specific conductance of the material between two electrodes. The measuring circuit is an alternating current Wheatstone bridge. The electrodes are most often made of stainless steel. Since conductivity varies with temperature, a temperature compensator may be used although as temperature rises, viscosity decreases. This compensates to some extent for the temperature increase as far as sugar boiling control is concerned. A major problem is scaling of the electrodes. They must be cleaned periodically to maintain reliable readings.
157
CONTROLLERS A system of instrumental control of an operation consists of a sensor to detect a condition and a transducer to convert the signal to a useful one and transmit it to a controller. The controller compares the signal with the desired (set point) and then transmits a controlling signal to a condition-varying mechanism such as a valve. The most important component, the sensing element, has already been discussed under Indicating Instruments. Only the basic considerations in the other elements of a control system will be discussed here. There are two instrumentation control systems - pneumatic and electronic. Pneumatic instruments are favored for a sugar factory because of ease of maintenance, relative ruggedness and simplicity. Although many electronic instruments are set up in modular form, a repair facility requires an expensive stock of spare modules. Even with modular parts, it is often difficult to locate malfunctions in electronic systems. Signals in a pneumatic system are transmitted by varying the air 2 2 pressure. The standard range is 0.2 to 1.0 k g / c m (3-15 lb/in. ) which would correspond to a 0 to 100 control range. In a typical system, a process transmitter, receiving a signal from a sensor, transmits a proportional air pressure to the receiving controller. The controller compares the signal with the set point and sends back an adjusting air pressure signal to the control element. The mechanism consists of two pairs of opposed bellows at the respective ends of a force beam which rests on a movable pivot. At one end of the force beam is a flapper which changes the back pressure on the detector air nozzle of a pressure regulator (booster). In the four bellows, the air pressures are those of the set point, sensor signal, reset and feedback functions. The force beam will be in equilibrium when the forces from all four bellows come to balance. A change in force on one of the bellows causes beam movement and a change in the position of the flapper. This causes a change in the feedback pressure, restoring the balance. Unit operations where controllers are essential are: 1. 2. 3. 4. 5. 6. 7. 8.
Mill and diffuser cane feed. Boiler pressure, excess air draft, water feed and safety interlocks. Clarification liming pH. Juice heater temperature. Evaporator vacuum, feed, level, syrup density and vapor loop pressure. Vacuum pan, vacuum and feed. Crystallizer temperatures. Centrifugal sequence cycle.
In choosing control instrumentation for these operations, it is wise to standardize as much as practicable in order to reduce the complexity of spare parts inventory. Recorders are advisable on all these basic controllers.
158
OPERATION The most important single factor in the use of instruments is that the operator has confidence in their value. This means that the instruments must be reliable. Not only must they indicate actual conditions but also, it is necessary that there be immediate indication of a malfunction. The job of the operator should be limited to using the instruments - not servicing them. Servicing should be in the hands of a qualified technician who is available immediately when instrument failure occurs. The technician requires a stock of parts so that repairs can be made at once. An inventory of spare instruments is necessary for particularly vital control devices. For pneumatic systems, since clean air is of prime importance, there should be an air compressor used only for this purpose. Daily monitoring of the air quality should be made. Electronic controllers must be housed in dust-proof Likewise, they should be isolated from heavy vibrations.
compartments.
159
Chapter 21
EQUIPMENT MAINTENANCE* EFFECTIVE MAINTENANCE High time efficiency, good sugar recovery and low production costs necessitate an effective maintenance program. Good maintenance is also reflected in good factory housekeeping. Cost effective maintenance is the result of planning and effort by all personnel. Maintenance is on-going and not a project with a start and completion date. Over a period of time individual factories have developed their own systems and procedures for maintenance. The elements common to effective maintenance programs are the following: Records: Complete data on equipment including purchase orders, specifications, drawings, operating manuals, maintenance instructions, spare parts lists and operational history. Inspection program: equipment. Work order system: maintenance.
Schedules
and
procedures
for
inspecting
Procedures for organizing and accomplishing
Spare parts: Controls to maintain critical parts.
inventory of fast
moving and
Investigation and analysis: Investigations of equipment stoppages and breakdown and analysis to prevent reoccurrence. Equipment and technology: Evaluation of new equipment technology, as possible replacement of high maintenance items.
and
Supervision and training: Supervision to insure proper maintenance and training to improve skills. A program of scheduled inspections and repairs is an important part of maintenance control. Judgment by experienced personnel is the key to the program. The balance between too much repair and not enough is elusive and practical judgment is necessary to execute the program to minimize stoppages while controlling production costs. A daily, period and annual off-season schedule should be set up and followed. All inspections and repairs should be recorded. A daily inspection by trained personnel walking through their area visually checking the equipment, touching motors and bearing housings, listening to pumps and gear boxes, and doing something about this information is vital to reducing lost time. Most breakdowns do not happen John W. Herkes assisted in the preparation of this chapter.
160
suddenly, but are caused by wear over a period of time and often this wear can be detected visually or by a change in temperature, sound, or vibrations well ahead of the actual breakdown. This prior warning, if acted upon, usually results in less damage and gives the staff time to prepare for the work beforehand. Many minor repairs can be accomplished on the run instead of waiting for scheduled shutdowns. Period repairs made during scheduled shutdowns include routine maintenance and jobs which have been shown necessary during the daily inspections. A work sheet lists these and shows priorities. The annual off-season maintenance schedule should be planned carefully, charted, and status followed. The chart should include job identification, estimated man hours, and personnel assignments. GUIDELINES FOR MAINTENANCE SCHEDULING Conveyors After the first month of operation, new conveyors should be inspected for abnormal wear. This inspection is important as it forms the basis for any corrective action which may be necessary. While operating, check heavy duty conveyors hourly for broken links and missing pins. Medium and light duty conveyors may be checked once a shift. During the scheduled shutdown replace worn or bent slats and pins.
make
a complete inspection
and
During the annual shutdown overhaul completely, replacing worn chain links and pins, slats, wear strips, idlers, bearings and sprockets. Keep records of these repairs as this information is helpful for future purchase as well as scheduling overhauls. Knives While operating check the external area hourly, particularly for vibration and hot bearings. Take advantage of the longer mill delays by stopping the knives and replacing broken blades giving first consideration to balancing the set. During the scheduled shutdown inspect knives for wear and replace worn ones. During the annual shutdown replace bearings, couplings and lubrication system.
knives,
worn hubs,
inspect
Shredder While operating check hourly for vibration and hot bearings. During the scheduled shutdown inspect hammers and grates for wear and replace with built-up sets as required before excessive wear takes place. Check lubrication system and grease hammer bushings.
161
During the annual shutdown replace hammers and grates. Inspect the bearings and couplings and change the oil and grease in the lubrication system. Mills While operating check hourly. Inspect intermediate carrier chains, hydraulic system and lubrication system. Check bearing temperatures on mills and gearing. Lubricate immediately squealing bearings. Once a shift check for loose hook bolts, eccentric pins, bearing wedges and shims. Build up grooves of rolls as necessary. Mild steel or hard facing rod may be used. During the scheduled shutdown clean the mill thoroughly, check and adjust the mill openings, check the turner plates and inspect the underside of the mills. Inspect condition of the mill grooving, repairing any areas damaged by tramp iron and build up grooves. Check lubrication and hydraulic systems. Inspect intermediate carriers and pumping systems. Test overspeed safety tripout on turbines while shutting down for weekend. Schedule replacement of worn rolls and turner plates. During the annual shutdown, dismantle the mills for a complete overhaul. Inspect the cheeks, foundation bolts and juice pans. Install new rolls or regroove existing rolls, replace turner plates and scrapers and fit all bearings. Overhaul all carriers and chutes. Overhaul cush cush screens, pumps and valves. Check nitrogen pressure on Edwards hydraulics. Clean lubrication system. Check gearing for wear and alignment. Check foundations for settling. Diffuser System (Silver) Buster and Fiberizer While operating check the oil temperature bearings hourly and the heat exchanger daily.
of the
main
rotor
During the scheduled shutdown inspect hammers and grates for wear and replace with built-up sets as required before excessive wear takes place. Grease bushings on each hammer. Check the lubrication systems. During the annual shutdown replace the hammers and grates. Inspect the bearings and change the oil and grease in the lubrication systems. Diffuser While operating check the entire lubrication system daily. Inspect all pumps daily. During the scheduled shutdown inspect the entire lubrication system and supply required oil and grease. Inspect all conveyor chains, sprockets, idlers and slats. Check the rubber belt and idlers. Oil the equalizer chains. Inspect V-belt and couplings on hydraulic drive. Clean and check juice diverter and clean activating cylinder. Check distribution drain plugs. Inspect each of the lower discharge screw bushings at least once a month. Inspect the discharge screws and build-up with hardfacing as required.
162
During the annual shutdown inspect all bearings and couplings and change the oil and grease in the lube system. Calibrate Weightometer and check counter weight assembly. Inspect all rubber seals at sprockets, levellers and idlers. Dismantle and inspect the discharge screw drive couplings and the two flexible couplings above the speed reducers. Inspect the bearings, seals and gearing on the speed reducers. Inspect the power unit, cylinder, valves, filter and piping on the diffuser feed plate. Replace rubber side seals. Change hydraulic oil on the hydraulic drive. Boilers Refer to: (a) (b)
American Society Mechanical Engineers (A.S.M.E.) Boiler and Pressure Vessel Code VII, Suggested Rules for Care of Power Boilers. Instruction Manual for boiler provided by the manufacturer.
Shut down a new boiler for a few days during the second month for a complete inspection. Check the furnace refractories, grate, tubes externally for movement, erosions and soot deposits, drums internally for scale, baffles for looseness or wear, and the air and hot gas systems for abnormalities. During the annual shutdown, inspect boilers with special attention to the following points: 1. Any metal in the furnace for breakage or burning such as grate bars, links and supports. 2. Refractories opposite oil burners, adjacent to doors and small openings, and near the bottom of the furnace when subjected to damage from cleaning tools. 3.
