PART – II ROTARY EQUIPMENT
CHAPTER - 1 PROCESS PUMPS
1.1
Introduction To enable the procurement of a pump, it is necessary to convey to the vendor all pertinent process information. This process information is conveyed to the Specialist/Mechanical group of TPPL in the form of Process group's standard Pump Data Sheet. Based on this information, the Mechanical group prepares the pump specification sheet to be sent to the vendors. The pump calculation sheet helps the process engineer in assembling the required data in the pump data sheet. Following guidelines should be used in preparing pump data sheets.
1.2
Types of Pumps Pumps generally used in process industries are of centrifugal, reciprocating (piston, plunger or diaphgram type) or rotary (gear, screw, lobe or others) types. A proper selection of the type of pump to be used in a particular service has to be made.
1.3
Selection of Pump Type Except for special purpose pumps for specific services, most of the process pumps are made in standard sizes, and hence, the main problem is to select the size and type that most nearly fits the service in question. Though the final selection of the pump will be made close cooperation between the vendor and the Mechanical group, a preliminary selection of the type of pump required is made by the process engineer. Selection of thc pump type can be made on the basis of capacity-head requirements or fluid properties e.g. viscosity, solid content and corrosive or erosive nature. Use Figure 1.1 for pump type selection based on head-capacity requirements.
Centrifugal type pumps will handle liquids having viscosity up to about 200 cst. For higher viscosities, specify screw, gear or reciprocating types. If solids are present in the stream to be handled, the choice of pumps is further restricted. When solids are present, all internal passages should have adequate dimensions. If solids are abrasive, close internal tolerances between stationary and moving parts are undesirable. Generally, centrifugal type pump (especially open impeller type) is the choice for handling liquids present in them.
1.4
Fluid Properties
Properties of the fluid to be pumped require careful attention of the process engineer. The process engineer should give all the pertinent information asked for in the process data sheet regarding the fluid characteristics. When solids or abrasive material are present, specify the amount and size of particles. Similarly, if corrosive and toxic materials are involved, specify the exact nature and concentration. Specify the pour point, congealing point etc. accurately to determine the type of seal flush arrangement and any special jacketing or heating arrangement required.
1.5
Suction Conditions and NPSH Improper suction conditions are the largest source of pump troubles. Careful attention should be given to NPSH (Net positive suction head). NPSH is the net remaining pressure at the suction flange of the pump after all negative forces that restrict liquid from getting into the pump are subtracted from all the positive forces that assist liquid in getting into the pump. Two terms of NPSH are referred to:
NPSHA =
Net Positive Suction Head available in the system expressed as meters of liquid.
NPSHR =
Net Positive Suction Head required by the pump expressed as meters of liquid.
Calculate NPSHA in the system as follows:
NPSHA =
Terminal Pressure in the Vessel + Height of Fluid above Pump Centre Line (see note) + Atmospheric Pressure – (Vapour Pressure of Liquid + Friction Loss in Suction Piping up to Pump Centre Line + Entrance and Exit Losses From Vessel + Loss in Suction Filter + Loss in Control Vale, Exhaner etc., if any)
Note : 1. The height of liquid in the vessel should be taken to be at the vessel bottom tangent line. 2. Pump centre line should be considered at 3/4 meter above the ground level.
Calculate NPSHA carefully considering all conditions i.e. start-up, original fill of lines, winter operation, all control valve pressure drops, summer operation, piping conditions, exit and entrance losses. Provide NPSHA at 1 meter over worse conceived NPSHR curve
by the manufacturer. In general, NPSHR of pumps should be considered as 3.0 meters, though the exact requirement will be specified by the pump manufacturer. In case of submersible pumps, the available submergence should be specified.
NPSH is a function of flow. It should, therfore, always be determined at design capacity regardless of the total head required. It. is not uncommon for NPSHR curve to turn upward at low flows. Low flow effects can often be amplified by selecting too large a pump which forces recirculation within the pump.
When liquids at their bubble points are pumped from closed vessels, NPSHA is only the static liquid head above the pump centre minus the friction losses in the suction piping. In such cases, it is usual to elevate the vessels suitably (or sometimes cool the liquid before it enters the pump) to get a margin of NPSHA over NPSHR. Sometimes, booster pumps (low head, high capacity pumps which require low NPSH) are used ahead of main line pumps to improve the NPSHA for the main pump which has high capacity high head requirements.
1.6
Pumping Capacity Proper care should be taken in establishing the rated capacity of the pump. Normal flow rate required by the process to be pumped should be estimated. A margin over this normal should be added to take care against pump runout, internal leakage/ slippage in the pump and contingencies. Following guidelines should be useful for specifying the capacity:
Type of Pump
% margin to be added to normal flow requirement
Centrifugal 3.
