Chapter 3 Clinker

Process Diagnosis Handbook

Chapter III: Clinker

Actualization Date: Nov 04 Code DP-03-1

Page 1

Part 1- Clnker. Its design and kiln processes.............................................................................................................3 1.1 Description of the clinker production process................................................................................................................5 Part 2- Combustion, Flame and Heat Balance..........................................................................................................23 2.1 Combustion...............................................................................................................................................................23 2.1.1 2.1.2 2.1.3 2.1.4

Solid Fuels...................................................................................................................................................................................24 Liquids........................................................................................................................................................................................30 Gases..........................................................................................................................................................................................31 Alternates....................................................................................................................................................................................31

3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 3.1.6 3.1.7

Cyclones towers...........................................................................................................................................................................42 Calciner or fuel burning in the smokes chamber.............................................................................................................................44 Kiln Tube.....................................................................................................................................................................................48 Coolers........................................................................................................................................................................................51 Production Level..........................................................................................................................................................................58 Operational Usage........................................................................................................................................................................59 Electrical Power Consumption.......................................................................................................................................................59

2.2 Flame and burner......................................................................................................................................................32 2.3 Heat Balance.............................................................................................................................................................39 Part 3- Operation and Kiln Control at Regime..........................................................................................................42 3.1 Kiln parts and their functions.....................................................................................................................................42

3.2 Control Variables.......................................................................................................................................................60 3.2.1 3.2.2 3.2.3 3.2.4

What What What What

speed does the kiln must have?...........................................................................................................................................61 draft level does the kiln must have?.....................................................................................................................................65 fuel level is the required one?..............................................................................................................................................67 feeding level does it admit?..................................................................................................................................................68

3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.3.7 3.3.8

Torque or rotation moment (motor amperage)...............................................................................................................................70 NOx............................................................................................................................................................................................71 O2 and pressures..........................................................................................................................................................................71 TV camera, zone temperature, and hot stage temperature.............................................................................................................72 Analytical.....................................................................................................................................................................................72 Exit gas temperature....................................................................................................................................................................73 Kiln view......................................................................................................................................................................................73 Weight/Liter.................................................................................................................................................................................73

3.3 Control Parameters....................................................................................................................................................69

3.4 Control.....................................................................................................................................................................73 Part 4- Kiln control and operation under Special conditions...................................................................................80 4.1 Cooling, heating and kiln start up...............................................................................................................................80

4.1.1 Starting up. Establishment level, and regime recovery....................................................................................................................86 4.1.2 Starting up abacus and kiln recovery.............................................................................................................................................88

4.2 Kiln with a low production..........................................................................................................................................89 4.3 Process protection interlocks......................................................................................................................................95 Part 5- Refractory and Kiln Repairs Protocol ..........................................................................................................96 5.1. Refractory coating....................................................................................................................................................96 5.1.1 5.1.2 5.1.3 5.1.4 5.1.5 5.1.6 5.1.7

Benefits.......................................................................................................................................................................................96 Types..........................................................................................................................................................................................96 Refractory Brick...........................................................................................................................................................................97 Bricks formats............................................................................................................................................................................101 Refractory brick placing systems.................................................................................................................................................102 Refractory life – campaign..........................................................................................................................................................104 Hot spots...................................................................................................................................................................................105

5.2. Refractory Concrete................................................................................................................................................107 5.2.1 Refractory concrete drying..........................................................................................................................................................109

5.3. Refractory Specifications.........................................................................................................................................110 5.3.1 5.3.2 5.3.3 5.3.4

Chemical request – chemical wear...............................................................................................................................................111 Thermal request – thermal wear.................................................................................................................................................114 Mechanical request – mechanical wear........................................................................................................................................116 Kiln tube corrosion.....................................................................................................................................................................119

5.4. Ovalness and kiln aligment......................................................................................................................................122 5.4.1 Kiln aligment..............................................................................................................................................................................123

5.5. Prolonged shut down of an installation (greater than 2 months)................................................................................124 5.6. General repairs protocol..........................................................................................................................................125 5.7 Refractories consumption and campaign durations....................................................................................................131 Part 6- Emissions....................................................................................................................................................133 Appendix A Part I. Potencial composition statement by Bogue.............................................................................143 Appendix B Pare I. Calibration of FRX and DRX equipments.................................................................................154 Appendix A Part II. Combustion.............................................................................................................................160 Appendix B Part II. Heat Balance...........................................................................................................................164 Appendix C Part II. Flame Calculations..................................................................................................................167 Appendix D Part II. System Temperatures Calculation.........................................................................................169 Appendix E Part II. Thermal Conductivity and Insulation.....................................................................................170




Page 2

Part 1- Clinker. Its design and kiln processes The main objective of this chapter is to expose in a simple and operative way, all those basic concepts to design clinker and to operate kilns in a satisfactory way. Regarding the clinker design, it can not be disconnected from the cement design or the availability or characteristics of the raw materials, but otherwise, as it will be seen ahead, from the raw materials nature a range of possible raw meal / clinker designs will arise, which tested in the cement desing will provide an optimal produced clinker for the cement quality to be commercialized and a total plant cost. The calculation tools which will be described are thought, as in the rest of the chapters, to be simple, of minimum data and easy to obtain, and giving results which allow planning and/or designing. That is why approaches, which can not be seen from a total inoperative point of view, are used, as well as not bibliographical equations empirically obtained which try filling dangerous emptinesses. Kiln Balance Tool



General Repairs Tool


Page 3




1.1 Description of the clinker production process The diagram exposes the main processes involved in clinker production in a dry method kiln.

Page 4




Page 5

All the phase formation processes which happen in the kiln are solid-solid reactions (schematized in the previous picture) which will depend on the diffusion speeds. The fastest to diffuse is the calcium, in Ca2+ form, then the aluminum and iron, in form of Al 3+, Fe3+ and, finally, the silica, in SiO 42- form. Without iron or aluminum, the C 2S + CaOL solid-solid reaction which gives C3S is produced over 2000°C. Given the relation between the CaO and SiO 2 molecular weights, it is established that the optimal lime, corresponding to saturation without excess or shortage for the C 3S generation, is: 3 Mw CaO VS 1 Mw SiO2 (Mw = Molar weight) and therefore: Optimal CaO = [3.Mw CaO / 1.Mw SiO2]. SiO2 = 2.8 SiO2 The aluminum presence favors the solids diffusion in this flux, lowering the temperature, which takes place over 1470°C, being the relation for optimal lime: Optimal CaO = 2.8 SiO2 + 1.18 Al2O3 Finally, the iron incorporation lowers the clinkering temperature again to a value around the 1450°C, being the optimal relation between elements: Optimal CaO = 2.8 SiO2 + 1.18 Al2O3 + 0.65 Fe2O3 The 1.18 y 0.65 values are empirical as the phases are not the pure ones which are formulated by Bogue (see Appendix A, p.129)

In this way a relation between the total real lime and the optimal one is established, denominated “saturation degree” which expresses the lime saturation with respect to the rest of elements. LSF (Lime Saturation Factor) = 100 * CaO / CaO optimal. LSF = 100 * CaO / (2.8 SiO2 + 1.18 Al2O3 + 0.65 Fe2O3) As greater the LSF level seeked, more energy will be required for a greater amount of carbonates to be decarbonated, excepting the CaO which comes from other minerals different to carbonates. The inclusion of other elements modifies this Lea&Parker module; so: 

The magnesium presence rises this degree when incorporated (up to 2%) or substituting the calcium in crystal lattices, forming analog compounds to the C3S, C2S, etcetera (M3S, M2S...)