Inside walls and outside walls for excessive bulging.
4. Tubes (outside) for sagging, erosion, corrosion near the lower drum and soot deposits. Tubes (inside) for scale and plugging. 5. Drum (outside) for corrosion and drum (inside) for scale, corrosion, and loose tubes. 6.
Support steel for sagging and burning.
7. Check level indicators, baffles, tuyeres, oil burners, soot blowers and flues, relief valves and fusible plugs. Make the necessary repairs and also anticipate the major repairs for the following year. This allows time to order material and plan the projected work load. Turbines Inspect the turbines daily.
163
During the annual shutdown: First Year: Remove top casing and inspect internals. glands, seals, bearings, valves and governor.
Check blading,
Second Year: Complete inspection and overhaul. Following Years: every 2 to 4 years.
Minor inspection
each year.
Complete overhaul
This applies to all turbines in the factory, but the annual minor inspection varies depending on the type of turbine and duty conditions. The three major classifications are: 1. Powerhouse Turbines: minor inspection would cover bearings, glands, governor and blading if machine is equipped with an inspection port. 2. Mill Turbines: minor inspection would cover bearings, governor and the valves which receive heavy service. 3. Pump Turbines: minor inspection consists of checking the bearings and governor. These simple rugged machines require a minimum of maintenance. Consult manufacturers manuals for specific recommendations. Generators Twice a shift log information from a visual inspection and temperature readings. During the annual shutdown: First Year: Complete inspection and maintenance. Following Years: Minor inspection each year. Complete overhaul whenever the drive unit is overhauled; at least every 3 years. When not in service, keep covered and use space heaters. Electric Motors All electric motors require a daily visual inspection and a temperature check. During the annual shutdown: High contamination area: should be opened and cleaned every year. Moderate contamination area: should be checked opened and cleaned approximately every two years.
each
year
and
Clean area: should be checked every year and opened and cleaned every three years. During the off-season, large motors should be covered and protected from moisture with space heaters.
164
Gear Reducers Mill and Powerhouse Twice a shift, visual inspection and temperature check should be made. On a new installation, check teeth alignment and general conditions after one month of use. A t the end of the first year, there should be an overhaul with special emphasis on teeth wear and foundation settling. After the second year, make a complete inspection. For the following years, have a minor inspection with a complete overhaul every third year. Conveyors Daily visual inspection and temperature check. Minor inspection every year. Complete overhaul every third year. Scales Before operations begin, all scales should be dismantled, repaired, cleaned and adjusted. Beam scales must be calibrated with known quantities. Make a visual inspection and check the tare on beam scales each shift. As very little can be done on a daily basis for the fixed quantity type scales, schedule another filling test for the second half of the year. Whenever welding in the area of scales, be sure the ground lead is on the section being welded to prevent any current flowing through the knife edges. Heaters Check condensates for sugar contamination. During shutdowns, clean the tubes by circulating hot caustic solution (over 40%). After the cleaning a simple static head water check is advisable. This can be done by pumping water through the heater and opening the condensate line to check for water. During the annual shutdown, inspect the tubes making the necessary replacements, change gaskets and seals and before closing check with a pressure test. Clarifiers At the end of the season, after liquidating the clarifier, immediately fill with water until time is available for cleaning. Early cleaning is advisable but if delayed, add lime or caustic soda to the water. Check scrapers, arms, shaft guides, trays, door seals, gas vents and make the necessary repairs. Paint an epoxy band approximately three
165
feet wide at the juice level to prevent corrosion caused by a fluctuating level. Filters Daily: Check screens, scrapers, juice passages, vacuum gauges and general conditions. Broken screens should be repaired or replaced as soon as possible to prevent plugging. Clean water spray nozzles. During the annual shutdown, make a complete inspection and repair any damaged screens. Replace the scraper and check the trunnions, tubing, deck plates and agitator. Periodically, a major overhaul is required changing tubing, deck plates, valves and seals. Evaporators Check condensates for sugar contamination. Clean the tubes during the weekly shutdown circulating hot caustic soda (over 40%) followed by inhibited (ferric sulfate) sulfamic acid. It is important to wash the vessels free of sugar before adding the caustic. It is also important to stop all water leaks into the system during cleaning as this will dilute the caustic solution. After cleaning make a simple static head water check. This can be done by installing a wing-bolt type manhole on the bottom of the cell. Open this manhole and after the steam chest is completely full of water, check for leaks. During the annual shutdown, replace the worn and leaking tubes, check the save-alls and feed inlet pipes, inspect domes and connecting piping, especially if constructed of mild steel, replace door gaskets and conduct a pressure test. Clean the tubes on the vapor side with alkaline permanganate followed by acid ferrous sulfate, or proprietary compounds. Pans Check condensates for sugar carryover. When the tubes become dirty, boil for one-half to one hour, with water or sulfamic acid between strikes. Additional cleanings with acid or caustic soda during the periodic repair stops should be standard practice. After cleaning, make a static head water check for leaks as with evaporator bodies. During the annual shutdown, check and replace all bad tubes, check the save-alls and feed inlets, inspect the domes and vapor pipes - especially if constructed of mild steel - check manhole gaskets and pressure test the calandria. Clean the tubes on the vapor side as with evaporators. Centrifugals Semi-Automatic Batch Daily inspection of screens, drives, brakes and general operating conditions. Gear driven batteries require additional checks and adjustments of the friction bands every week.
166
During overhaul the cylinders. A machines are
the annual shutdown, inspect the machines thoroughly and braking system, rubber buffers and governors and actuating complete overhaul should be made every two years or, if the old, every year.
Automatic Batch Daily visual inspection of screens, drives, braking, hydraulic and pneumatic actuators and general operating conditions. It is important that the electrical components (relays, contacts, timers, etc.) be kept in a clean atmosphere. During the annual shutdown, inspect the machines thoroughly and overhaul the braking system, rubber buffers, hydraulic and pneumatic actuating cylinders, solenoid valves, relays, timers and contacts. A complete overhaul should be made every two years. Continuous Daily visual inspection of the general operating conditions. A good time for this is in conjunction with the screen washing. Then, with a strobe light, the rotating basket can also be checked before opening the feed valve. Check molasses from each machine every shift for sugar crystals which would indicate a broken screen. The main basket bearings must be oil-protected during shutdowns. Therefore, on some designs, the oil pump must be run daily for a few minutes. Check screen clamping ring to insure that the screen fits tightly against basket to prevent screen flexing and breakage. If basket is worn turn down the ring to the same size. Every year the machines should be thoroughly inspected, in particular the baskets for stress corrosion. Perform a complete overhaul every two years. Crystallizers Daily visual check of the crystallizers plus a water test of the empty units looking for leaks. Repair any leaks at once. During the annual shutdown, half of the crystallizers should emptied, inspected and the necessary repairs completed.
be
Pumps On a new installation, open up the pump for an inspection after one week to one month of operation depending on the amount of wear expected. This inspection will be the basis of the repair program. All pumps should be visually inspected daily for abnormal sound and temperature rise. Many pumps utilizing alloy steel shafts and impellers on nonabrasive service can operate two or three years without an overhaul. However, during the annual shutdown, take the cover plate off and check the impeller, casing and shaft play. Where the duty is severe, an annual
167
overhaul is in order. In very severe cases, complete inspections and impeller changes are required during the season. Tanks Tanks should be washed out during the annual shutdown, then inspected. Tanks holding alkaline or near neutral liquids normally need not be painted. With good paints many tanks will go two or three years without repainting. Piping Most piping in the factory can be shutdown, after flushing out with water exception is the syrup piping. In most scraped clean. Periodic washing with cleaning will make this unnecessary.
left in at the cases, caustic
place during the annual end of the season. The this must be opened and soda during evaporator
Valves Most valves should be taken out, repaired and tested by a simple static water test during shutdown. Valves in severe service should be checked more often. Fans T
On a new installation a fan should be checked after a month s operation to determine wear. In severe wear cases, the blades or the complete rotor may have to be replaced periodically during the season. During the annual shutdown, inspect the blading, housing and bearings, making all necessary repairs. Instruments Automatic controls, properly installed, will function only with proper care. Basic rules must be adhered to or control will soon become unsatisfactory. Refer to the instrument manuals for specific instructions. An adequate supply of clean air is essential for pneumatic instruments. Clean air means free of foreign particles as well as oil and water. A separate oil-free air supply system for instruments is recommended. Pneumatic No repair work of any kind should be done by unauthorized personnel. Perform the routine maintenance that some installations such as cleaning pH electrodes, screens and diaphragms.
require
At least once a week, the instrument should be inspected, cleaning the air filters, moisture traps and pens. A t the same time check the system while raising and lowering the set point and air regulator. During the annual shutdown, clean and inspect the instruments paying special attention to the ink system, control relay intrument balance - zero and span - and tubes. Check all tubing for air leaks and plugs, and drain water from traps.
168
Electronic Electronic instruments should only be serviced by trained electronic technicians. For this reason the technician must be available at all times. An inventory of essential components, particularly of modules, must be maintained. Electronic instruments must be kept in a dust-free and vibrationless cabinet.