Single stage
10
4.
Multi stage
10
Reciprocating
20
Rotary
25
Pumps should not be oversized beyond the above recommended margins. Operating at reduced capacities can result in increased bearing loads, reduction in pump efficiency and sometimes increased NPSH requirement. When a future capacity increase in anticipated, find out whether it can be accommodated by adding one more pump in future or using a new wide impeller instead of specifying the pump for the future requirements and running it at reduced flows now.
1.7
Discharge Conditions The suction pressure and discharge pressure at the pump suction and discharge flanges respectively should be estimated by drawing the system sketch in the pump calculation sheet. All pressure drops in the system should be considered for various conditions – start up, shutdown, fouled condition of piping/ equipment, start of run, end of run etc. The difference between the discharge pressure and suction pressure so calculated is the differential pressure or differential head when expressed in meters of liquid. The differential head required for different conditions should be estimated and specified in the pump process datasheet after taking about 10% margin over the normal differential head requirement.
A centrifugal pump will not produce a higher pressure than its shut-off pressure even if the discharge line is completely blocked. On the average, shut-off pressure is the maximum suction pressure plus 1.2 times the differential pressure depending on the characteristic curve of the pump. To avoid changes in the design at a later stage, the pump shut-off pressure should be considered as maximum suction pressure plus 1.25 times the differential pressure. All downstream equipment in the system e.g. piping, valves, exchangers, vessels etc. should be designed for this shut-off pressure. During detailed engineering, this shut-off pressure should be checked against the actual shut-off pressure specified by the vendor.
1.8
No. of Pumps and Sparing Philosophy
1.8.1 Pumps in Regular Use
Normally, only one pump is provided for regular use. In some cases, however, when pumping capacities are very large and a single pump is not available, more than one operating pump may be specified. Where substantial future capacity has to be catered for, it is advisable to put additional pump in future (for which space should be kept in the system) than to oversize the present pump.
1.8.2 Sparing Philosophy
A combination of factors including desired process reliability, service factor, service factor, operating conditions, customers maintenance philosophy and cost etc. should be considered to establish pump sparing philosophy. In general, 100% spare should be
provided for essential services and hot (above 200°C), high pressure, dirty, clogging and polymerising service. For clean fluids and non-essential services, 50% spares can be specified. If more than two pumps are running in parallel for the same service, a common standby pump can be specified. Sometimes a common spare can be provided for two compatible services.
Following criteria can be used for identifying the various service conditions : •
Essential Services ∼ Furnace and reactor charge pumps ∼ Product and reflux pumps
•
Non-essential Services ∼ Product transfer pumps ∼ Chemical and additive injection pumps ∼ Product blending and circulation pumps
•
Dirty/Clogging Services ∼ Pump feeding from tank bottoms ∼ Pumps taking suction from tower ∼ Bottoms slurry service
•
Pump Reliability ∼ Pumps having mechanical seals are generally more reliable than pumps with stuffing boxes. ∼ Rotary and positive displacement pumps are less reliable than centrifugal pumps.
1.8.3 Warm-up Connection for Standby Pumps
Standby pumps under hot services (temperature above 150°C) should be always kept heated up by providing a warm-up connection. Refer to Chapter 7 of Volume-IV of the Process Engineering Manual titled P&ID Development for typical sketches of these warmup connections.
1.8.4 Pumps Running in Parallel
If two identical pumps are simultaneously working in parallel, they will deliver the sum of the capacities to the same total head. The reasons for parallel pump operation can be:
1. Economy - two pumps and one standby can be more economical than one larger pump and a standby
2. Two or three flow rates
3. Flow rates too large for a single pump to handle
Some cooling water and fire water pumps are typical examples.
1.9
Characteristic Curves of Centrifugal Pumps Figure 1.2 is a typical characteristic curve of a centrifugal pump. The nature of head capacity curve, efficiency and NPSH curves can be different depending upon the type of pump. For a given system, the head-capacity curve of the system is superimposed on the head-capacity curve of the pump and the point of intersection of the two curves is the duty point at which the pump will operate at its best efficiency. The system head curve is a function of the system static head and pressure head which are constant and the friction head which varies with the flow.
Pump vendors will supply the characteristic curve for each pump. Following points should be considered while finally accepting the pump: • The slope of the head capacity curve should not be too steep if the pump delivers into a distribution system, since a small change in flow will cause a large change in delivery pressure. • The slope should not be too flat if the pump capacity is to be controlled by throttling discharge (manually or by a control valve). • When pumps are to run in parallel, the head-capacity curve should be stable i.e. a curve with the head constantly increasing as one approaches zero capacity. A dropping type curve (where the shut-off head is less than the maximum head) gives an unstable operation. This requirement is also valid for all pumps within battery limits having standby. • Manufacturers supply characteristic curves of pumps based on water as test fluid used in the testing shop. While handling viscous fluids in actual practice, certain corrections
need to be made for viscosity to get the actual head, capacity and efficiency. Use Figure 1.3 for correction charts for pumping viscous fluids.