In the other hand, the presence of calcic sulphate in the clinker lowers the lime for the clinkering reactions, so it has to be considered in its stoichiometric relation CaO/SO 3.

In this way, the LSF is corrected to the expression: LSF = 100 * [CaO* + 0.75 MgO* - 0.7 SO3 excess] / [2.8 SiO2 + 1.18 Al2O3 + 0.65 Fe2O3] *Note: Just up to 2% of MgO, since the rest is not incorporated to the clinker phases. If to the total CaO the generated free lime is eliminated, an operative LSF is given, the one the kiln “enjoyed”. The excess SO3 with respect to the alkali.




Page 6

This relation between the common elements in all the phases, the CaO, and the rest, can not be characterized by its own, a raw meal or a clinker, being necessary more information relative to the easiness (temperature) to be combined. Then other two modules arise: Silica Module (MS): is the relation between the solids and liquids which favor their diffusion, and therefore will reduce the clinkering temperature.

MS 

C S  C2 S SiO2 Solids  3  Liquids C 3 A  C 4 AF Al 2 O3  Fe2 O3

It is an indirect measurement of the liquid phase porcetange which will be seen ahead. 

Inferior MS to 2.0: generate an excess coating thickness, burning easiness and liquid phase excess.



Superior MS to 3.0: generate a little coating thickness, little liquid phase and high thermal load in order to burn.

Therefore the MS is much related with the kiln temperature at which the clinkering will happen. Alumina Module (MA): is the relation between the two main fluxes, the alumina which melts at high temperatures and the iron which does it at low temperatures MA = Al2O3 / Fe2O3 Therefore the MA is much related with the section of the kiln where this liquid phase will happen specified by the MS. 

Superior MA to 2.3: the liquid phase begins too late (towards the kiln material discharge), the liquid phase will be a viscose one.



Inferior MA to 1.3: the liquid phase begins too soon (towards the kiln material entrance), the liquid phase will be a fluid one.

Summarizing the modules concepts which characterize a raw meal / clinker and the operation in its chemical part (later mineralogy and granulometry will be seen), we get:




Page 7

Therefore (without dealing with deatails which will be seen ahead) if the flux (Fundente) usage at low temperatures is abused like the Fe 2O3 and it is compensated with a lowering of Al 2O3 the same L.P. % can be obtained, but gotten at the rear, which will produce premature clinkering with rings formation at the middle of the kiln, and clinker overburning which will rest benefits. As the L.P. expression will show other minority elements are fluxes (Fundentes) at low temperatures, as MgO, K 2O, Na2O and SO3. The formulation of maximum and expected amounts, or phases POTENTIAL, is due to Bogue, who portioned the sotichiometric relation of pure compounds. Knowing how to formulate them and getting to the potential expressions is very important, since the inclusion of new compounds (for example the fluorite) and the compounds which will form, will vary the Bogue formulas and the modules formulation, therefore they will not have an operative validity; this defect is secured raising the new reactions. In the Appendix A part 1, is detailed the development until reaching the Bogue POTENTIAL expressions, as well as an operative formulation for the fluorite inclusion case (F 2Ca), a very high capacity flux(Fundente) which forms a mineralizing pair with the sulfur, speeding up or favoring the C3S formation. Calculations of these potential phases and other behavior data can be calculated in the Raw Meal Design tool. Calculus sheet example

Raw Meal design.XLS

Sheet: Desing and loading plan


0.53 Fe2O3 + 5.4 SO3 – 4.59 K2O – 6.93 Na2O – 2.541 Al2O3 C3A = 2.65 Al2O3 – 1.696 Fe2O3 – 10.13 SO3 + 8.61 K2O + 13.06 Na2OC4A3SO3 = 7.62 SO3 – 6.48 K2O – 9.83 Na2OC4AF = 3.04 Fe2O3SULFO-ALUMINOUS CLINKER

Cooking aptitude (BF) (With the new modules)AW = C4AF + C3A + 0.2 C2S + 2 Fe2O3 + 3C2S∙3CaSO4∙CaF2Coatingability coefficient (AW)Clinkering T(°C) = 1300 + 4.51 C3S – 3.74 C3A – 12.64 C4AF – 12 (3C2S∙3CaSO4∙CaF2)

LSF + 10 MS – 3(MgO + K2O + Na2O)Cooking aptitude (BF)AW = C4AF + C3A + 0.2 C2S + 2 Fe2O3Coatingability coefficient (AW)T(°C) = 1300 + 4.51 C3S – 3.74 C3A – 12.64 C4AFRegarding the minimum clinkering temperature

C3S = 4.071 [ CaO – CaO(L)] – 7.6 SiO2 – 4.48 Al2O3 – 2.86 Fe2O3 + Correction by othersC2S = 2.867 SiO2 – 0.754 C3SC2F = 1.7 Fe2O3 – 2.67 Al2O3C4AF = 4.77 Al2O3PORTLAND CLINKER (MA<0.64)

Process Diagnosis Handbook Actualization Date: Nov 04

Code DP-03-1 Page 8

C3S = 4.07 [ CaO – CaO(L)] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 + Correction by others

C2S = 2.86 SiO2 – 0.754 C3S – 0.515 (3C2S∙3CaSO4∙CaF2)C3A = 2.647 Al2O3 – 1.69 Fe2O3C4AF = 3.04 Fe2O3WITH MINERALIZING PAIR

= 8.2 Al2O3 – 5.22 Fe2O3 + MgO* + K2O + Na2O + SO3; For MA<1.38LP1338 °C = 6.1 Fe2O3 + MgO + K2O + Na2O + SO3 For MA>1.38Liquid Phase

C3S = 4.071 [ CaO – CaO(L)] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 + Corrección_por_otrosC2S = 2.867 SiO2 – 0.754 C3SC2F = 2.65 Al2O3 – 1.69 Fe2O3 C4AF = 3.04 Fe2O3CLINKER PORTLAND (MA>0.64)

2.95 Al2O3 + 2.2 Fe2O3 + MgO* + K2O + Na2O + SO3; For MA>0.64

2.85 SO3 + 3.05 MgO*



Actualization Date: Nov 04

Code DP-03-1 Page 9




Page 10




Code DP-03-1

Page 11

It is important to clarify that the evaluation must be done based on the clinker analysis, since this is where other compounds, with no relation with the raw meal, take part, as the fuel ashes. By a natural homogeneity, the marls (tending to the natural raw meal) are more appropiate for the burning than using pure raw materials. This is, for the same material used as a silicic component:

 Better burning High limestone + much clay  Worse burning (because of its little natural initial homogeneity) Limestone marl + little clay

To talk about a clinker analitycal is to talk about its three modules (LSF, MS, and MA) and also about its phases POTENTIAL. The real phases knowledge and their amounts through X Ray Difraction (DRX) allow to connect the kiln with grinding in order to anticipate and calculate grinded cement strength resistances (chapter 4: Cement). Although the range of working values for the modules is based on an industrial strip, the reality is that the clinker design does not begins with its chemical part, but mineralogical, this is (and saving the costs) according to the mineral which constitutes the raw materials for the raw meal, that is how the combination range of modules for getting a quality and operative clinker for the kiln will be. The granulometry will be the last set in order to save the minerals problems, remembering that the kiln processes are physical-chemical, solid-solid, therefore the particule size is especially important, and even determinant. This CHEMICAL-MINERALOGICAL-GRANULOMETRICAL set is summarized in the BURNABILITY or COOKING APTITUDE, which will express how easy or difficult clinkering will result, to a given resultant free lime and a given temperature. Of the two more used methods, one is theorical (F.L.S. method) which corresponds to correlations with empiric data, and other is purely empirical (Polysius method). Both use the free lime concept obtained in a kiln laboratory at a T°C temperature during 30 minutes in order to express its burnability, being the Polysius a several temperatures relation. “Generalized” F.L.S. method (not available in bibliography): CaOF (T°C) = aT * (LSF – LSFT) + bT * (MS – MST) + cT * SiO2 Chemical factor