169
Chapter 22
GENERAL RULES OF THUMB FOR APPRAISING F A C T O R Y OPERATIONS AND OPERATIONAL TARGETS GENERAL RULES OF THUMB FOR APPRAISING F A C T O R Y OPERATIONS General approximations which are useful in rapid appraisal of factory operations are listed here. They show the order of magnitude of average figures and give a clue as to reported figures which are out of line. Cane (Mature, sound stalk, 24 months) Fiber % Pol% Refractometer solids % Purity of juice
13 14.5 16 90
Pol = Pol in first expressed juice times 0.8 (Java Ratio). 3
3
No void density (for milling) 1121 k g / m (70 l b / f t ) . Tops (At break point, mature unburned cane, 24 months, dry conditions) % of cane Fiber % Pol% Refractometer solids % Purity of juice
5 22 4.5 9 50
Fibrous Trash (full burn, dry weather, including dead cane) % of rake harvested cane 7 Fiber as harvested % Dry conditions 70 Wet conditions 30 Bagasse Pol = Pol in last expressed juice times 0.75 (mill). Density Fresh Stored
3
3
112 k g / m ( 7 l b / f t ) 3 3 160 k g / m (10 l b / f t )
Fuel value changes 1% for each 1% change in bagasse moisture. Extraction changes 0.1% for each 1% change in bagasse moisture. Mixed Juice Purity = Purity first expressed juice minus 2.5. Purity = Purity disintegrator extract of cane minus 1.5.
170
Filter Cake Tons
Tons mixed juice % insoluble solids χ 4 plus tons mixed juice times 100 100
Tons bagasse fines required equals tons insoluble solids in mixed juice. Syrup Purity = Purity mixed juice plus 0.5. Massecuites 3
3
Weight of A and Β = 1.56 m /metric ton (50 ft /short ton) 96 DA sugar. 3
3
Weight of low grade = 0.34 m /metric ton (11 ft /short ton) 96 DA sugar. Low Grade Sugar (Remelt) Final molasses recirculated at 80 purity is twice that at 90 purity. Sugar 3
3
Density commercial raw = 881 k g / m (55 l b / f t ) . Attenuation index at 99 pol = Attenuation index crystal times 3. 99 pol contains 1.3% molasses. Final Molasses Refractometer plus 5.
sucrose purity = Refractometer deleaded pol purity
Refractometer solids = 1 to 1 spindle brix minus 2.5. Quantity changes 2.5% for each change of 1 point in syrup purity. Viscosity doubles for each 10°C drop in temperature. Viscosity changes 50% for each unit change in refractometer solids. OPERATIONAL TARGETS A summary of general operational targets for factory operations is listed here. The targets are intended to be used only as guides and operating conditions must be taken into consideration. Cane Preparation Displaceability index Mill Diffuser
87 94
Imbibition % of fiber milled
250
171
Bagasse Pol% Mill Diffuser
2 1
Last Expressed Juice Refractometer solids % Mill Diffuser
3 1.5
Filter Cake Pol%
1
Syrup Refractometer solids %
68
Massecuites Refractometer solids % Purity Crystal content
A
Β
C
92 85 50
92.5 75 40
96 58 35
Sugar Pol% Moisture % Crystal size mm Temperature ° C (maximum)
99 0.2 0.8 40
Final Molasses Refractometer solids % Purity above expected Temperature ° C (maximum)
83 6 40
Undetermined Loss Pol%
1.7
173
Chapter 23
CANE STRUCTURE The stalk of the sugar cane plant is a two-phase system, solid and liquid. The solid phase is a cellulose-lignin-pentosan complex known generally as fiber. The liquid is a water solution, the juice, containing a large variety of organic and inorganic substances, close to 90% of which is sucrose. The structure of the stalk is shown in the cross sectional drawing of an internode (Fig. 23-1). The hard rind encloses a matrix of thin-walled parenchyma cells in which are imbedded the vascular bundles. The rind and the vascular bundles constitute what is commonly called the "fiber" of the fibrous portion while the sheet-like parenchymatous tissue is referred to as "pith." There are thus two types of materials which make up the solid phase of the stalk reported as fiber in the customary methods of analysis. The parenchyma tissue forms the walls of storage cells which hold high density, high purity juice. Inside the relatively sturdy vascular bundles are the conducting vessels which extend throughout the plant from the roots to the leaves. Through this piping system, water and nutrients move from the roots to all parts of the plant and products of photosynthesis translocate from the leaves. The fluid in this conducting system is, therefore, highly variable in composition and is lower in purity and more dilute than the juice in the storage cells. Being protected by the thickwalled fiber bundles, it is not so readily expressed as the juice in the storage cells. There are, therefore, two types of liquid - juice in the storage cells, which might be called static juice, and juice in the circulatory system, which could be called transit or dynamic juice. The vascular bundles are closer together at the rind and in the nodes so the fiber content is higher. Also the juice in these regions is lower in purity and sucrose content because of larger amounts of vascular juice. The range in analytical values for a cane averaging 12.5% fiber, 15.5% pol and 90 purity is as follows:
Internode Rind section Center section Node Rind section Center section
Fiber
Pol
%
%
Purity
20-25 5.5-6.5
12-14 17-19
87-89 91-94
13-14 11-12
10-10.5 13.5-14.5
75-76 86-88
174
Fig. 23-1. Cross section of cane stalk.
175
It must be kept in mind that these are analyses based on separation of components by rough methods. Actually, the composition within the growing plant is unknown because the components are altered in the process of separation. For example, the juice expelled from a stalk of cane by the application of pressure usually has a pH in the range of 5.2 to 5.5. Some mechanism is active whereby sucrose can be stored at this pH for long periods of time without inversion. There is evidence, again from separated materials, that some of the water is not affiliated with the juice but with the fiber - the so called bound water which may be 20 to 25% of the weight of the fiber. JUICE The total juice in the cane is referred to as Absolute Juice or Normal Juice. In most factory control systems, the quality of this juice is calculated, based upon the analysis of mill juices. The assumption is made that the total juice in the cane consisted of the sum of the juice extracted, as appearing in Mixed Juice, and that remaining with the fiber in the bagasse. Furthermore, the juice remaining in the bagasse is usually assumed to have the same composition as that of Last Expressed Juice (juice obtained in dewatering bagasse). Since Last Expressed Juice is always of lower purity than Mixed Juice, the calculated purity of Absolute Juice is always lower than that of Mixed Juice. With the advent of direct analyses of cane and subsequent return in many factories of processing of cane by diffusion rather than milling, calculated figures for Absolute Juice were no longer necessary. Substantial differences were then noted to exist between calculated and measured values. Reasonable certainty exists that the composition calculated from Mixed Juice and Last Expressed Juice does not represent the original juice in cane, nor is that found in the direct analysis of cane the same. In the intact cane plant, juice and fiber have an equilibrium homogeneity. When separated, both change. Mixed Juice is all the juice recovered in the extraction plant. Its composition depends upon the quality of the cane, the efficiency of the extraction process and the changes occurring under the conditions of extraction. The changes are largely unknown, at least not established, but they indicate those caused by microbiological action, chemical changes such as inversion, and solubilization of solid material. The extent of these changes is variable but investigation has shown that Mixed Juice from a milling plant is about 1.5 points lower in purity than juice in cane as determined by disintegrator analysis. Tests on diffusion plants have shown similar differences. In this case, it has also been established that there is an increase in soluble solids in the diffuser, indication that solubilization is taking place. In the diffuser, there is no biological activity, but some chemical inversion can occur having an effect on purity. It is also a fact that in the process of direct analysis of cane by disintegration, some changes occur. So, it must be concluded that the
176
original juice in the cane is higher purity and lower soluble solids content than routinely reported. The juice is also affected by the quality and quantity of the trash in the cane entering the extraction plant. Cane tops in particular have a significant influence because they contain expressible juice of low purity (around 50). Even dry leaves, although containing no expressible juice, are subject to leaching, and nonsugars, especially colored substances, enter the juice. The result of all these effects is that the quality of juice from which the factory must make sugar is lower than that of the growing cane stalk. FIBER The two types of fiber - fiber bundles and parenchyma - differ in physical properties but chemically they are similar. The fiber bundles are long, tough and stringy. The parenchyma (pith) is thin, weak and paper-like. The fiber bundles, providing the structural integrity of the cane stalk, are concentrated mainly in the rind, while the parenchyma is the internal cell wall material. The relative quantities of the two vary therefore with the age of the cane, stalk diameter, growth pattern and variety. It is true also that the strength and toughness of the fiber bundles varies with these and in particular with variety. Some canes have strong fibers, others brittle and readily fractured ones. The disintegration and milling properties of a cane change with the quality of the fiber. A strong-fibered cane may be difficult to disintegrate but feeds well in a mill because of the long fibers. A weakfibered cane will disintegrate readily but has poor feeding characteristics because of the short and sometimes powdery fiber. VARIETIES Commercial cane varieties are bred primarily for optimum yields of sugar. The breeding program has led to canes which are at a relatively constant yield level and which do not vary much in composition. The fiber content, juice solids, purities and general characteristics are close to the same. The geneticists effort therefore is mainly in the direction of producing disease resistant varieties and new clones to replace those suffering from yield decline. So in general varieties do not play a major role in factory processing. The varietal characteristics which do show up are usually in milling where it is possible to make adjustments to accommodate fiber quality differences. REFERENCES 1
Martin, J. P., Sugar Cane Diseases in Hawaii, Advertiser, Honolulu, 1938, 295 pp
177
Chapter 24
REGIONAL VARIATIONS IN SUGAR CANE PROCESSING The Hawaiian sugar industry is characterized by: Administrative cane with few independent growers Two-year cropping Year-round factory operations with a shutdown for yearly maintenance of one to three months Minimum manpower with 40-hour labor week Mechanical harvesting, principally by pushrake Extremely varied rainfall and terrain Production of raw sugar only Other sugar regions have some of these conditions, none have all. Administrative cane, for example, is the rule in Brazil and Peru. Two-year cropping is the practice in Peru and to a limited extent in South Africa. Colombia and Peru harvest cane the year around. Peru harvests extensively by pushrake, a practice which is just beginning in Brazil. No other area attempts to harvest under heavy rainfall conditions. The sum of Hawaiian conditions imposes upon the factory the necessity of handling a robust cane which may contain more than 50% extraneous material tops, leaves, weeds, soil, gravel, rocks and tramp iron. The material entering the factory, therefore, is not cane but a crude ore varying in cane richness. A massive cleaning plant becomes a necessity for preparing the cane for milling and diffusion. These cleaners are the unique characteristic of Hawaiian factories. Beyond the same effective, extraneous
the cleaning plant the unit factory operations are essentially as in other sugar regions. Since no cleaner is completely however, the succeeding steps must handle a greater load of matter and equipment is subject to heavy wear.