1.10
Choice of Driver Electric motor is by far the most common drive for pumps in the process industry. Occasionally, special considerations such as reliability of power, safety considerations and criticality of service require turbine drives to be used. Sometimes, the plant utility balance makes turbine drives necessary on large pumps. Steam turbines are effective for standby pumps for a few larger, vital pumps such as charge pumps, cooling water pumps, unit pump out pumps or flushing out pumps. These considerations have to be firmed up during the design basis stage.
Steam turbines can be condensing or non-condensing type. The choice of non-condensing or condensing operation is affected by the demand for the exhaust steam. Generally, pump drives are not made condensing type without a rigorous review as the increased complexity or condensing is rarely worth the small savings achieved in the utilities.
In some cases like offshore platforms, oil and gas processing terminals etc., gas turbines are used as drivers for large pumps.
1.11
Utilities for Pumps All pumps will need some utilities – electrical power for motor driven pumps ; steam for steam-turbine driven pumps; cooling water for condensers of steam turbines; cooling water for bearings, casing pedestals and packing; steam for seal quench, jacketing etc. Though the exact requirements of different utilities needed by a pump will be specified by the pump vendor, it is essential for a process engineer to have a fairly good idea about these requirements so that they can be included in the utility summary sheet of the plant. These requirements can later be firmed up after getting the vendor information. Following guidelines can be used for estimating utilities required by the pump.
1.11.1 Power for Pump Driver
Compute the power consumption for the pump as shown in the pump calculation sheet. Refer Figures 1.4, 1.5 and 1.6 for efficiencies of different types of pumps and Table 1.1 for motor efficiencies. Take 120% of the power so computed for utility estimation.
1.11.2 Steam for Turbines
Steam requirements of a steam turbine can be estimated by using Figures 1.7 and 1.8. Following instructions apply for the use of these figures. 1. Compute the delivered horse power (BHP) as shown in the pump calculation sheet. 2. Determine turbine efficiency from table below : KW 150 From 151 – 500 From 501 – 2000 From 2001 – 6000 Above 6000
Overall Efficiency % 35 45 55 65 70
3. Determine steam rate from Figures 1.7 and 1.8. Table 1.1 : Electric Motors – Recommended Sizes and Efficiencies Pump Requirement at Probable Motor Motor Efficiency @ % of Full Design Conditions Rating Load 0 – 0.5 1 81 82 82.5 0.51 - 0.75 1.5 67 73 75 0.75 – 1.00 2 75 78 80 1.01 – 2.00 3 75 79 80 2.01 – 4.00 5 81 83 84 4.01 – 6.00 7.5 75 80 81.5 6.01 – 8.00 10 80 84 85 8.01 – 12.0 15 81 85 86.5 12.1 – 16.0 20 80 83 86 16.1 – 20.0 25 83 86.5 88 20.1 – 26.1 30 83 86.5 88.5 26.2 – 34.8 40 85 88 88.5 34.9 – 43.5 50 80 85 87.5 43.6 – 52.2 60 84 88 89.5 52.3 – 65.2 75 87 89.5 90.5 65.3 – 87.0 100 84 89 91 87.1 – 114 125 85 89.5 91.5 115 – 136 150 86 89 91 137 – 182 200 88 91 92.5 183 – 227 250 90 92.5 93.5 228 – 273 300 90.5 93 94 274 – 318 350 91 93 94 319 – 364 400 91 93 93 365 – 409 450 91 93 93 410 – 455 500 91.5 93 93.5 456 – 545 600 93 94 94.5 Note : This table applies to totally enclosed motors (i.e. explosion proof)
• For non-condensing turbines and inlet steam pressure up to 18.5 kg/cm2a use Figure 1.7. For all condensing turbines and non-condensing turbines with inlet system pressure up to 43.5 kg/cm2a, use Figure 1.8. • Determine heat theoretically available from saturated steam of the anticipated inlet and exhaust pressure from graph 1 of Figure 1.7. • Correct heat available for unsaturation or superheat as indicated on the figures. • Determine steam rate from efficiency and heat available by use of graph 2 of Figure 1.7. • Estimate steam requirements by multiplying the brake horse power and the steam rate calculated from Figures 1.7 and 1.8.