+45

+ dT * CaCO3 +125 + eT * Aq

mineralogical and granulometrical factor

where: Chemical part: Note: the MA does not influence significantly in the cooking aptitude, as will be seen later. LSF

= raw meal or clinker LSF

LSFT

= LSF reference to temperature T°C: LSFT = 25 + 0.05 T

aT

= temperature adjustment coefficient T°C: aT = 1.71 – 0.001 T

MS

= raw meal or clinker MS

MST

= MS reference to temperature T°c: MST = 0.40 + 0.001 T

bT

= temperature adjustment coefficient T°C: bT = 10.44 – 0.0059 T




Code DP-03-1

Page 12

Mineralogical and granulometrical part: a) 20 g. of sample are taken (raw meal) and sieved at 125 m. The retained is evauluated as in its percentage as well as its nature in the following way: Retained % = R

125

125 Seive m

It is analyzed

CaO % in that R 125

by FRX

Obtaining the Calcium Carbonate percentage OVERSIZED, this calcite which has an excessive size will difficult its decomposition; CaCO3 +125 = ( %R125 / 100 ) * [ (100/56) * % CaO

in R125

]

b) 5 g. of sample are taken (raw meal) and sieved at 45 m, evaluating the percentage that is retained (R45.) The retained is tried to dissolve with a chloridric acid dissolution (100 ml water + 20 ml HCl.) It is let boiling for 20 minutes with agitation and then it is filtered. What left as “insoluble residue” (RI) in the filter is dried for 1 hr. at 105 °C and it is weighted to obtain the %RI, and it is analyzed by RX in order to determine the SiO 2 porcentage (it will be the quartz) and the rest (other minerals which are difficult to decompose). So: HCl

%R45

Seive 45 m

%RI Rest (soluble)

% SiO2 Rest = 100 – SiO2

The “oversized” quarts that by its stability and overdimension will difficult its intervention into the reactions: SiO2

+45

=

%R 45  % RI  * % SiO 2 inRI  * 100  100 

Being stables and oversized the rest of the minerals: Aq =

%R 45  % RI  * (100  % SiO 2 inRI ) * 100  100 

And finally the adjustment coefficients are: cT = 5.35 – 0.0033 T dT = 1.87 – 0.0011 T eT = 3.98 – 0.0026 T




Page 13

With all these, for example at 1500°C, the burnability will be: CaOF 1500 = 0.21 * (LSF – 100) + 1.59 * (MS – 1.9) + 0.40 * SiO2 +45 + 0.22 CaCO3 +125 + 0.08 Aq If fluorite (F2Ca) is being used, the form calculation can be corrected (empirical not bibliographical): CaOF with F2Ca = (1 – 0.5 * %F2Ca) * CaOF without F2Ca As an example, if a raw meal has: LSF = 95, MS = 2.3 %R45 = 22.6 with RI = 5% (100% SiO2

 Aq = 0)

%R125 = 9.1 with CaO = 46.15 Its theorical burnability is: 1500°C = 1.6

1450°C = 2.8

1400°C = 4.3

1350°C = 6.0

1000°C = 26.1

Polysius Method: Uses as a burnability factor (BF) the following relation between burnabilities (Free lime) tested at different temperatures: BF = 3.73 *

CaO1350  CaO1400  2 * CaO1450  3 * CaO1500 (CaO1350  CaO1500 ) 0.25

Using the previous example, a BF of 53 will be obtained. An approximation of the relation between BF and CaO1500 could be: BF



45 + 14.45 * CaO1500

From the evaluation of the different raw meals it is observed that the importance of the different parameters regarding the burnability is the following: 1) What most affect are the oversized quartz followed by the LSF and the oversized calcite and finally the MS, except this implies quartz movement. 2) The oversized quartz is not reduced significantly with overgrinding, this is, trying to lower this oversized quartz will become uncostable. 3) If the oversized calcite is reduced considerably when overgrinding the raw meal. 4) The Alumina Module (MA) does not affect the burnability, just when there is a low MS has an effect, burning better if the MA is low. Note: the relation alumina/iron (MA) does not affect the burnability in a laboratory kiln (discontinuous operation) but it affects in an industrial one, where the time implies movement (position) along the kiln. 5) The raw meal fine particules just affect a premature nodulation if they are highly contained by clays. The anticipated nodulation would require a greater kiln speed in order to avid ring formations, while the rest of products will require a slower one for their nodulation, having a conflict with the fact that the kiln only has one transmission.




Page 14

It is important to clarify that in the laboratory test is observed how from a 5 minutes burning to a 10 minutes one, there is a free lime lowering of 50%, being the tested temperature of 1200 to 1450°C, but from there the change is small, chosing 30 minutes as a secure point for the burnability evaluation. In the bibliography other expression can be found to evaluate the burnability as well as the Ludwig cooking aptitude. CaOF

1500°C

= 0.01 * T

Ludwig

+ 0.7

Where the Ludwig “T” has the following expression: T Ludwig = 55.5 + 11.9 * R96 + 1.56 * (LSF – 90)2 – 0.43 * FL2 With LSF = 100 *

CaO  0.75MgO 2.8SiO 2  1.18 Al 2 O 3  0.65Fe 2 O 3

R96 = raw meal retained at 96 m. LP (Liquid Phase) = 1.56 * ( 3 Al2O3 + 2.25 Fe2O3 + MgO + K2O + Na2O + SO3 ) All the previous for the Ludwing calculation is based on the raw meal analitycal and not on the clinker one. From the practical point of view, it is needed to watch the kiln and to study it stage by stage, but not to the control parameters depth. Z1 – Drying zone: The raw meal humidity degree will affect on the fuel consumption and will be seen when studying the kiln energy balances. Z2 – Clays dehydration zone: The energy (heat) consumed at unbinding the water from the clays, is a small problem as they are easy to decompose and to react. In other words, the SiO 2 parts from the clays are more reactives than the quartz SiO2, as it has already seen at the Quarrys chapter Z3 – Calcination zone: The decarbonation is the most expensive process, thermally talking, of those which happen in the kiln, so all the CaO contributed which does not come from carbonates is an important saving. The decarbonation process can finish before entering the kiln tube in the case of kilns with calciner, keeping in its hot stage very high temperatures (  900°C) which, except in towers with a lot of stages, will derive in a high escape gas temperature and therefore in a thermal inefficiency. In the conventional kilns it happens otherwise, but the decarbonation consumes a lot of kiln meters, leaving less kiln for the rest of the processes, so its production levels are clearly lowered. LOI= 0.786CaO + 1.08 MgO

Z4 – Nodulation zone or liquid phase formation zone or fusion zone: As its name indicates is the zone where the compounds (C3A, C4AF) are formed and when they are fused will form the liquid phase which will favor the CaOF and C2S solids diffusions for their combination in C 3S. In this zone C2S is formed and C3S begins to appear, as nodules which size will depend on the zone length. A bibliographical definition of sintering is, “process which converts melted dust into solids”. The sintering efficiency (Clinkering) is higher when it happens as nodules, so it is very important to nodulize before getting into sintering temperatures. As smaller this zone (small diameter kilns) the nodulation begins