In the following sections is a comparative discussion of practices with those in other sugar regions. CONTROL By far the greatest amount of sugar cane in the world is grown on a one-year cycle, harvested by hand and delivered to the factory relatively free of extraneous matter. Harvesting is done in dry weather and the factory operates less than six months of the year. Under these conditions it is possible to start factory control accounting with the weight of cane. This being impossible in Hawaii because of the large amount of extraneous material, control starts with the weight and analysis of extracted juice. The quantity of cane can only be arrived at by calculation and estimation. This constitutes the principal difference in Hawaiian control methods.
178
Recovery balances are made on the basis of the weights of juice, final molasses and sugar and the analysis of the streams from samples taken by continuous or intermittent means. The basic analyses are pol and refractometer solids. Mixed juice must be analyzed for insoluble solids in order to correct the gross weight to weight of liquid. MILLING Cane milling in Hawaii is characterized by low mill speeds. Extraction efficiency suffers from high roll wear caused by abrasive soil constituents brought in by mechanical harvesting. Otherwise, milling is much the same as in other areas. Major emphasis is on good cane preparation and low bagasse moisture. Shredders are an essential part of the milling tandem and are commonly placed after a two-roll crusher. This arrangement protects the shredder hammers from large rocks and also reduces the power necessary for the shredders. The most common feeding aid is the under feed roll, although there are several heavy-duty pressure feeders in use. Coarse circumferential grooving is not used because of the adverse effect on extraction. Although groove wear is severe, periodic arc hardfacing helps in maintenance. Messchaert grooves are standard. A hot compound maceration system is routine with dilution in the range of 200 to 250 percent on fiber. DIFFUSION Present-day basic principles of cane diffusion were etablished in ! Hawaii in the 1950 s. The importance of proper cane preparation for bed permeability and high extraction and the advantage of nonmoving cane bed are features which have been followed. Although the term diffusion is universally used, the process in Hawaii is recognized by a more proper description as juice displacement - a process involving the separation of the juice from fiber by displacement with water. In several regions a hybrid system of combined milling and diffusion has been introduced and is referred to as bagasse diffusion. In this, the cane first passes through one or two mills, then through a diffuser, then through mills for dewatering. This combination of milling and diffusion requires extra equipment and runs counter to sound technology. It does not improve extraction. The Hawaiian diffusion plant consists of leveling knives, two heavy-duty shredders in series for cane preparation and the stationary bed Silver ring diffuser. Moving bed diffusers, which were the principal types installed in the decade 1970-1980, have some advantage in flexibility over the fixed bed type. With a moving bed, however, it is not possible to control the liquid level and avoid channelling, particularly on the sides of the bed, so efficiency falls below that of a fixed bed diffuser. Bagasse dewatering remains the major unsolved problem in diffusion. The original installations in Hawaii avoided the use of mills. The extrusion screw press and the cone press proved to be effective
179
dewatering devices and were used for many years. Continuing high maintenance costs led to their replacement by mills. Both of these machines, nevertheless, are attractive in principle and it is reasonable to predict that inventive effort will bring the return of something similar with the obsolescence of massive mills. CLARIFICATION Since only raw sugar is produced, simple clarification is employed in Hawaii. To minimize scaling, it is normal to use a mixture of lime and magnesium oxide added as a slurry to cold juice. Heating and clarifier settling follow. Polyelectrolytes are added to the heated juice at a rate of about.two parts per million. Similar procedures are followed in other sugar areas where raw sugar is the product, except that magnesium oxide is rarely used. Liming may be done on hot juice. Any advantage of this has not been universal and mechanically it is difficult to perform. Although a wide variety of clarification aids may be found in use, the economic value of any but polyelectrolytes is problematical. In phosphate-deficient juice, increasing the level of phosphate may be helpful, however. In factories producing a direct consumption sugar (plantation white), the most common practice is sulfitation. Sulfur dioxide is added to the juice followed by liming and heating. In comparable procedures, the juice may be limed partially or completely first, followed by sulfur dioxide treatment. The remainder of the clarification, settling and filtration, remains the same as in simple clarification. If a better quality of sugar is desired, syrup sulfitation is an added step. SUGAR BOILING Sugar boiling procedures are universal in principle, but the chosen system of handling massecuites in many areas appears to follow the fancy of the technologist. The final result can be the same regardless of the route involved in going from syrup to final molasses by removing sugar by crystallization. The efficiency of operation depends upon the extent that conditions are controlled. Thus instrumentation is a vital necessity and those factories with advanced control instruments outperform those depending upon manual control. The problem with instruments in many areas, however, is the absence of local maintenance facilities, so it becomes necessary to rely on skilled operators. Hawaiian cane yields high purity juices making a 2-massecuite boiling system impracticable, so a standard 3-massecuite system is used. Material flow is contercurrent with flow of low purity effluent toward final molasses. Α-strikes are boiled primarily on syrup. B-strikes are boiled on Α-molasses along with the remelted low grade sugar. A l l strikes are seeded with a slurry of sugar in isopropyl alcohol ground in a ball mill. A and Β sugars together constitute commerical sugar, which averages close to 99 pol. No crystallizers are used on commerical sugar. No sugar dryers are used and sugar is stored and shipped in bulk.
180
Low grade massecuites are cured in water-cooled crystallizers for relatively long periods (48 to 72 hours). This period is necessary with characteristically slow crystallizing massecuites from high salt content juices. Massecuites are reheated to close to the saturation point in massecuite heaters (up to 60° C) before centrifuging. In many sugar locales where cane growing conditions are less favorable, such as those with winter frost, or at the other extreme low-lying locations near the equator, juice purities are lower. If they fall below 80, it is possible to use a two-boiling system making only one commercial strike and one low grade strike. This is often done with satisfactory recoveries. It has the advantage of simplicity and of producing only one quality sugar. The practice of using crystallizers on A - and B-massecuites as well as low grade is also common throughout the world. Retention time is short (2 to 6 hours) and often continuous crystallizers are used. Sugar dryers are common both for raw and plantation white sugars. STEAM GENERATION AND UTILIZATION Because many Hawaiian factories must supply considerable power for the plantation irrigation system and furnish surplus electricity to the community utility grid, they have been traditionally energy-efficient. Boilers are high pressure for the cane industry, ranging from 32 to 85 2 2 k g / c m (450-1200 lb/in. ) (3103-8274 kPa) pressure. Most furnaces are designed to burn the bagasse in suspension, which is possible with the finely divided bagasse. Air preheaters and economizers are normal. Boilers are completely instrumented and control is automatic. In most instances there is only one boiler per factory. The cooperative tie-in with the public power utility permits the factory to sell surplus power to the utility and buy power when the factory is not running. Boilers are equipped to burn oil when necessary. Recent design changes have been directed toward burning of waste products in order to reduce oil consumption. This is in sharp contrast with most cane sugar areas where little or no 2 electricity is exported. Boilers are usually low-pressure, 10 to 16 k g / c m 2 (150-225 lb/in. ), (1034-1551 kPa), and multiple smaller units are employed. Furnaces tend to be hearth-burning. Instrumentation is minimal with manual control. Supplemental fuel, commonly wood, is burned when necessary. Whereas in Hawaiian unit factory operations, close attention is paid to steam economy; such has not been the rule in many other areas. Ordinarily, there should be ample bagasse to supply internal factory needs without much attention to economy. With efficient boilers and sound boiling house steam utilization, there should be an excess bagasse of the order of 25 percent.
181
MANPOWER Hawaiian factories are noted for the efficient use of manpower. Workers are employed the year-round whether the factory is operating or not. Their retirement benefits are close to their pay while working. The standard work-week is 40 hours, so factories normally operate on a 5-day week. Factory operations are automated as much as practicable making manpower requirements minimal. Total manning, including shops, for a factory processing 200 tons cane per hour would average 120. Actual operators per 8-hour shift would be under 20. Such a system requires close attention to maintenance of equipment. Outside contractors are utilized extensively for major repairs and new installations.