1.11.3 Cooling Water Pumps
Cooling water may be required for a pump for cooling its bearing, pedestral, gland packing, etc. This water can be fresh raw water, circulating cooling water or sea water. Generally, the water after cooling the various parts of the pumps is routed to drain. Use the following general guidelines for estimating water requirement for pump cooling :
Up to 120°C Up to 250°C Above 250°C
0.5 m3/hr 1.5 m3/hr 2.0 m3/hr
Some minor leakage of oil from the pump can be expected which will also be routed to drain along with the above water. Take about 200 ppm of oil content in the outlet water to drain for the purpose of effluent summary of the plant.
1.11.4 Steam for Quenching and Jacketing
Pumps handling high pour point products like Bitumen, LSHS, DMT, Phthalic Anhydride and other such products are often steam jacketed. Specify steam conditions available in the datasheet. The approximate requirement of steam for each such pump may be taken as 200 kg/hr.
1.12
Pump Sealing System Process data sheets for pumps should indicate the type of sealing to be provided. Use the following guidelines for selection of type of sealing:
Service Type of Sealing External Flushing Clean hydrocarbons and non-corrosive Single Mechanical Required chemicals which will not solidify at Seal ambient conditions at temperatures upto 300°C and pressure upto 40 kg/cm2a Clean liquids which will solidify at Single Mechanical Required ambient conditions, temperature upto Seal 300°C and pressures upto 40 kg/cm2a. Very corrosive, dirty or high pressure Double Mechanical Required service Seal Slurries
Packed Box
Not Applicable
Water clean hydrocarbons, non-corrosive Packed Box chemicals at moderate pressures
Not Applicable
Pumps taking suction from pumps where Packed Box leakage can be redirected to the sumps
Not Applicable
Typical mechanical seal arrangements for centrifugal pumps are shown in Figure 1.9. Where external seal flushing is required, the process engineer should make suitable provision to supply the flush liquid. API-610 provides recommendations on the seal flushing plan for centrifugal pumps for different services (refer Figures 1.10 and 1.11)
The bubble temperature of the external flush fluid should be about 10°C higher than the maximum operating temperature at operating pressure to avoid vapourisation of the fluid and consequent vapour locking.
1.13
Material of Construction In the petroleum and petrochemical industries, the selection of material of construction of a pump is usually dictated by considerations of corrosion, erosion, personal safety and liquid contamination. Tables 1.2 and 1.3 based on the recommendations of API-610 can serve as a useful guide to the process engineer for selection of material of construction of pumps. Obtain the recommendations of Specialist/ Mechanical group to verify these materials of construction and specify these recommendations in the process data sheets.
Table 1.2: Material Class and Material Class Abbreviation Table 1.3 : Material Recommendations for Different Services Service Fresh water, condensate, cooling-tower water Boiling water and process water Boiler feed water - Axially split - Double casing (barrel) Boiler circulator Foul water, reflux drum water, water draw and hydrocarbons containing these waters, including reflux streams Propane, butane, liquefied petroleum gas and ammonia (NH3) Diesel oil; gasoline; naphtha; kerosene; gas oils; light, medium, and heavy lube oils; fuel oil; residuum; crude oil; asphalt; synthetic crude bottoms Noncorrosive hydrocarbons, e.g. catalytic reformate, isomaxate, desulfurized oils Xylene, toluene, acetone, benzene, furfural, MEK, cumene Sodium carbonate, doctor solution Caustic (sodium hydroxide), concentration of ≤ 20% Seawater MEA, DEA, TEA – stock solutions DEA, TEA – lean solutions MEA – lean solution (CO2 only) MEA – lean solution (CO2 and H2S) MEA, DEA, TEA – rich solutions Sulfuric acid concentration 85% 85% - 15% 15% - 1% 1% Hydrofluoric acid concentration of >96% Notes : 1.