Page 15

before so it is easier to find clinker as balls in small kilns, than in those with a bigger diameter, which are very prone to generate clinker dust and also very sensible to MA module changes. Therefore, high flux (Fundentes) levels at low temperatures (Fe 2O3, K2O, Na2O, MgO, SO3) cause early nodulation which can derive in thick coating and large clinker balls formation, and because of this obstruction they can end being enormous, jumping the obstacle and moving forward with danger of damaging the burner. For these reasons, and when these elements levels rise, for example the SO 3 with the pet coke usage must be compensated lowering other element as the Fe 2O3. Thus, the MA relation becomes smaller so the Liquid Phase % must be evaluated at 1338°C and moving the individual element levels according to their relative influences which are contemplated in that expression. For example the relation in clinker of Fe 2O3 with respect to SO3 is 6.1 Fe2O3 vs 1 SO3. The alkali (K2O and Na2O) effects are even more delicated since they are flux (Fundentes) at 800°C, so they can generate selective nodulation at very low temperatures. The inclusion of other fluxes (Fundenetes) as F2Ca must be taken into consideration in the same line, since it seems to work towards the Al2O3 line and, therefore, towards kiln zones closer to the kiln exit. Z5 – Clinkering zone: The formed C3S amount will depend on every factor which affects the extension T C (líquid ) of its solid-solid reaction in a liquid medium (C2S(s) + CaOL(s)    C3S(s)), this is: 1) Relation between the elements (Modules): at higher modules (LSF, MS) higher the C3S potential but worse burnability, so in some cases it is convenient to lower the LSF in order to combine better and rise the C3S.

Kcal/kg

LSF 99 96

99/2.2

96/2.5

MS 2.2 2.5

C3S

2) Residence time: is defined as the necessary permanence time for the obtention of a free lime equal to 2%. This residence time is defined by some authors as: t(min) = 19 * L /(n * D *i) where: L D i n

= = = =

kiln length (m) effective internal kiln diameter, without considering the refractory (m) inclination in % rotation speed of the kiln in rpm




Code DP-03-1

Page 16

Other more complete expressions will be seen later. 3) Reaction speed, which depends on: 3.a) Clinkering temperature, being the minimum theorical: Minimum clinkering T = 1300 + 4.51 * C3S – 3.74 * C3A – 12.64 * C4AF In the appendix A it is presented an alternative when F 2Ca is used. It is important not to forget than when the thermal load is very high it results difficult to keep oxidant conditions, being possible to become a reducing condition (Fe 2+) with low strength resistances in cements. 3.b) Liquid phase viscosity. The viscosity increases when the Al 2O3, Na2O, K2O, and MnO are increased and when the F2Ca y Fe2O3, MgO y SO3 is reduced. A viscosity decrement favors the solid diffusion for the C3S formation, but raises the danger of refractory infiltrations. The low surface tension of the liquid phase also favors, being lower as the MgO and the SO3 becomes higher. 3.c) Liquid Phase %. A high L.P. % improves the medium for the C3S formation, so it is improved when raising the Fe2O3, Al2O3, MgO, K2O y SO3. 3.d) The particle size and, therefore, the raw meal granulometry. The size must be the adequate one to the raw materials mineralogy, just as it was seen in cooking aptitude, this is, it will be possible to control by sizes as long as they are not quartz or calcite. The normal values are: Seive ()

% Retained

% Passed Through

297

0.5

99.5

200

0.5 – 5

99.5 – 95

125

5 – 15

95 – 85

90

10 – 20

90 – 80

75

15 – 25

85 – 75

45

25 – 35

75 – 65

Note: being the more recommendable, the finest values of these intervals.

Understanding that one thing is to control by the most operative sieves and other thing is that they are not binded at 45 and 125 m towards the cooking aptitude. In other words, with the complete curve the two sieves which control the raw meal and its relation with 45 y 125 m are determined in order to go from its desired values into equipment control values in other sieves. Therefore the resuides at 45 y 125 m can be controlled, controlling a specific sieve, usually the 75 or 90 m. As an example it is shown common relations of raw meal: P45 = 0.077 * P751.55

P45 = 0.0018 * P902.37

P125 = 12.2 * P750.46

P125 = 3.88 * P900.71

In the other hand, the relation between liquid phase or flux(Fundente) and the raw meal avergae size is very importat in order to characterize the expected at nodulation.




Page 17

Raw Meals Behavior 1.- FL1338 = -0.04Φ50 + 28 2.- FL1338 = -0.02Φ50 + 30

% F.F. 28 20

3.- FL1338 = 0.0075Φ50 + 8

A

4.- FL1338 = 0.0435Φ50 -1.75

1338

=

3 B

8

Note: FL

1

2 C

40

A: Effective nodulation (Nodules)

D

B: Incomplete nodulation (Dust)

D

C: Sticky Raw Meal D: Paste or suspension

4 E

200

E: Melted raw meal

500

 50 (m)

6.1  Fe2 O3  MgO  K 2O  Na 2O  SO3 MA  1.38  8.2  Al O - 5.22  Fe O  MgO  K O  Na O  SO MA  1.38  2 3 2 3 2 2 3

Example: Having a raw meal with a Φ50=21 microns which generate a clinker with LP 1338 = 26% will be between the curves 1 and 3, and therefore, in the area A of “Effective nodulation”.

Note: Seive that retains and let pass a half of raw meal. Φ50 = e(50-)/By RRSB approximation b = Rx – a*Ln (x) a = R y - Rx Ln(y) – Ln (x) Seive Y (m) > Seive X (m) Rx = retained in X m

Ry = retained in Y m

If there is not a granulometer, the approximation for the raw meal can be: Φ50 = R75 – 2 = R90 + 2 The raw meal use to have an average diameter of 10 to 25 , in the white raw meals around 10  and in the gray ones 15 to 25 , so, in order to get an effective nodulation, their liquid phase must be around 23 and 28 %. With all these it can be seen that the election of an adequated chemistry for a given mineralogy, is not something limited in selecting the three modules, or electing it trying to get a maximum Bogue potential, since there is a point where the LSF relation, for a given MS, with the calorific kiln consumption becomes exponential, needing to know the economic limit for the C 3S (which use to be a LSF of 95 to 99) and in that search it is important to clarify that movements in MS decimals are equivalent, in burning hardness, to LSF points. This is, as some authors have described the chemical burnability as: AS (Sintering aptitude) = LSF + 10*MS – 3*(MgO + K 2O + Na2O), between 100 and 120 with ideal 108. So in order to keep the same burnability, if a MS decimal is incremented 1 LSF point must have to be decreased.