183 Chapter 25
ENGINEERING
SPECIFICATIONS
Factory installations should be designed to conform to accepted engineering standards. These standards are well established and are updated periodically. In the United States the essential standards are those of: American National Standards Institute (ANSI)* 2 American Society of Mechanical Engineers (ASME) American Society for Testing and Materials (ASTM) American Petroleum Institute (API) American Welding Society (AWS) National Electrical Manufacturers Association ( N E M A ) Other manufacturing countries have analogous and usually equivalent standards. Although the design specifications for each piece of equipment must be detailed to fit the individual requirements, all are subject to general standards. The most basic of these standards are summarized in this chapter. VESSELS Vessels include non-pressure tanks and unfired pressure vessels such as juice heaters, evaporator bodies and vacuum pans. Non-Pressure Tanks Non-pressure tanks are fabricated from mild steel with a minimum thickness of 6 mm in accordance with A P I standard 12F, for tanks smaller than 5 m diameter, and A P I standard 12D, for larger size. Pressure Vessels Pressure vessels are fabricated in accordance with the ASME Boiler and Pressure Vessel Code, Section VIII, Rules for Construction of Unfired Pressure Vessels. Construction may be based on Division 2 of the Code which is more rigorous than Division 1, but Division 1 is usually adequate. Vessels subject to internal pressure are designed for the maximum operating pressure plus 10% or 2 kg/cm (19 kPa), whichever is greater. Vessels subject to external pressure are designed to withstand atmospheric pressure. A minimum corrosion allowance of 2 mm is added to the thickness of the plate.
calculated
184 Juice Heaters Standard juice heaters are fabricated of mild steel with stainless steel 2 tubes and designed for working pressures of 6 k g / c m (59 kPa) on the juice 2 side and 2 k g / c m (19 kPa) on the steam side at a maximum juice velocity of 2 m/sec. Tubes are 38 mm outside diameter made of 18 gauge type 304 stainless steel, annealed and pickled. ASTM-A-249 is a suitable specification. Tube sheets have a minimum thickness of 28.5 mm. Evaporators Standard evaporator bodies are fabricated from mild steel with stainless steel tubes and designed for a working pressure of 2 kg/cm (19 kPa). Tubes are 38 mm outside diameter made of 18 gauge type 304 stainless steel, annealed and pickled. ASTM-A-249 is a suitable specification. Minimum thickness of steel plate is 16 mm for shell, 19 mm for top and bottom and 28.5 mm for the tube sheets. One front sight glass is located just above the tube sheet, with a light source at the back at a height of 1.5 m above the tube sheet. Vacuum Pans Vacuum pans are fabricated from mild steel with stainless steel tubes 2 and designed for a working pressure of 2 k g / c m (19 kPa). Tubes are 100 mm outside diameter made of 18 gauge type 304 stainless steel, annealed and pickled. ASTM-A-249 is a suitable specification. Minimum thickness of steel plate is 16 mm for shell, 19 mm for top and bottom and 28.5 mm for the tube sheets. Optimum tube length is 800 mm. Striking capacity is a maximum of 1.5 m above the tube sheet. Bottom is stream-flow design. Front sight glasses provide visibility from the tube sheet to the striking level, with a light source at the back above the striking level. A 508-mm diameter man-door is fitted above the tube sheet. A top-driven mechanical circulator is standard. PIPING Process and utility piping is designed in accordance with ANSI-B-31.3 Code for Pressure Piping. General piping is fabricated from schedule 40 ASTM-A-53 black steel pipe. Minimum wall thickness for exhaust, vapor and vacuum lines is as follows: Diameter mm
Wall thickness mm
400 - 720 720 - 1,150 over 1,150
The minimum corrosion allowance is 1.25 mm.
6 8 10
185 Design standards are for the velocities.
following maximum allowable stream
Stream
Maximum velocity m/sec
Exhaust steam Vapor (pressure) Vapor (vacuum) Air Water Juice Syrup Molasses Massecuite
30 30 60 30 2 1.5 1 0.5 0.2
Piping, larger than 50 mm diameter, is generally all-welded. Stainless-steel piping is appropriate for piping subject to rapid corrosion such as mill juice lines. Piping for temperatures exceeding 50°C requires insulation. VALVES Valve specifications include hydrostatic testing by the manufacturer at twice their normal working pressure. All valves working at pressures above 10 kg/cm (98 kPa) are steel-body type. Valves over 40 mm diameter require flanged connections. Low-pressure flanges are ASTM-A-181 grade 1/11 standard; high-pressure flanges are ASTM-A-105 grade 1/11 standard. Gate Valves Gate valves are used for complete shut-off purposes. They are commonly used on steam, vapor and water lines; they may be used for massecuites and other streams. Valves on steam lines over 200-mm diameter are fitted with a pressure-equalizing bypass and globe bypass valve. Valves on exhaust steam and vapor lines over 300 mm diameter are similarly equipped. Butterfly Valves Butterfly valves are all metal construction with an indicator showing the position of the vane. They are used primarily for control purposes. When used on exhaust steam and vapor lines over 300 mm diameter, they are fitted with a pressure-equalizing globe bypass valve. Diaphragm Valves Diaphragm valves are fitted with synthetic rubber discs with sealed bonnets and bleed hole for indication of rupture. They are of general use for juices, syrup, molasses, milk of lime and caustic soda.
186 Globe Valves Globe valves are used primarily for flow control but also, should be used on bypass lines on pressure-reducing valves. Check Valves Check valves up to 50 mm diameter are all brass swing type. over 50 mm are swing type but have iron bodies bronze fitted.
Valves
Pressure Reducing Valves Pressure reducing valves are made with steel bodies and hardened stainless steel or stellite discs. They are fitted with a globe bypass valve. Control Valves Control valves used for pressure reducing or flow control are provided with a bypass to permit servicing while the equipment is in operation. PUMPS Pumps, in general, conform to A P I Standard 610. Recommended design capacity is 25% greater than the maximum working capacity. Selection is made on the basis of pump performance curves, with care, to remain below the point of maximum efficiency. Materials of contruction for most uses consist of a cast-iron casing, bronze impeller and carbon steel shaft. These pumps can be used on water, clarified juice, syrup and molasses. Pumps for mill juices, mixed juice *nd muds are designed of corrosion and abrasion-resistant materials //ith stainless steel shafts. Glands are water-sealed. Use of mechanical seals is limited. Pump drives are in accordance with A P I Standard 601. frame-mounted with flexible couplings.
Drives are
ELECTRICAL All electrical equipment and material is designed in accordance with the standards of ANSI and N E M A and designated for continuous operation. Motors Standard motors are 440 enclosed type, fan cooled.
V, 60
cycle,
3 phase, induction,
totally
Motors up to 15 kw may have direct, on-line starters. Larger motors have separately housed control, center-type combination starters with circuit breaker, magnetic switch and 3-element thermal relay, designed to operate in the range of 10% overvoltage to 15% undervoltage. Motors larger than 15 kw are fitted with tubular-type heaters which will hold the windings higher than ambient temperature but not above the operating temperature.
187 Low Voltage Switchgear Switchgear is designed in accordance with ANSI C37.13 and N E M A SG3 standards. Elements are removable and consist of a fuseless 3-pole air circuit breaker, overcurrent relay, interpole barriers and quenches, mechanical pushbutton trip and space heater. GEARS Gears are designed to ANSI B6 standards. Gears are designed for continuous service and to withstand stall conditions. General service factors, based upon the power rating of the prime mover are: Service Factor Mill, knives and shredder drives Cane, bagasse and sugar conveyors Massecuite pumps Cane feeder table and crystallizers Fans Centrifugal pumps and others
2.00 2.00 2.00 1.75 1.75 1.25
TURBINES Steam turbines are specified according to N E M A SM-20 or A P I 615 standards. Turbines for mill drives are capable of 200% starting torque. Normal operating speed is set at 80% of maximum speed with a capability of maintaining full load at 50% of maximum speed. Normal fittings include the following: Partial load and overload hand valves. Woodward oil relay governor with electrically 2-to-l remote speed changer and hand-operated speed changer for full range at the turbine. Independent, overspeed governor with hand-operated trip at turbine, operating an independent shut-off valve.
the
Relief valve on the exhaust line ahead of the shut-off valve. Independent lubrication system with integral shaft-driven oil pump and electrically-driven auxiliary pump with duplicate elements including low oil-pressure and high oil-temperature alarms, low oilpressure automatic trip, cooler, relief valve and gauges. Remote control solenoid trip. - Excess back-pressure automatic trip. High-pressure and exhaust-steam isolating valves. Complete instrumentation including tachometer; pressure gauges for high-pressure steam, exhaust steam and nozzle-pressure; and oilpressure gauge.
188 BOILERS Boilers are designed to ASME PG and PW standards. Bagasse-fired boilers are designed to burn bagasse of 50% moisture content at a boiler efficiency of 68% with 42% excess air. Supplementing oil burning capability is provided. Standard furnace equipment includes spreader-stoker bagasse feed, oil burner, traveling grate, forced- and induced-draught fans, wet scrubber, air pre-heater and economizer. A feedwater deaerator and continuous blowdown equipment are necessary. Boiler control is automatic with safety trips for low water level, induced draught fan failure, instrument air failure and oil burner flameout. REFERENCES 1 2 3 4 5 6
American National Standards Institute (ANSI), 1430 Broadway, N e w York, N Y 10018 American Society of Mechanical Engineers (ASME), 345 East 47th Street, New York, N Y 10017 American Society for Testing and Materials (ASTM), 1916 Race Street, Philadelphia, P A 19103 American Petroleum Institute ( A P I ) , 1801 Κ Street, N . W . , Washington, D . C . 20006 American Welding Society ( A W S ) , 2501 N . W . 7th Street, Miami, FL 33125 National Electrical Manufacturers Association ( N E M A ) , 2101 L Street, N . W . , Washington, D . C . 20037.
189
SELECTED REFERENCE BOOKS FOR THE TECHNOLOGIST Introduction to Cane Sugar Technology
G. H. Jenkins, Elsevier, Amsterdam, 1966, 478 pp.
Cane Sugar Handbook
G. P. Meade ôc J.C.P. Chen, 10th Ed., John Wiley <5c Sons, New York, 1977, 947 pp.
Handbook of Cane Sugar Engineering
Principles of Sugar Technology
E. Hugot, 2nd Ed. with collaboration and translation by G. H. Jenkins, Elsevier, Amsterdam, 972, 1079 pp. P. Honig, Elsevier, Amsterdam, Vol. I 1953, 767 pp, Vol. II 1958, 567 pp, Vol. Ill 963, 711 pp.
The Manufacture of Sugar from Sugarcane
C. G. M. Perk, Sugar Milling Research Institute, Durban, 1973, 212 pp.
Manufacture and Refining of Raw Sugar
V. E. Baikow, 2nd Ed., Elsevier, Amsterdam, 1982, 588 pp.
Machinery and Equipment of the Cane Sugar Factory
L. A . Tromp, Norman Roger, London, 1936, 644 pp.
Cane Sugar
Noel Deerr, 2nd Ed., Norman Rodger, London, 1921, 644 pp.