Temperature Range (°F) <212 <250 250 – 350 > 350
Material Class I-1 or I-2 I-1 or I-2 S-5 C-6
>200 >200 >200 <350 >350 <450 <450 450-700 >700 450-700
C-6 S-6 C-6 S-3 or S-6 C6 S-1 S-1 S-6 C-6 S-4
<450 <350 <210 ≥210 <200 <250 <250 175-300 175-300 <175
S-1 I-1 S-1
<100 <100 <100 <450 <100
S-1 S-1 S-9
See Note 4 4 4
5
5,6 5 6
7 8 9
S-1
7 7 7,10 7
S-1 A-8 A-8 A-8 S-9
5 5 5 5 5
Separate materials recommendations should be obtained for services not clearly identified by the service descriptions listed in this table. 2. Cast iron casings, where recommended for chemical services are for nonhazardous locations only. Steel casing (S-1 or I-I) should be used for pumps in services location near process plants or in any location where released valeased vapour for a failure could create a hazardous situation or where pumps could be subjected to hydraulic shock, for example, in loading services. 3. Mechanical seal materials, for streams containing chlorides, all springs and other metal parts should be Alloys 20 or better. Buna-N and Neoprene should not be used in any service containing aromatics. Viton should not be used in services containing aromatics above 200°F. 4. Oxygen content and buffering of water should be considered in the selection of material. 5. The corrosiveness of foul waters, hydrocarbons over 450°F, acids, and acid sludges may very widely. A material recommendation should be obtained for each service. The material class indicated above will be satisfactory for many of these services but must be verified. 6. If product corrosivity is low, Class S-4 materials may be used for services at 451-700°F. A separate materials recommendation should be obtained in each instance. 7. For temperatures greater then or equal to 160°F, all welds should be stress relieved. 8. Alloy 20 or Monel pump material and double mechanical seals should be used with a pressurized seal oil system. 9. For seawater services, the purchaser and the vendor should be agree on the construction materials the best suit the intended use. 10. Class A-7 material should be used, except for carbon steel casings.
CHAPTER – 2 PROCESS COMPRESSORS
2.1
General To enable the procurement of a compressor, it is necessary to convey to the vendors all pertinent process information. This process information is conveyed to the Specialist/ Mechanical group of the company in the form of process groups standard "Compressor Data Sheet". Based on this information, the Specialist/Mechanical group prepares the compressor specification sheet to be sent to the vendors. Following guidelines should help the process engineer to prepare process data sheets for compressors.
2.2
Type of Compressors Compressors generally used in the process industry can be classified as positive displacement, dynamic or thermal types as shown in Figure 2.1.
2.2.1 Positive Displacement Type Compressors
Positive displacement type compressors compress a constant volume of fluid in each stroke or rotation. They include reciprocating, rotary and diaphragm compressors.
Reciprocating compressors are the oldest and still the most widely used type of compressors. They consist of one or more cylinders each with a piston or plunger that moves back and forth, displacing a positive volume with each stroke. They are available from less than 1 hp to more than 3000 hp and for pressures ranging from below atmospheric to 50,000 psig (about 3520 kg/cm2a ). They run at speeds from 150 rpm to 1000 rpm. Overall efficiency of reciprocating compressors range from 75-80%. Vacuum pumps are a variation of the reciprocating compressor. Suction pressure of0.5 to 1.5 inches of mercury absolute are quite common. Vacuum pumps are characterised by high compression ratio.
Diaphragm compressors use a hydraulically pulsed flexible diaphragm to displace the fluid.
Rotary compressors cover lobe-type, screw type, vane type and liquid ring type, each having a casing with one or more rotating elements that either mesh with each other such as lobes or screws, or that displace a fixed volume with each rotation. The screw type is probably the most important for general refinery and petrochemical plant application.
Screw compressors run at speeds from 2000 to 15000 rpm. These compressors can operate on relatively dirty gases and are by nature nonlubricated. Efficiency is somewhat lower than those for comparable reciprocating compressors.
Figure 2.1 Types of Compressors 2.2.2 Dynamic Type Compressors
Dynamic type compressors include radial-flow (centrifugal), axial-flow and mixed flow machines. They are rotary continuous-flow compressors in which the rotating element (impeller or bladed rotor) accelerates the fluid as it passes through the element, converting the velocity head into static pressure, partially in the rotating element and partially in stationary diffusers or blades.
Centrifugal compressors have variable head-capacity characteristics. These are used for handling large volume of gas at relatively intermediate discharge pressures. These compressors are simple, require less maintenance and have longer reliability factors.Axial compressors are somewhat similar to centrifugal compressors, but instead of having impellers moving inside diffusers, they consist of a series of stator and rotory blades. These are used for handling very large flows at relatively low heads. Generally, they do not have a stable operating range over as wide a capacity and are therefore, considered less versatile.
2.2.3 Thermal Type Compressors
Ejectors are “thermal” compressors that use a high velocity gas or steam jet to entrain the inflowing fluid, then convert the velocity of the mixture to pressure in a diffuser.
2.3
Selection of Compressor Type Though the final selection of the compressor for a specific requirement will be done by close cooperation of the vendor and the Specialist/Mechanical group, a preliminary selection of the type of compressor required is made by the process engineer based on his process needs. Figure 2.2 should be used for the general range of application of the various compressor types.
Figure 2.2: General Range of Application of Various Compressor Types
The main differences between reciprocating and centrifugal compressors (the two most commonly used type of compressors) are summarised in Figure 2.3.