Page 18

In the other hand, it is important not to forget that in all these movements, the coating profiles may vary in thickness and position, being important, the coating and the assurance of the kiln campaing, at the time of selecting the modules. So, a MS above 2.5, besides difficulting the burnability by flux shortage and generating a dusty clinker (which effects will be seen ahead), causes a little coating. This high thermal load which forces a high MS is less important in calciner kilns, since the works between tube and tower are better distributed. Despite the kiln operation with high MS’s becomes more difficult. For the cases on wich there is a flux shortage at the zone and require to operate with higher LSF’s and MS’s, it is recommended to use fluorite, which added in very small amounts, will generate great liquid phase as it will be seen ahead. On the contrary, low MS modules (< 2.0) besides favoring the operations at low thermal load, produce a very granulated clinker, or even with big balls formation, hard to grind and a lot of coating which not only uses free tube section and therefore production capacity, but also the coating weight lowers the refractory life and the kiln gets instable due to continuous coating formation-falling cycles. In bibliography is commonly used as coating coefficient (AW), the following expression: AW = C3A + C4AF + 0.2 C2S + 2 Fe2O3 (*note: C2F in case of MA < 0.64) Being normal AW’s between 20 (less becomes a shortage) and 35 (higher forms excessive coating). Or the Turbain coating index ic: ic = 2.8 (SiO2 – 20) + Fe2O3 If ic < 7 it is difficult to generate coating, being the optimal between 10 and 15. The coating formation aptitude is a relation binded to the MA because it is very influenced to the refractory material adhesion at the Fe2O3 level. Therefore a low MS and a little Fe 2O3 (high MA) produce a greater kiln instability since any MA movement will generate a great change to the material amount which incorporates or leaves the coating. Therefore the MA plays a very important role, being its recommended levels close to 1.63 since at that relation is when all the Fe2O3 and Al2O3 melt at the lowest temperature of 1338°C when constituting the EUTETIC. Above 1.63 all the Fe 2O3 melts at that temperature, but not all the Al 2O3, and under that value the opposite happens. Z6 – Cooling zone: The importance of an effective cooling at the freezing of the reversible reaction of C3S formation has already been seen. Lowering the formed clinker temperature as fast as possible until 1170°C will allow cristallizing the liquid phase and catching C3S phases formed and the C2S and CaO without combine, not letting the reaction to be inverted. If there is a very slow cooling the decomposition will be done and secondary C2S and CaO will be generated, in the C2S case it will not be in a structure with resistances potencial, as in the  (beta) phase, but it will be transforme into  (gamma) which is stable under that temperature and does not have hydraulic properties. According the bibliography, this cooling speed is more critical in the case of actived clinkers with fluorite.




Page 19

It is industrially assured, excepting operations cases extremely punctually rarified, as in grates coolers as well as in satellites ones the freezing is effective, being more risky in the satellites ones since the exchange is worse and because the cooling begins before already at the kiln tube. The theory establishes that as faster is the mentioned cooling: 

The crystal sizes are smaller and with a greater tension accumulation, which will make easier the clinker grinding.



The C3S mineral is bigger when its formation reaction gets frozen more effectively.



More MgO is incorporated to the clinker phases, leaving less as free expansive magnesia. It is verified that as a result of a fast cooling, the MgO solidifies in glassy state in very small sizes (< 5) which not generate expansive phenomenons in cement. Otherwise, by slow cooling, will adquire bigger sizes that will affect.



The C3A is less reactive to the setting, decreasing the fast setting or flash tendency.

Also it is true, that at industrial levels there are few justified differences on the cooling type, while in laboratory tests, testing different cooling speeds, the informations (provided by articles) are contradictories even more in terms of C3S amount. As in the flame, regarding its easiness of transmiting heat by radiation, as well as cooling, regarding the heat exchange effectiveness between air and clinker, are very affected the dust level in the clinker, mainly in kiln with satellites coolers, proner to reintroduce in the kiln the clinker dust. That dust does not only interfere in the kiln making it to consume more fuel, but also inestabilize the coating, trying to increase the product losses in open warehouses and in this clinker rehandling. The size or classification by size in the clinker use to be expressed in terms of high dust level, understanding this as the one formed by particules smaller than 1mm, being a habitual classification: % passed through 1 mm

Clinker type

<1

Very coarse

6  28  50

Coarse Fine or with little dust

> 80

Only dust

Very dusty




Page 20

Being an optimal clinker, under this premise, the one that pass through 1 mm in the order of 15 to 20%. From the clinker grindability point of view, the bibliography is very contradictory, just agreeing in a general rule: At higher quality, beter grindability. This is, the grindability improves when increasing the C3S % and decreasing the C2S %, when reducing the alite (C3S) and belite (C2S) size and when reducing the liquid phase, some bibliography indicates that the MgO is a grindability facilitator. The clinker density (if porouser better) and the pores distribution (as more distributed better) play an important role, and in this aspect a wrong raw meal homogenization will cause clinkers to be harder to grind. An excess of liquid phase will increase the density and with it the grinding will be harmed, also the overburning and larger and weaker flames which enlarge the belite and CaO (F) crystals forming fewer alites and with a bigger size. According the bibliography the alite particles size is smaller when the lime content in the raw meal is higher and when it is finely grinded. Not confirming in all cases that the cement strength resistances increase when decreasing the alites size, and if happens, is at early ages and not at late ones. Regarding the duration and temperature at the sintering zone, the crystals “usually” become bigger since they have greater residence times and greater temperatures, the term “usually” is used since other authors, the duration (greater residence time) at the preparation zone (800 °C – 1300 °C) generate big crystals, but in the sintering zone (1300 °C – 1450 °C) are generated small ones, therefore a balance is constituted. For all these aspects the microscopy has generated much information: Microscopy and crystals size. Ono Method. Operative parameter Heating speed: by alite size (C3S)

Fast: 15 to 20 m

Slow: 40 to 60 m

Burning time: by belite size (C2S)

Long: 25 to 40 m

Short: 5 to 10 m

Burning temperature: by alite birefringence

High: 0.01 to 0.008

Low: 0.005 to 0.002

Fast: light

Slow: amber

Cooling speed: by belite color.

Definitely, in order to get quality and grindability it is needed to control the following points: 1. Intense and short flames, difficult and unnecessary in big diameter kilns with calciner, therefore it is better to say: “There is no interesting in too large and loose flames”, “There is no interesting in too slow heatings and coolings”. 2. High LSF and MS modules (LSF > 95 y MS > 2.2) and relatively low MA (< 1.6). % grindability ≈ f( C3S %) 3. Small raw meal particule sizes (Φ50 < 25), “Savings on the raw meal mills will reflect on expenses in the kiln and cement mills”




Page 21

The microscopy technic gives and excellent investigation work to specific problems, but not, for the control of the calcining procces for which, technics as the DRX give more benefits. Concluding this section, it not easy to determine a priori which is the more convenient clinker to produce in a specific installation in a given geographical point, here is a scheme that might help:

It is important not to forget the importance of the chemical part, mineralogical and granulometical stability chosen finally (LSF < 1, MS < 0.05,MA < 0.05, P75 < 1, constant minerology) To accentuate the importance that the mineralogy has in the cooking aptitude, it is presented the following study done by RDX vs. Cooking Aptitude Test, on which the different hardness are shown based on the presence of different minerals, as: the ordered calcic albite, which as not being considered bibliographically as a hard mineral towards the calcinations process, opens a whole investigation line through the minerals identificaction and quantification.




Page 22

So, the new way to talk about (new language) a raw material, will not only be by its chemical composition, but also mineralogically, as the following phrase: silicic component denominated Sandy Mud, of chemical characteristics: SiO2 = 61.76 MS = 2.78 withouht harmful components; and of mineralogical characteristics: Tufa with a %x of ordered calcic albite (from the Feldspar family) which confers a high burning hardness. (Note: being x the semiquantitative percentage that will determine the DRX equipment software) With this information, this silica corrector not only will not allow reaching high MS modules without other corrector, and besides it will contribute to the raw meal a bad cooking aptitude.