Principles of Cane Sugar Manufacture
J. G. Davies, Norman Rodger, London, 1938, 144 pp.
Technology for Sugar Refinery Workers
Oliver Lyle, 3rd Ed., Chapman <5c Hall, London, 1957, 663 pp.
Evaporation
A . L. Webre, C. S. Robinson, The Chemical Catalog, New York, 1926, 500 pp.
The Efficient Use of Steam
Oliver Lyle, Her Majesty s Stationery Office, London, 1947, 912 pp.
!
190
Beet-Sugar Technology
R. A . McGinnis, 2nd Ed., Beet Sugar Development Foundation, Fort Collins, 1971, 835 pp.
By-Products of the Sugar Cane Industry
J. M. Paturau, 2nd Ed., Elsevier, Amsterdam, 1982, 366 pp.
Sugar Factory Analytical Control, Official Methods of the Hawaiian Sugar Technologists
J. H. Payne, ed., Elsevier, Amsterdam, 1968, 190 pp.
System of Cane Sugar Factory Control
J. L. Clayton, ed., 3rd Ed., International Society of Sugar Cane Technologists, Brisbane, 1971, 87 pp.
Sugar Analysis ICUMSA Methods
F. Schneider, ed., ICUMSA, Peterborough, 1979, 265 pp.
Laboratory Manual for Queensland Sugar Mills
5th Ed., Bureau of Sugar Experiment Stations, Brisbane, 1970, 250 pp.
Laboratory Manual for South African Sugar Factories
2nd Ed., South African Sugar Technologists Association, Durban, 1977.
Basic Calculations for the Cane Sugar Factory
J. Eisner, Booker Brothers, McConnell, London, 1958, 24 pp.
Handbook of Cane Sugar Technology
R. B. L. Mathur, Oxford <5c IBH, New Delhi, 1975, 498 pp.
191
SUBJECT INDEX
A C O N I T ATE Crystals, calcium magnesium, 114 ANALYSIS Procedures, 1-8 Schedule, 9-11 ASH Molasses, 117-119 conductivity, rate of crystallization effect, 125 exhaustibility effect, 117 rate of crystallization effect, 118, 119 Sugar, 131 BAGASSE Analysis, 4 Calorific value gross, 138; net, 138 Dewatering, 39, 45 Drying, 143 Fiber, 49 Fiber Index, 28 Fines, filter, 64 Fractional Fiber Content, 28 Measuring, 3 Moisture diffusion, 45 extraction effect, 50 fuel value effect, 51 milling, 49 Operational targets, 171 Pol diffusion, 41, 59 milling, 50, 51, 55, 59 Rules of thumb, 169 Sampling, 3 Screening, 64, BOILER, 137-145 Control, 137 Design, 137 Efficiency, 137, 138
Engineering specification, 188 Heat balance, 138, 140 Maintenance, 162 Operation air supply, 142 bagasse quality, 142 oil firing, 143 stack emission, 143 water 143 Sensible heat loss, table, 139 Water analysis, 8 blowdown, 142 composition, 140 control, 143 deaeration, 141 heating, 141 sampling, 8 treatment, 140, 141 BOILING POINT E L E V A T I O N , 77, 83 BOILING, SUGAR, PROCEDURES, 88, 102 BOOKS, R E F E R E N C E , 189-190 BUSTER,35 CANE Cleaning, 15-21 diagram, 16 efficiency, 18 fibrous trash removal, 15, 17 losses, 20 rock, gravel, sand removal, 15 thinning, 15, 17 washing, 15, 17 waste disposal, 20 Composition, 19, 173 absolute juice, 175 fiber, types, 176 fiber bundles, 176 juice types, 175 parenchyma, 176
192 total juice, 175 Deteriorated, 119, 125 crystallization rate effect, 125 crystal elongation effect, 125 Direct analysis, 175 Drought affected, 126 Dry weight produced, 19 Extraneous matter delivered to factory, 19 diffusion effect, 56 fuel value, 144 milling effect, 54 removal, 15-21 rule of thumb, 169 Field analysis, core, 2 analysis, direct, 2 extraneous matter, 19 measuring, 1 sampling, 1 Preparation cell rupture, 24, 43 diffusion, 35 displaceability index, 24, 31, 43 milling, 24 operational targets, 170 shredders, 24 Prepared analysis, 3 measuring, 3 sampling, 3 Processing, regional variations clarification, 179 control, 177 diffusion, 178 manpower, 181 milling, 178 steam generation, 180 sugar boiling, 179 Stalk analysis, 3 cross section, 174 rule of thumb, 169 sampling, 3 structure, 173 Tops composition, 19 fuel value, 144
percent, 19 rule of thumb, 169 Trash delivered to factory, 19 diffusion effect, 56 fuel value, 144 milling effect, 54 removal, 15-21 rule of thumb, 169 Varieties, 176 C A N E C L E A N E R , 15-21 carding drum, 15 cascader, 17 collared rolls, 17 combing drums, 17 diagram, 16 efficiency, 18 losses, 20 Olsen rolls, 17 operation, 18 sink-float bath, 15 tumbling conveyor, 15 velocity bath, 15 waste disposal, 20 water reuse, 20 C A N E S A L V A G E R , 18 C A R D I N G D R U M , 15 C A S C A D E R , 17 C E L L R U P T U R E , 23, 24, 31, 43 CENTRIFUGALS Batch machines, 113-115 Continuous machines, 115, 116 Maintenance, 165 C E N T R I F U G A T I O N , 113-116 Commercial sugar, 113 Low grade sugar, 114, 115 Operation, 115, 116 C I R C U L A T I O N , V A C U U M P A N , 86, 87 CIRCULATOR, MECHANICAL, V A C U U M P A N , 87, 102 C L A R I F I C A T I O N , 61-70 Bagasse fines, 64 Clarifiers, 68 Filters, rotary vacuum, 64-68 Juice heaters, 63 Liming, 61 Magnesium oxide, 62 pH, 61 Phosphatation, 62
193 Polyelectrolytes, 61 Sulfitation, 63 C L A R I F I E D JUICE Screening, 70 CLARIFIERS Flash tank, 68 Holdover juice, 69 Maintenance, 164 Operation, 69 Settling area, 69 Settling space, 69 C L E A N E R , C A N E , 15-20 C O L L A R E D R O L L S , 17 COLOR Clarification effect, 61 Juice, 61 Sulfitation effect, 63 Sugar, 128, 129 C O M B I N G D R U M S , 17 CONDENSATE Analysis, 8 Boiler use, 140 Composition, 140 Exhaust, 140 Removal, 75 Sampling, 8 Vegetal, 140 CONTROL Boiler, 137 Boiler water, 143 Diffuser, 46 Factory, 1 Milling, 32 C O N T R O L L E R S , 157 CONVEYORS Maintenance, 160 CRYSTAL Color, 128 Content, 86 Elongation, 119, 125 Growth rate, 107 Measurements, 107 Size, 81, 128 Yield, 106 CRYSTALLIZATION BY C O O L I N G , 95-99 C R Y S T A L L I Z A T I O N R A T E , 95, 105-107, 118 Ash effect, 118, 119 Conductivity effect, 125
Crystal elongation effect, 119 Crystal growth rate, 107 Deteriorated cane effect, 125 Deposition and diffusion, 95 Purity effect, 106, graph 108 Reducing sugar-ash ratio effect, 125 Supersaturation effect 105, 107 Temperature effect 105, 107 CRYSTALLIZER, LOW GRADE Design, 96 Instrumentation, 97 Maintenance, 166 Operation batch, 97 continuous, 98 continuous, diagram, 99 C R U S H E R , 24 DENSITY Bagasse, 169 Cane, 169 Sugar, 170 DESSIN F O R M U L A , 78 DETERIORATION FACTOR, S U G A R , 130 DIFFUSER Flow balance diagram, 42 Gradients, graph, 38 Juice percent fiber, diagram, 44 Silver ring design, 36 operation, 41 DIFFUSION, 35-47 Bagasse dewatering, 39 Cane preparation, 35 Comparison with milling, 59 Diffuser gradients, graph, 38 Effect of extraneous matter, 56 Flow balance diagram, 42 Juice displacement, 36 Juice percent fiber, diagram, 44 Maintenance, 161 Operation, 41-47 bagasse dewatering, 45 bed depth, 41 bed permeability, 41
194 cell rupture, 43 control, 46 dilution, 43 diffuser speed, 41 draft, 41, 43 flooding, 45 press return, 43 temperature, 45 Press juice percolation rate, 40 treatment, 39 Recovery and losses, 59 Silver diffuser design, 36 operation, 41 DILUTION Diffusion, 43 Milling, 30, 31 Operational target, 170 D I S P L A C E A B I L I T Y I N D E X , 24, 31, 43 D O U W E S - D E K K E R , formula, 120 D R A F T , 41, 43 D R A I N A G E , 25, 27 DRYING Bagasse, 143 Sugar, 115 EISNER, 148, 150 ENGINEERING SPECIFICATIONS, 183-188 Boilers, 188 Evaporators, 184 Gears, 187 Juice heaters, 184 Motors, 186 Piping, 184 Pumps, 186 Switchgear, 187 Tanks, 183 Turbines, 187 Vacuum pans, 184 Valves, 185 ENTRAINMENT Evaporator, 76 Separators, 76 EQUILIBRIUM RELATIVE H U M I D I T Y , 133 ESCRIBED V O L U M E , 28 E V A P O R A T I O N , 71-80