Figure 2.3: Comparison of Reciprocating and Centrifugal Compressors CENTRIFUGAL vs RECIPROCATING The advantages of a centrifugal compressor over a reciprocating machine are : a. Lower installed first, cost where pressure and volume conditions are favourable. b. Lower maintenance expense. c. Greater continuity of service and dependability. d. Less operating attention. e. Greater volume capacity per unit of plot area. f. Adaptability to high-speed-low-maintenance-cost drivers. The advantages of a reciprocating compressor over a centrifugal machine are : a. Greater flexibility in capacity and pressure range. b. Higher compressor efficiency and lower power cost. c. Capability of delivering higher pressures. d. Capability of handling smaller volumes. e. Less sensitive to changes in gas composition and density.
2.4
Fluid Properties All the properties of fluid to be compressed should be clearly specified in the process datasheet of the compressor. The composition of the fluid, moisture content, molecular weight, ratio of specific heats (K = Cp/Cv), critical temperature and pressure and the compressibility factor (Z) should be known. Volume I of the Process Engineering Manual titled Process Databook can be used for properties for pure gases. The GPSA Databook can be used to supplement the Process Databook. In case of mixed gas, calculate the properties of the mixture by using the following guidelines. Further, any impurities like solid particles, polymers, corrosive and erosive elements in the gas should be clearly defined.
2.4.1 Adiabatic Exponent (K = Cp/Cv) To calculate K, the adiabatic exponent i.e. the ratio of specific heats (Cp/Cv) of a gas mixture, first calculate MCP for the mixture based on mole fractions of individual components: n
(MCP )mix = ∑ yi(MCP )i i =1
where n is the total number of components in the mixture, yi is the mole fraction of the ith component and (MCp)i is the molar heat capacity of the ith component. The adiabatic exponent is then calculated as below: K mix =
(MCP )mix (MCP )mix − R
If MCp is in Btu/ibmol °R or kcal/kgmol °C, the value of R is 1.986. Refer to Figure 2.4 for molar heat capacities of individual hydrocarbons. Since the heat capacities vary considerably with temperature, K is normally determined at the average of suction and discharge temperatures. Aproximate adiabatic exponents for hydrocarbon gases based on average molecular weights are given in Figure 2.5.
2.4.2 Average Compressibility Factor (Zavg)
Compressibility factors, which modify the ideal gas law are important primarily for multistage reciprocating and centrifugal compressors where volumes must be calculated for each succeeding stage. For perfect gases like H2, N2, air etc., compressibility factor is 1.0. Calculate average compressibility factor of non-ideal gas/gas mixture as below:
Z avg =
Z1 + Z 2 2
Z1 = Compressibility factor at suction conditions Z2 = Compressibility factor at discharge conditions Use the following for calculating the discharge temperature T2 based on the suction temperature T1 and the suction and discharge pressures (P1 and P2) :
Figure 2.4: Molar Heat Capacities Figure 2.5: Approximate Adiabatic Exponents for Hydrocarbon Gases
P T2 = T 2 P1
K −1 K
Charts given in section 7.8 of the Process Databook can be used for calculating compressibility factors for hydrocarbon gases of different molecular weights.
2.4.3 Composition of gas
The composition of gas to be compressed should be known and the same is to be indicated in the data sheet of the compressor. In case of a mixture of gases, give the composition of the individual constituents of the mixture and the molecular weight of the mixture. Also specify the water content in the gas, if any. Moisture content is very important because when gases are compressed, their ability to hold water or other potential condensates decreases at a given temperature and provision must be made for separating the condensed material between different stages of compression. For air compressors, the inlet relative humidity is required from which the amount of water present in the air can be determined.
2.4.4 Compressor Capacity
The normal flow rate required by the process should be estimated in kg/hr.
2.5
Compressor Calculations A process engineer is required to make certain basic thermodynamic calculations for estimating the head and horsepower requirements so that he can make a preliminary assessment of the various utilities needed by the compressor (as described in section 2.10). The amount of work required in compressing a gas is dependent on the polytropic curve involved and increases with increasing values of N. Refer Figure 2.6. The path requiring the least amount of work is N = 1 which is equivalent to isothermal compression. For adiabatic (isentropic) compression, N is equal to the Adiabatic Exponent (K). It is usually impractical to build a isothermal machine. Most machines tend to operate along a polytropic path which approaches the isentropic. Most compressor calculations are therefore based on an efficiency applied to account for true behaviour. Certain type of
Figure 2.6: Polytropic Compression Curves compressors, like reciprocating and single stage centrifugal, follow an adiabatic compression cycle while others like multistage centrifugal, follow a polytropic compression cycle. The following equations are useful. The legend for the various symbols used is given at the end of this section.