Page 23

Regarding the mineralizers: are those which, besides being normally fluxes (lowrering the fusion temperature of the liquid phase), are capable of activating or accelerating the formation of the C3S mineral, being one of its mechanisms to lower the liquid phase viscosity favoring the solid-solid reaction, this is, CaO F – C2S (the binds with K, Na favor the penetration attacking the brick at temperatures around 1450°C, causing corrosion at the high transition). Some of these mineralizers are fluorides CaF2 (1350°C, shellphous), MgF 2 (1350°C), SrF2 (1190°C), BaF2 (1285°C), alkalines or alkaline-terreous (see periodic table), P 2O5, Cr2O3, MnO and also seem to be the sulfides (BaSO 4, CaSO4- CaF2) and chlorides (CaCl2, 773°C). That is where the mineralizer “pair” F 2Ca+ CaSO4 rises, which in a 8:1 proportion (in clinker) allows that little amounts of F 2Ca be mixed with CaSO4 and C2S, forming compounds with a high molecular weight and therefore generators of great liquid phase amount. From these, the most used, because of its abundance, is the CaF 2 or fluor-spar, which added to raw meal with a liquid phase deficit, in low proportions (0.2 % – 0.4 %) they generate new phases (3C2S ٠3CaSO4 ٠CaF2, C11A7CaF2) which give a great liquid phase amount, facilitating a higher conversion to C3S, being of great utility for raw meals with a high LSF and MS as well as quartz presence (see appendix A part 1). The CaF2 behavior is not linear, this is, below the 0.2 % does not offer benefits and neither above 0.6 % (Except with high SO3 levels), besides being detrimental above the 0.6 %, it instables the C 11A7F2Ca phase in the clinker and generates low strength resistances and cracks the dough as a “Crocodrile skin”. There are other mineralizers which on the contrary are detrimentals, as the phosphates. The El P 2O5 is a mineralizer which stables the C2S being difficult its transformation to C3S.




Page 24

Part 2- Combustion, Flame and Heat Balance. In this section, it is tried to expose briefly the most essential conepts and to give a calculation tool, which will only ask for simple data of easy obtentainment, leaving the appendixes to describe the calculation models.

2.1 Combustion It is the combination of fuel with a combustive agent or oxidant (typically air) in order to generate heat. The main involved reaction is the carbon oxidation to carbon dioxide, always passing through carbon monoxide. C + O2



CO2

94 Kcal/mol (Exothermic)

The reaction might not be completed by different causes as: Oxigen shortage: C + ½ O2



CO

26.4 Kcal/mol (Exothermic)

This reaction is a low exothermic one and may have an explosion risk. An excess oxigen may suffocate the flame and generate CO presence. Contact of the incandescent carbon (activated) with other compounds generated at the combustion or present in the environment: Combustion gases: C + CO2

 2CO

-41 Kcal/mol (Endothermic)

Humidity: C + H2O



CO + H2


Calcium carbonate to be decarbonated: C + CaCO3



CaO + 2CO


These last three reactions, as being endothermic, entail an energetic efficiency lost. The other 2 combustion reactions which follows in importance are the hydrogen ones (H 2 + ½O2  H2O 58 Kcal/mol) and the sulfur ones (S + O2  SO2 71 Kcal/mol), which are exothermics. The fuels used in the cement industry are derived from organic compounds (Coal) and petroleum (Coke, Fuel Oil and Gas) and they are divided by their physical state in Solids, Liquids and Gases where the different alternate fuels will be included. The hydrocarbons amount and therefore the C/H relation is determinant in order to classify the combustion difficulty: Solids C/H >> 8 (more difficult to combust) Liquids C/H ≈ 8 Gases C/H ≈ 3 (less difficult to combust)




Page 25

Having each one different flame characteristics: Flame Fuel

Particule Size

Length

Luminosity

Emisivity

Natural Gas

Amströng

Short ≈ 15 m

Little

Low

Fuel Oil

Microns

Medium ≈ 20 m

Much

Medium

Coal/Coke

Microns

Long ≈ 25 m

Very Much

High

In the other hand the air has a main composition of 79 % N 2 and 21 % O2 in volume percentage and 77 % N2 and 23 % O2 in mass percentage.

2.1.1 Solid Fuels a) Coal The coals are composed of an aromatic part (High carbon percentage) and other hidrocarbonic part called “Bitumen”, which binds the aromatic part.

As the coal gets older, it loses its “Bitumen” through the following process described by Mackenzie & Taylor: Wood: CH4 lost in acid and anaerobic medium

Turf Lignite Soft Coal Anthracite




Code DP-03-1

Page 26

In order to difference them, 2 types of analyses are done: Ultimate analysis: Determination of the coal chemical compostion analyzing the carbon, hydrogen, nitrogen and sulfur percentage. From this analysis the Low Heat Value can be obtained according reception (LHVSR) through the following formula: PCISR = 7838 (%C/100) + 28899 (%N2/100) + 2218 (%S/100) – 584 (%H2O/100) (Kcal/kg) Example: A coal with the next compositon: % C = 69.0 % N2 = 1.44 % S = 1.29 % H2O = 5.0 % Ashes = 23.27 (ashes: inorganic not combustible components: CaO, SiO 2, Al2O3,P2O5, etc.) Will have an approximate Heat Value LHV = 5824 Kcal/kg. Proximate analysis: Determination of the humidity percentage and once the material gets dry the ashes, volatile and fix coal amounts can be determined. This kind of analysis is complemented with the sulfur percentage (by its effect in the process), the hardness (by Hardgrove method) and the calorific power evaluated in a calorimetric pump Soft Coal or Bituminous Coal Lignite

¾ greasy

½ greasy

Greasy

Anthracite

Coke

Total Humidity

10 - 25

2–6

2-6

2–6

<2

<1

Hygroscopic Humidity

10 - 30

10 - 20

10 - 20

10 - 20

5 – 10

<1

Volatile

40 - 50

25 - 40

10 - 25

5 - 10

<5

< 15

These analyses are identified in dry method or according reception. In the Heat Value case it is specified if its High (H) or Low (L), since in the High it is included the needed heat for the water evaporation formed by the hydrogen oxidation and the hygroscopic humidity contained in the fuel, which was incorporated (at molecular level) in the coal formation process and therefore it is inherent to the fuel. The subscripts DB and AR mean “Dry Base” and “As Recieved” respectively. LHVDB = 8200 – [1.8*(15 - % Volatile)2 * (100-% ashes)/100] (Kcal/kg) LHVDB = PCIDB +[70*(%Volatile)0.4] (Kcal/kg) LHVAR = PCIDB*(1-%H2O/100) (Kcal/kg) LHVAR = PCSDB*(1-%H2O/100) (Kcal/kg)




Code DP-03-1

Page 27

Example: Coal

Calculations

(dry base) C (weight %, DB)

66.6

LHVDB = 8200 – [1.8*(15 – 28.3)2 * (100-18.4)/100

H (weight %, DB)

3.99

LHVDB = 7940 Kcal/kg

N (weight %, DB)

1.07

S (weight %, DB)

1.22

HHVDB = 7940 +[70*(28.3)0.4]

O (weight %, DB)

8.85

HHVDB = 8207 Kcal/kg

Ashes (weight %, DB)

18.4

Total elemental Analysis

100

18.4 LHVAR = 7940*(1-2.35/100)

Volatile (weight %, DB)

28.3

Fix carbon (weight %, DB)

53.3

Total immediate Analysis

100

H2OTOTAL (weight %)

2.35

LHVAR = 7741 Kcal/kg HHVAR = 8207*(1-2.35/100) HHVAR =8002 Kcal/kg

b) Coke The coke is the concentrated residual in fixed carbon from the thermal cracker of the petroleum destilation residual (concentrated in fixed carbon) in which it is seeked to extract the rest of volatile compounds and of high petroleum value (Gases and gasolines). During the petroleum destilation the more volatile compounds are being liberated until the destilation residual is obtained (tars), for finally elevate its temperature to liberate the rest of volatile and to obtain the solid residual or coke. The coke also can be obtained by coals pyrogenicity (Heat breakage), normally anthracites. The same analyses as the coals can be applied.