Multiple effects, 71 Vapor bleeding, 72 EVAPORATOR Capacity, 73 Calculations, 77-80 Engineering specifications, 184 Maintenance, 165 Operation control, 74 condensate removal, 75 condenser, 74 entrainment, 76 malfunctions, 77 non-condensible gas, 75 scaling, 75 scale removal, 76 EXPECTED PURITY, FINAL MOLASSES Douwes-Dekker formula, 120 T Hawaiian Sugar Planters Association formula, 117, 118 Hugot formula, 120 Sugar Research Institute, Australia, formula, 119 EXHAUSTIBILITY, F I N A L MOLASSES, 117 Douwes-Dekker formula, 120 Hawaiian Sugar Planters' Association formula, 117, 118 Hugot formula, 120 McGinnis, chart, 109 Reducing sugar-ash ratio, 117 Sugar Research Institute, Australia, formula, 119 Viscosity effect, 117-119 E X H A U S T I O N , MASSECUITES, 106, chart 109 EXTRACTION Bagasse moisture effect, 50 Diffusion, 35, 59, 60 Extraneous matter effect, 55 Milling, 23, 59, 60 EXTRANEOUS MATTER Diffusion effect, 56 Fuel value, 144 Milling effect, 54 Removal in cleaner, 17-20 Rule of thumb, 169
195
E X T R U S I O N , 29, 31 FALSE G R A I N , 129 FANS Maintenance, 167 FEEDING DEVICES Over feed roll, 25 Pressure feeder, 26 Two-roll feeder, 26 Under feed roll, 25 FIBER Bagasse, 49 calorific value, 138 fiber index, 28 Fractional Fiber Content, 28 Cane, 176 fiber bundles, 176 internode, 173 node, 173 parenchyma, 176 Fuel leaves, 144 stalk, 144 top, 144 Trash, fibrous, 54 FIBERIZER, 35 FIBER I N D E X , 28 FIBROUS TRASH Delivered to factory, 19 Diffuser effect, 57 Fuel value, 144 Milling effect, 54 Removal in cleaner, 17 Rule of thumb, 169 FILTER A B I L I T Y , S U G A R , 130 FILTER, R O T A R Y V A C U U M , 64-68 Bagasse fines, 64 Cake pol, 68 quantity, 68 washing, 67 Maintenance, 165 Mixing fines, with settlings, 67 Operating action diagram, 66 Speed, 67 Station diagram, 65 Vacuum, 67 FILTER C A K E Analysis, 5
Measuring, 5 Operational target, 171 Pol, 68 Quantity, 68 Rule of thumb, 170 Sampling, 5 Washing, 67 FLASH T A N K , 68 F L A S H I N G , 68 F L O O D I N G , DIFFUSER, 45 F L U E G A S , 138-140, 143 F O R W A R D SLIP, 29, 31 F R A C T I O N A L FIBER C O N T E N T , 28 FUEL VALUE Bagasse, 51 Fibrous trash, 144 GAS Flue, 138-140, 143 Non-condensible, 75 GEARS Engineering specifications, 187 Maintenance, 164 GENERATORS Maintenance, 163 G R A I N , FALSE, 129 GROOVING Chevron, 27 Circumferential, 26 Juice, 27 Messchaert, 27 1
HAWAIIAN SUGAR PLANTERS ASSOCIATION Molasses exhaustibility formula, 117, 118 HEAT Latent, evaporation, table, 149 Sensible, flue gas, table, 140 Utilized, steam, table , 151 HEAT B A L A N C E Boiler, 137-140 Factory, 80 HEATING SURFACE Vacuum pan, 87, 102 HUGOT Final molasses expected purity formula, 120 HUMIDITY
196 Sugar storage effect, 133 HYDRAULIC LOADING Mills, 31 IMBIBITION Effects, 31 Operational target, 170 Quantity, 30 Temperature, 30 INCLUSIONS Crystal, 128, 129 I N S T R U M E N T A T I O N , 153-158 Controlling, 157 Crystallizer, 97 Indicating, 153 Vacuum pans commercial, 88 low grade, 102 INSTRUMENTS Controlling, 157 Indicating conductivity, 156 consistency, 156 density, 156 flow, 154 level, 155 pH, 156 pressure, 154 thermocouples, 154 thermometers, 153 Maintenance, 167 Operation, 158 INVERSION Diffusion, 59 Milling, 59 Processing, 61 JUICE Absolute, 175 analysis, 4 calculated, 4 Clarified analysis, 5 sampling, 5 screening, 70 Diffusion percent fiber, diagram, 44 Filtrate analysis, 5 sampling, 5
First expressed analysis, 4 sampling, 4 Holdover, 69 Last expressed analysis, 4 sampling, 4 operational target, 171 Mixed analysis, 5 measuring, 5 sampling, 5 rule of thumb, 169 Press return analysis, 5 sampling, 4 treatment, 39 Total in cane, 175 JUICE DENSITY C U R V E S , 32, 33 JUICE D I S P L A C E M E N T , 36 JUICE HEATERS Engineering specifications, 184 Juice temperature, 63 Juice velocity, 64 Maintenance, 164 Scaling, 64 KNIVES, 23 Maintenance, 160 KRAJEWSKI C R U S H E R , 24 LIMING, 61 LOSSES Clarification, 121 Diffusion, 59 Filter cake, 68, 171 Inversion, 59, 61 Mechanical, 121 Milling, 59 Molasses, 121, 171 Processing, 121 Sugar crystallization, 121 Undetermined, 121, 171 LOW G R A D E S U G A R C R Y S T A L L I Z A T I O N , 95-103 Crystallization by cooling, 95 Crystallizer operation, 97 Example, 102 Instrumentation, 102 Procedure, 102
197
Seed slurry calculation, 102 Vacuum pan crystallization, 98 Vacuum pan operation, 100 M A C E R A T I O N , 30, 31, 170 M A G N E S I U M OXIDE, 62 M A G O X , 62 MAINTENANCE, EQUIPMENT, 159-168 Boiler, 162 Centrifugal, 165, 166 Clarifier, 164 Conveyor, 160 Crystallizer, 166 Diffuser, 161 Fan, 167 Filter, 165 Gear reducer, 164 Generator, 163 Heater, 164 Instrument, 167 Knives, 160 Mill, 161 Motor, 163 Piping, 167 Pump, 166 Scales, 164 Shredder, 160 Tank, 167 Turbine, 162 Vacuum pan, 165 Valve, 167 MASSECUITE A-massecuite boiling, 81-86, 88 quantity at 85 purity, 90 quantity at 80 purity, 91 seeding, 85 B-massecuite boiling, 86, 88 quantity at 85 purity, 90 quantity at 80 purity, 91 C-massecuite boiling, 98-103 quantity at 85 purity, 90 quantity at 80 purity, 91 Analysis, 6 Operational targets, 171 Quantity calculations, 89 Rules of thumb, 170
Sampling, 6 Three-massecuite system, chart, 110 Viscosity crystal content effect, 105 purity effect, 105 solids effect, 105 temperature effect, 105 M E S S C H A E R T G R O O V E S , 27 MC GINNIS, MOLASSES EXHAUSTIBILITY C H A R T , 109 MILL R A T I O , 28 M I L L I N G , 23-34 Comparison with diffusion, 59 Extraneous matter effect, 54 General effects, 31 Mill setting, 28-30 escribed volume, 28 fiber index, 28 fractional fiber content, 28 mill ratio, 28 procedures, 29 reabsorption factor, 29 turner plate, 30 work opening, 28 Operation, 27-33 Power, 31, 32 Recovery and losses, 59 MILLS, 23-25 Maintenance, 161 MOISTURE Bagasse, 49 extraction effect, 50 fuel effect, 51 Sugar, 130 MOLASSES A , B, C analysis, 6 purities, 86 sampling, 6 MOLASSES, F I N A L , 117-120 Analysis, 7 Ash, 117 Decomposition, spontaneous, 135 Drought cane effect, 126 Exhaustibility ash effect, 117 Douwes-Dekker formula, 120
198 Hawaiian Sugar Planters' Association formula, 117, 118 Hugot formula, 120 McGinnis chart, 109 reducing sugar-ash ratio, 117 Sugar Research Institute, Australia, formula, 119 viscosity effect, 117-119 Handling, 134 Measuring, 6 Operational target, 171 Percent pol in syrup, table, 123 Percent 96 D A sugar, 124 Purity-refractometer solids, 1.2 supersaturation, 101 Recirculation, low grade sugar, 106 Reducing sugar-ash ratio, 117 Rules of thumb, 170 Sampling, 7 Storage, 134, 135 Temperature effect, 135 Viscosity-temperature relations, 96 MOTORS, E L E C T R I C A L Engineering specifications, 186 Maintenance, 163 M U L T I P L E EFFECTS, E V A P O R A T O R , 71 N O N - C O N D E N S I B L E G A S , 75 N O N S U G A R S , 89-92 N U C L E A T I O N , 81, 85 OLSEN R O L L S , 17 O P E N C E L L S , 23, 24, 31, 43 OPENING Mill, work, 28 Turner plate, 30 O P E R A T I O N A L T A R G E T S , 170-171 Bagasse, 171 Cane preparation, 170 Filter cake, 171 Final molasses, 171 Imbibition, 170 Last expressed juice, 171 Sugar, 171 Syrup, 171 Undetermined loss, 171
P A N D R O P , 86 PAN, VACUUM Circulation, 86, 87 Design, 87, 102 Heating surface, 87, 102 Instrumentation, 88, 102 Maintenance, 165 Mechanical circulator, 87 PERCOLATION RATE Press juice, 40 PERK, 105 pH Boiler water ,141 Clarification, 61 Deteriorated cane, 119 Diffuser, 38, 45 Holdover juice, 70 Molasses, 61 Process streams, 61 Syrup, 68 Unlimed juice, 59 P H O S P H A T A T I O N , 62 PIPING Engineering specifications, 184 Maintenance, 167 POL Bagasse diffusion, 41, 59 mill, 59 targets, 171 Cane, 169, 173 Diffuser juice, 38 Filter cake, 68, 171 Open cells, 23, 24, 31, 43 Sugar, 127, 171 Tops, 169 Undetermined loss, 171 P O L Y E L E C T R O L Y T E S , 61 POWER Boiler design, 137 Factory requirements, 148 Milling, 31, 32 Relations of steam, 151 P R E P A R A T I O N , C A N E , 24 Diffusion, 35, 43 Displaceability Index, 24, 31, 43 Milling, 24, 31 Open cells, 23, 24, 31, 43 PRESSURE F E E D E R , 26 PUMPS
199