2.5.1 Adiabatic Compression
(a) Adiabatic Head:
K −1 K P2 − 1 Z1 + Z2 1545 P1 H ad = T1 2 M K − 1 K
(b) Adiabatic Gas Horse Power :
HPad =
WH ad 3300ηad
Adiabatic efficiency (ηad) can be taken as 65-70% for most of the cases for preliminary estimation.
(c) Adiabatic Discharge Temperature :
(T2 ) ad
P = T1 2 P1
K −1 K
2.5.2 Polytropic Compression
(a) Polytropic Head :
H poly
N −1 N P2 − 1 Z + Z 2 1545 P1 = 1 T1 2 M N − 1 N
(b) Polytropic Gas Horse Power :
HPpoly =
WH poly 33000η poly
Polytropic efficiency (ηpoly) can be taken as 70-75% for most of the cases for preliminary estimation.
(c) Polytropic Discharge Temperature :
P (T2 ) poly = T1 2 P1
N −1 N
2.5.3 Relationship of Polytropic and adiabatic Exponents
The value of quantity N in polytropic relations is found as :
N − 1 K − 1 1 = N K η poly H poly η poly
=
H ad η ad
Legend of symbols used HP
=
Gas Horse Power
H
=
Head, ft lb/lb
Z1
=
Compressibility Factor
M
=
Molecular Weight of gas
T
=
Temperature of gas in °R
K
=
Ratio of specific heats Cp/Cv, adiabatic exponent
N
=
Polytropic exponent
W
=
Weight flow, lb/min
η
=
Compression Efficiency
Subscripts
2.6
ad
=
Adiabatic Compression
poly
=
Polytropic Compression
1
=
At suction of compressor
2
=
At discharge of compressor
Compression Ratio Compression ratio of any stage of a compressor is the ratio of absolute pressure at the discharge of that stage to the absolute pressure at the suction of that stage. It is normal practice to balance the cmpression stages so that each stage is designed for approximately the same compression ratio. This is done partly for economic reason but also to limit maximum interstage discharge temperatures. For example, discharge temperatures in lubricated reciprocating type air compressors are limited to 160°C to avoid the potential
danger of fires from a reaction of oxygen with hot oil. In general, it is not advisable to use a single stage compressor for compression ratio above 3.5 at relatively low pressures. For suction pressures above 70 kg/cm2a, a compression ratio of not more than 2.5 per stage should be considered. When multistage compressors are involved, care should be taken to provide means for routing the interstage condensate to some appropriate location in the process flow scheme.
2.7
Compressor Capacity A process engineer will usually indicate his requirement of flow in the process data sheet expressed as kg/hr. In case of recycle streams, the material balance of various recycle streams should be indicated in the process data sheet by making a system sketch. In case inlet flow in volumetric units is required, the following expression can be used :
Q 1 = V1 * W
V1 =
Z1 RT1 144P1
where, Q = inlet flow ft3/min
2.8
W
=
Weighted flow lb/min
Z1
=
Compressibility factor at suction temp.
R
=
Gas constant (1544/Mol. wt.)
T1
=
Suction Temperature °R
P1
=
Suction Pressure psia
Choice of Driver Electric motors, stream turbines and in some cases gas turbines and oil/gas engines are used to supply motive power to compressors. The selection of a particular drive is a function of utilities available, cost and preference of owners. These alternate choices should be reviewed at the stage of fixing up the design basis.
2.8.1 Reciprocating Compressors
In case of reciprocating compressors, oil/gas engine drives have certain advantages in cases where gas or oil is in abundance and no other power supply is conveniently available. This might be the case in oil or gas field installations or for pipeline pumping and compression units. Direct steam engine drive is also used frequently for reciprocating compressors. It
should be recognised, however, that waste steam or condensate from such units contains oil and cannot be directly returned to a condensate system. Steam or gas turbines are not very common drives for reciprocating compressors. If used, care must be taken to provide speed reducing gears, flywheels, torsional couplings etc. Electric motors are used quite frequently as drives for reciprocating compressors in the process plants. Motors for big compressors are generally synchronous type, whereas small compressors may use a V-bolt drive, gear reducer or direct valve with an induction motor.
2.8.2 Centrifugal Compressors
Electric motors, steam turbines and in some case, gas turbines are used to supply motive power to centrifugal compressors. Since these compressors generally operate well above 3600 rpm, speed-increasing gears are employed with motor drives. Steam turbines can either be condensing, non-condensing or extraction type, depending on such matters as steam cost and need for process steam at different pressure levels.
Gas turbines may be used to supply power to large (5000-7000 hp) centrifugal compressors, for instance in ethylene plants, offshore gas compression units, etc.