Coal

Code DP-03-1

Page 28

Calculations

(dry base) C (weight %, DB)

89.5

LHVDB = 8200 – [1.8*(15 – 10)2 * (100-0.5)/100

H (weight %, DB)

3.08

LHVDB = 8155 Kcal/kg

N (weight %, DB)

1.71

S (weight %, DB)

4.0

HHVDB = 8155 +[70*(10)0.4]

O (weight %, DB)

1.11

HHVDB = 8331 Kcal/kg

Ashes (weight %, DB)

0.5

Total elemental Analysis

100

0.5 LHVAR = 8155*(1-1.5/100)

Volatile (weight %, DB)

10.0

Fix carbon (weight %, DB)

89.5

Total immediate Analysis

100

H2OTOTAL (weight %)

1.5

LHVAR = 8033 Kcal/kg HHVAR = 8331*(1-1.5/100) HHVAR =8206 Kcal/kg

These fuels by their solid nature will have a different behavior in their combustion depending on their granulometry. Before grinding these fuels they need to be dried, but leaving their hygroscopic humidity, since this process requires a heat amount which could ignite the fuel. That is why it is recommended that the residual humidity of the ground product to be approximately equal to the hygroscopic humidity: H2O residual = H2O hygroscopic In a practical way the coke will not be dried above the 1.5%. The order of greater sensitivity to the cokes grindability in the different mills is the following one: Vertical centrifugal (Williams) >> Vertical Roller Mill >> Balls Mills That is why the balls mills complied better with cokes or coals changes. According to the kiln combustion a grinding fineness will be seeked in function to the volatile (which are the first to burn) with the following approximation: % passed through 90 microns = 100 – 0.50 * % volatile % passed through 75 microns = 100 – 0.85 * % volatile




Page 29

The combustion process is a physical – chemical process, where the speed on which the heating liberation and volatile ignition happen and the coke or coal combustion depend on the particule size. CO – CO2

Ignition Pyrolisis1 The bitumens Volatile are liberated burn as gases (volatile) Tignition = 300 °C 100 ms

Heating 50 – 100 ms

O2

Coal combustion 1s

The burning time is a direct function of the average particule diameter (d 50), wich oscillates between 20 and 30 (m). Since the obtention of fine fraction is difficult in cokes, it is recommended to use as working sieve, one not superior to 75 microns, being recommendable to observe the particule d 50 as a fineness indicative. The solid fuel grindability depends on, the coal age and the cracking process from which the coke was obtained. In the coal, as getting older, the bituminous part begins to be lost, until reaching the anthracites with low content of volatile and high hardness. In the coke case, these can be produced by batches (Delayed), giving as result a varied result, depending on the cracking time, with HGI hardnesses (Hardgrove Index) between 35 and 70. The fluid coke (or Flexicoke) is produced continuously and it is harder, with a HGI below to 40. The fuel composition is important to define the oxidant amount to be used in order to obtain a complete combustion (stoichiometric O2, stoichiometric air); a) By elemental analysis % MassDB

Kg Minimum air/Kg fuel

Nm3 Minimum/Kg fuel

%C

11.42*%C/100

8.83*%C/100

%H2

34.33*%H2/100

26.55*%H2/100

%S

4.28*%S/100

3.31*%S/100

%O2

- 4.29*%O2/100

-3.31*%O2/100

Amin

Sum

Sum

b) In an approximate way by low heat value in dry base (Expressend in Kcal/kg) Minimum air or stoichiometric (Nm3/kg fuel): Amin = 1.01*PCI/1000 + 0.05 Gases volume of stoichiometric combustion (Nm3/kg fuel): V0R = 0.89*PCI/1000 + 1.65 (With a composition of 18 % of CO2 + SO2) The excess air with respect to the stoichiometric minimum one (Lamda, λ) varies the total amount of available air for the combustion: 1

Pyrolisis: heat breakage




Page 30

Total Air Atotal = λ * Amin Total volume of combustion gases VRtotal = λ * V0R With an approximate density of the combustion gases of 1.5 – 0.01*(%O 2) (Kg/Nm3) λ = 1/[1-(79*O2)/(21*N2)] (Requires orsat analysis N2 = 100 – %CO2 – %CO - %O2) λ = 1+0.09*(%O2) (%O2 = Oxigen excess, there where the λ is being evaluated.) In order to expose them in terms of clinker kilograms, the relation between the fuel calorific power and the kiln specific consumption must be multiplied: VR (Nm3/kg clinker) = VR (Nm3/kg fuel) * (CEH/PCI) (Kg clinker/kg fuel) CEH = specific heat consumption (kcal/kg clinker) Finally, and as a necessary heat balance data, the specific heat under constant pressure (Cp) for coal and coke is: Cp = 0.262 + 390x10-6 T(°C); Kcal/kg°C T = coke temperature/coal fed to the burner Also, the dew points for the combustion gases vary, because of the hydrogenated volatile levels, of 50 °C for lignites and 30 °C for anthracites and coke; and for the last ones less temperature will be required at the collector to avoid condensations.




Page 31

2.1.2 Líquids The most common is the fuel oil. They are hydrocarbons of very close composition for diesel, light and heavy oil. % C ≈ 86, % H ≈ 12, % S ≈ 2. The densities ρ have important variations which not differ a lot with the temperature: Deisel oil ≈ 0.88 g/cm3 Light oil ≈ 0.9 g/cm3 Heavy oil ≈ 0.95 g/cm3 However the heat value of different types can be related according their density through the following formula: LHV (Kcal/kg) = [12400 – (2100* ρ2)] – 800 Regarding the minimum air, even if it is too close to the 11 Nm3/kg fuel it can be expressed as: Amin (Nm3/kg fuel)

= 10.8*(LHV/9750) = 0.85*(LHV/1000) + 2

V0R (Nm3/kg fuel)

= 1.11*(LHV/1000); with an approximation of 14 % CO2 + SO2

The ignition viscosities and temperatures are very bound: Viscosity (Cst)2

Ignition T °C

Deisel oil

< 10

< 30

Light oil

10 – 60

30 – 50

Medium oil

60 – 150

50 – 60

Heavy oil

> 150

> 60

Being very important the viscosity variation with the temperature towards a good atomization and therefore a good combustion. Then, if a heavy oil at 120 °C has a viscosity (3°E ≈ 23 Cst) in the range of the light oils, which allow to be transported and injected at a working pressure of 35 – 40 Kg/cm 2 will generate, by an adequate atomization, an adequate flame for the kiln. Other characteristic of other fuel kinds is the dew point of their combustion gases very close to 50 °C. The specific heat under a constant pressure Cp is approximately: Cp = (0.403 + 0.008*T)*(1/ ρ0.5), (T = Temperature in °C and ρ = density in g/cm 3 at 15 °C)

2

Cst: centi-stockes




Page 32

2.1.3 Gases The most used are the natural gases, which in this case vary highly in composition and properties. Regarding the composition this can be expressed in: %CH4 (Methane) (80 % – 95 %) %C2H6 (Ethane) (2 % - 10 %) %C3H8 (Propane) (0 % – 3 %) %C4H10 (Butane) (0 % - 2 %) Being their heat values, expressed in Kcal/Nm 3 gas, in the order of 7000 – 10,000 Kcal/Nm 3 (Natural Gas 8450 Kcal/Nm3). The minimum air is in a range of 10 – 14 Nm 3/Kg gas, being able to be calculated in function of its low heat value in the following way: Amin = 1.09*LHV/1000 + 0.25

Nm3/kg gas

And its volume of stoichiometric combustion gases V0R = 1.14*LHV/1000 + 0.25 Nm3/kg gas; (With a CO2 percentage in the smoke gases in the order of 10 % and a approximate dew point of 60 °C.) Finally and as a simple rule (not very precise), valid for solids, liquids and gases, the stoichiometric minimum air and combustion gases for each 1000 Kcal is: Amin/1000 Kcal = 1.4 Kg/Kg fuel or 1.1 Nm3/Kg fuel V0R/1000 Kcal = 1.5 Kg/Kg fuel or 1.2 Nm3/Kg fuel Natural gas density at 20°C and 1 atm is 0.694 g/cm 3 2.1.4 Alternative Fuels The alternative fuels judgment entails not only thecnological and economical aspects, but also the enviromentals (Legislation) and socials (Communities), as it will be seen in part 6. a) Liquid Alternative Fuels The main points to consider are: 



Heat Value o

Motor oils ≈ 9600 Kcal/kg

o

Dissolvents ≈ 9000 Kcal/kg

Viscosity: must be below 300 cst, depending on the transport method and nebulization (Atomization) o

Water percentage: Must not surpass 0.0005 Kg water/MW of burner

o

Particles in suspension.