Engineering specifications, 186 Maintenance, 166 PURITY Cane juices, 175, 176 sections, 173 Crystallization rate effect, 106 Drop, massecuites, 86-87 Final molasses continuous centrifugal increase, 116 exhaustibility, 117 expected, Douwes-Dekker formula, 120 expected, Hawaiian Sugar T Planters Association formula, 117, 118 expected, Hugot formula, 120 expected, Sugar Research Institute, Australia formula, 119 operational target, 171 recirculation, low grade sugar, 106 rule of thumb, sucrose-pol, 170 Juice rules of thumb, 169, 170 Low grade sugar molasses recirculation effect, 170 Massecuites A-boiling, 81 B-boiling, 86 C-boiling, 100 operational targets, 171 quantity, 89-91 viscosity effect, 105 Molasses boiling point-refractometer solids at 1.3 supersaturation, 84 refractometer solids, 1.2 supersaturation, 101 SJM formula, 121 Syrup molasses quantity effect, table, 123
molasses-sugar ratio effect, 124 recovery, effect, table, 122 rule of thumb, 170 QUALITY Cane, 119 Sugar, 127 R A T E OF C R Y S T A L L I Z A T I O N Ash effect, 118 Crystal growth, 107 Crystal elongation, 119 Diffusion and deposition, 95 Purity effect, 106; graph, 108 Supersaturation effect, 105 Temperature effect, 105 R E A B S O R P T I O N F A C T O R , 29, 31 R E C I R C U L A T I O N OF MOLASSES Low grade sugar, 106 R E C O V E R Y F A C T O R S , 121-126 REDUCING SUGAR-ASH RATIO, 117-120 R E F E R E N C E BOOKS, 189-190 R E F R A C T O M E T E R SOLIDS Diffuser juice, 38 Mill juice curves, 32 Molasses purity-temperature relations, 101 REGIONAL VARIATIONS, CANE PROCESSING, 177-181 Clarification, 179 Control, 177 Diffusion, 178 Manpower, 181 Steam generation and use, 179 Sugar boiling, 179 REPORT Daily, form, 12 Recovery and Loss, form, 14 Weekly, form, 13 RILLIEUX P R I N C I P L E , 71 R U L E S OF T H U M B Bagasse, 169 Cane, 169 Fibrous trash, 169 Filter cake, 170 Final molasses, 170
200 Low grade sugar, 170 Massecuites, 170 Mixed juice, 169 Sugar, 170 Syrup, 170 Tops, 169 SAMPLING Procedures, 1-8 Schedule, 9-11 SCALE, EVAPORATOR Composition, 75 Removal, 76 SCALES Maintenance, 164 SCHEDULE, MEASURING, SAMP L I N G , A N A L Y S I S , 9-11 SCREENING Bagasse, 64 Clarified juice, 70 SEED S L U R R Y Particle size, 106 Preparation, 106 Quality, 129 Quantity commercial sugar, 88 low grade sugar, 102 SEEDING Commercial sugar, 85, 88 Low grade sugar, 100, 102 SETTING, MILLS, 28 SETTLINGS, C L A R I F I E R Filtering, 64 Mixing fines, 67 SHREDDERS, 24, 35 Displaceability Index, 24, 35 Silver Buster, 35 Silver Fiberizer, 35 Maintenance, 160 SILVER Buster, 35 Fiberizer, 35 Ring Diffuser, 35, 161 S I N K - F L O A T B A T H , 15 SJM F O R M U L A , 121 SPECIFICATIONS, E N G I N E E R I N G , 183-188 Boilers, 188 Evaporators, 184 Gears, 187
Juice heaters, 184 Motors, 186 Piping, 184 Pumps, 186 Switchgear, 187 Tanks, 183 Turbines, 187 Vacuum pans, 184 Valves, 185 SPEED, MILL R O L L , 31 STEAM Balance, 148 Power relations, 150 Pressures, 147 Process requirements, 148 Use, 147-152 evaporator effects, 149 evaporator supply, 148 exhaust, 148 high pressure, 148 juice heating, 149 processing, 148 sugar boiling, 150 STEAM G E N E R A T I O N , 137-145 Boilers control, 137 design, 137 efficiency, 137 operation, 142-143 water, 140-141 STORAGE Molasses, 134 Sugar, 133 SUCROSE CRYSTALLIZATION Data, 107 crystal growth rate, 107 single crystal measurements, 107 Generalizations, 105-106 crystal yield, 106 exhaustion, 106 purity effect, 106 supersaturation effect, 105 temperature effect, 105 viscosity effect, 105 SUGAR Commercial analysis, 7 ash, 131
201 caking, 134 centrifuging, 113-115 color, 128 crystal size, 128 crystallization, 81-93 decomposition, spontaneous, 134 deterioration factor, 133 filterability, 130 handling, 133 high pol, 110, 114 measuring, 7 moisture, 130, 133 operational target, 171 pol, 127 quality, 127-131 rule of thumb, 170 sampling, 7 storage, 133-134 washing, centrifugal, 114 Low grade analysis, 7 centrifuging, 114-116 crystallization, 95-103 recirculation of molasses, 106 sampling, 7 washing, centrifugal, 116 Recovery, 121 Remelt analysis, 7 sampling, 7 S U G A R BOILING A-massecuite, 81 B-massecuite, 86 C-massecuite, 98 Example, 88, 102 Instrumentation, 88, 102 Low grade massecuite, 98 Procedures, 88, 102 Three-massecuite system, chart, 110 SUGAR CRYSTALLIZATION A-massecuite, 81 B-massecuite, 86 C-massecuite, 95 Commercial sugar, 81-93 crystal content and pan drops, 86 example, 88
instrumentation, 88 operational target, 171 procedures, 88 quantity, 89-93 rule of thumb, 170 seed slurry calculations, 88-89 vacuum pan 81-88 Low grade, 95-103 crystallization by cooling:, 95 crystallizer operation, 97 example, 102 instrumentation, 97, 102 procedures, 97, 98, 102 quantity, 89-93 rule of thumb, 170 seed slurry calculations, 102 vacuum pan, 98-100 Rate, 95, 105-107, 118 Vacuum pan crystallization, 81-88, 98-100 control, 82-86 boiling point elevation, 83 consistency, 84 electrical conductivity, 83 refractometer solids, 82 S U G A R Q U A L I T Y , 127-131 Ash, 131 Color, crystal, 128 Color, whole, 129 Crystal growth, 128 Crystal size, 128 Crystal uniformity, 128 Filterability, 130 Inclusions, 128 Moisture, 128 Operational targets, 171 Pol, 127 Seed, 129 S U G A R R E C O V E R Y , 121-126 Percent pol in syrup, 122 Purity effect, 121 S U G A R R E S E A R C H INSTITUTE, AUSTRALIA
202 Expected purity molasses formula, 119 S U L F I T A T I O N , 63 SUPERSATURATION Crystallization rate effect, 105 Purity, refractometer solids, saturated, 82 Purity, refractometer solids, 1.2 supersaturation, 83 Purity, refractometer solids, 1.3 supersaturation, 84 Sensing boiling point elevation, 83 consistency, 84 electrical conductivity, 83 refractometer solids, 82 SWITCHGEAR Engineering specifications, 187 SYRUP Analysis, 6 Operational target, 171 Refractometer solids, 71, 74 Rule of thumb, 170 Sampling, 6 T A N D E M , M I L L I N G , 23 TANKS Engineering specifications, 183 Maintenance, 167 T A R G E T S , O P E R A T I O N A L , 170-171 TOPS, C A N E Composition, 19 Fuel value, 144 Percent of cane, 19 Rule of thumb, 169 T R A S H , 53-57 Fibrous delivered to factory, 19 diffusion, effect, 57 fuel value, 144 milling effect, 54 removal in cleaner, 17-20 rule of thumb, 169 Diffuser effect, 56-57 Milling effect 54-56 Removal in cleaner, 17-20 Rule of thumb, 169 T U M B L I N G C O N V E Y O R , 15 TURBINES Engineering specifications, 187
Maintenance, 162 T U R N E R P L A T E , 30 U N D E T E R M I N E D LOSS Operational target, 171 V A C U U M FILTERS, 64-68 Bagasse fines, 64 Cake pol, 68 quantity, 68 washing, 67 Mixing fines with settlings, 67 Maintenance, 165 Operating action diagram, 66 Speed, 67 Station diagram, 65 Vacuum, 67 VACUUM PANS Circulation, 86, 87 Design commercial sugar, 87 low grade sugar, 102 Engineering specifications, 184 Heating surface, 87, 102 Instrumentation, 88, 102 Maintenance, 165 Mechanical circulator, 87, 102 Procedures, 88, 102 VALVES Engineering specifications, 185 Maintenance, 167 V E L O C I T Y B A T H , 15 VISCOSITY Final molasses exhaustibility effect, 117-119 purity effect, 105 rule of thumb, 170 temperature relations, 96, 105 Massecuites crystal effect, 105 purity effect, 105 solids effect, 105 temperature effect, 105 WATER Boiler analysis, 8 blowdown, 142
203
composition, 140 control, 143 deaeration, 141 heating, 141 sampling, 8 treatment, 140, 141 Cane cleaner, 20 Condensate analysis, 8 composition, 140 exhaust, 140 removal, 175 sampling, 8 vegetal, 140 Condenser analysis, 8 sampling, 8 Drain analysis, 8 sampling, 8 Imbibition effects, 31 operational target, 170 quantity, 30 temperature, 30 Maceration, 30 WORK O P E N I N G S , 28