2.9
Number of Compressors and Sparing Philosophy Generally, the reliability factor for reciprocating machines vary from 95% to 98% and for centrifugal machines from 99.5% to 100%. If the client has no particular preference, following criteria (based on economic considerations) can be used for selecting the number of compressors. However, the actual number of spares will depend on the clients philosophy on operation and maintenance.
Reciprocating Up to 100 KW
One running (100%) + one standby
(100%)
From 100 KW and above
Min. of 33-50% as spare capacity
Centrifugal Since centrifugal compressors are of large size and have comparitively higher service factor, it is not conventional to have a spare unit. Also, upto 15000 hp, it is conventional to put in one full size machine rather than two 50% size machines with adequate spare parts including a complete rotor to reduce the downtime of the machine.
2.10
Utilities for Compressors All compressors need utilities like electric power, steam or gas for their drivers, cooling water for condenser, jacket, bearing, intercooler, aftercooler and oil cooler etc., fuel gas and starting air for gas turbines. Though the exact requirement of various utilities needed by a compressor will be specified by the compressor vendor, it is essential for a process engineer to have a fairly good idea about these requirements at the stage of process design work so that these can be included in the overall utility summary sheet of the plant. These requirements can later be firmed up after getting the vendor information. The process engineer should indicate the utility conditions in the compressor process data sheet. Following guidelines can be used for estimating utilities required by the compressor.
2.10.1 Power for Driver
Following steps are involved in estimating the electric power required by the compressor: (a) Calculate the gas horse power of the compressor as given in section 2.5. (b) Calculate the BHP of the compressor by dividing the gas horse power by the mechanical efficiency. Take 95% as mechanical efficiency for centrifugal compressors. Use following table for mechanical efficiencies for reciprocating compressors.
Total Compression Ratio 1.5 2 3 4 5 6 7
Mechanical Efficiency 65 74 81 83 82 81 79
(c) Applying the overload capacity requirement of the compressor and the service factor of electric motors, estimate the driver horse power as follows : Power requirement of Motor = BHP / (0.9 x Motor efficiency)
Motor efficiencies can be taken as 95-98%
2.10.2 Steam Requirement for Turbines
(a) Reciprocating Compressors : Steam requirement for the estimating BHP is calculated from Figure 1.7 and 1.8 given in Chapter 1 for pumps.
(b) Centrifugal Compressors : The turbine is specified to develop 120% of the BHP required at the design conditions. The theoretical steam rate in kg/HP (hr) can be obtained from the expression
643 h1 − h 2
where h1 – h2 is the enthalpy change of steam expressed in kcal/kg. Actual steam rate can then be obtained by dividing the theoretical steam rate by the turbine adiabatic efficiency which is given below:
BHP 500 800 1000 1200 1500 2000 2500 3000 & above
Adiabatic Efficiency 61 66 67 69 70 72 73 74
2.10.3 Fuel Gas for Gas Engines and Turbines
Gas Engines Use an average figure of 2400 kcal/BHP/hr. Estimate the fuel gas requirement based on this figure and the calorific value of the fuel gas available for use. Gas engines can be run at fuel rated power when burning any fuel gas having a heating value of 7100-23000 kcal/m3, but no more than 10 mole% butane. If it contains more than 10 mole% butane, the engine must be derated by 10% i.e. a 500 hp engine will be suitable for 450 hp.
Gas Turbines
Use an average figure of 3000-3200 kcal/BHP/hr. Estimate the fuel gas requirement based on this average figure and the calorific value (net) of the fuel gas available for use. Horse
power rating of gas turbines are based on 26.5°C combustion air temperature and 300 meter altitude. They should be derated by 5% for every 5°C increase in air temperature.
2.10.4 Cooling Water Requirements
Different types of compressor and their drives require cooling water. Estimate the cooling water as follows. Consider a 9°C rise in temperature of cooling water from inlet to outlet.
Compressor Type/ Drive
Cooling Water
Type of Water
Requirement Reciprocating Compressors - Compressor Jacket
125 kcal/BHP/hr
Treated
- Inlet Cooler
250 kcal/BHP/hr
Treated
- Aftercooler
250 kcal/BHP/hr
Treated
- Gas Engine Cylinder Jacket
1000 kcal/BHP/hr
Closed System
20 m3/hr (min.)
Treated
Centrifugal Compressors - Oil Cooler -
2.10.5 Cooling Water for Turbine Condenser Cooling water is required for condensers of stem turbine driven compressors. Considering that condenser will operate at 750 mm Hg, calculate the change in steam enthalpy from intel steam conditions to the condenser conditions. From the steam rate found in section 2.10.2, find the total heat to be removed. Considering a temperature rise of 9 oC in cooling water inlet to outlet, estimate the cooling water requirements of the condenser.