Page 33

b) Solid Alternative Fuels Size (mm)

Substitution % Max.

Polyolephines (Polyethylene and polypropylene)

<7

20

Synthetic gums polybutadiene)

<7

20

Plastic or expanded polystyrenes

<7

18

Polyvinyl chloride (PVC)

<3

5

Acrylics (Fibers y plastics)

<7

9

Cellulose plastics

<7

5

Wood/paper/cardboard

<7

20

Polyester

<7

15

Phenolic or Epoxy resins

<2

12

Polyamides (Nylon, synthetic wool)

<2

12

Polyurethane

<7

9

< 100

10

-

5

(Poly-isopropene

Triturated tires Tires (Not shredded)

and

LHV (Kcal/kg)

8000 - 9600

3600

8000 - 9600

6700 – 8000

2.2 Flame and burner The clinkering process requires to obtain at some kiln point that temperature at which sintering is obtained, transforms a combustion problem into other one, the flame. It is known, since long time, that even the flame is obtained with only one fuel, really it is not only one, there are 2 flames in this concept. The fuel, once out of the burner, is heated rapidly and adquires its ignition temperature, and the combustion with the available oxigen in this stage begins, which is the primary and transport air (If it is a solid fuel). Later the combustion products surround the fuel decresing the O2 diffusion towards the fuel, therefore delaying the combustion. So that is why there is a zone where the combustion becomes slow and in consequence the generated heat is lowered. Once the particule finds O2 again the combustion is reactivated. That is why the flame has 2 temperature peaks, being visible only the first one.




Page 34

The decrease of the zone saturated with combustion products and therefore the decrease of flame total length will depend on: 1. The burner technology a. Generated turbulence. (by radial air or swirl and its approximate angle of 40°) b. Internal air as 3rd channel. (Recirculation of hot unburned material) c.

Primary air total flow. (Push or impulse)

d. Primary air speed. (Push or impulse) 2. The fuel to be burnt a. Injection speed. (Push or impulse) b. Granulometry (smaller size of atomatization drop or solid particle) c.

Mixture or not of several fuels. (Multiple flame)

3. Free space for the flame a. Kiln diameter b. Free section losses by -

Production level

-

Coating/obstcules

-

Apparent density (Kiln producing clinker or material almost raw)

It will be derived from the flame quality: 1. The clinker quality 2. Volatile drain (sulfur cycle) 3. Refractory duration 4. NOx emissions




Page 35

Towards improving the contact between the oxidant and the fuel, the burners are designed to have parallel and concentric flows. So even 3 air types are defined (radial, axial and internal) each one will be radial, axial or internal regarding each fuel channel. The following picture shows a multichannel burner:

Axial Radial Solid Fuel Central

The flame maximum position will be very influenced by the flow amount of radial air or swirl and by the internal air, as well as the burner position and the fuel kind. This dependence is of the type: Flame Max Position2/Flame Max Position1 = %Radial13/%Radial23 If well the aggressiveness on the refractory is increased as increasing the radial air %. 3rd generation burners = Hollow conic flame The IDF draft displaces the combustión gases, allowing the renovation of the flame air.

Secondary Air.

   

Radial air Fuel Internal air Axial air

The radial air opens the the axial flow in a conic way (Its rosettes must be in the rotating direction)

Secondary air.

Turbulence created by the speeds difference of the primary and secondary air.

The flame opening creates a pressure at the center, reintroducing hot unburned material which stabilize the flame. The slower internal air favors this effect.




Page 36

The turbulence created by the radial air, intensifies the combination between oxigen and fuel, being this turbulence the one which shortens the flame (And decreases the residence time) and intensifies it, favoring the C3S formation and decreasing its crystals size (Better reactivity and grindability) in order to favor an aprubt change during cooling and therefore freezing better the C3S formation reaction. Several concepts are used to calculate the primary air suitability (including transport air) towards the flame stability. Burner push or impulse = Force in Newtons of the total primary air. Is the most common and effective for calculation, as it will be seen later. Impulse I = mass flow (Kg/s) * speed (m/s), [N] Kinetic energy or flame moment = Energy in Joules of the total primary air. Kinetic energy = mass flow (Kg/s) * speed 2 (m/s) [Watts] As it has been mentioned the most effective, is the total primary air impulse or push, to which the one generated by the fuel is added. Next it is shown an example of a burner push calculation with the following characteristics: Burner point

Passing area A m2

Mass flow M Kg/s

Speed V m/s

Impulse I (N)

Axial Air

0.0153

3.5

210

735.8

Radial Air

0.0201

3.91

184

719.5

Interno Air Transport Air Fuel Total I =

1455.3

Conversion formulas of volumetric flow to mass flow: Flow in Nm3/hr: M = (Nm3/hr) * 1/3600 (h/s) *1.293 (Kg/Nm3) result in Kg/s Flows in m3/hr: M = (m3/hr) *

273

 273  Tf 



13.6  760  e

0.0001255H

10333

(Kg/m3) * 1/3600(h/s) result in Kg/s

Speeds V (m/s) = (m3/hr) / (A (m2)* 3600 (s/h)) Impulse (N) = M * V H = Height in m SNMM Tf = fluid temperature in °C LHVpq = Fuel calorific power at burner point TPH = fuel T/h burner output

8,180 13

I

Specific impulse = LHV   TPH/1000  pq

13.7

N/(Gcal/hr)

Kcal/kg Coke ≈ 6 – 8 Coal ≈ 5 – 6 Oil fuel ≈ 4 – 5 Gas < 4




Page 37

According the transport air of the solid fuel, this should not be too high (< 2% of the Amin), decreasing primary air to hold the flame (if dust injection speed is needed to avoid deposits –DUNESin the duct, approximately 23 m/s), not as low, causing PISTON flow with PULSANT flame, which is avoided lowering the injection speed (approximately 20 m/s). The speed which the fuel must be injected is very bound with the transport air amount, in order to avoid causing piston flow. V (m/s) = 6* dust Kg/air Kg + 17 In primary air percentage terms (See calculation Appendix A part 2). The relation between the primary air % and the speed (m/s) in order to get a good combustion, with CO absence is: Vap (m/s) = f*236*%Ap-0.4 ; being f = 1 for old burners and 1.5 for modern ones (working at high pressures (from 200 to 250 mbars)). Therefore each Ap % is associated with a speed for the impulse maintenance (Represented by (Vap) * (% Ap)) which holds the burner flame. If adjusted in an unbalanced way a CO presence will appear: % CO = 2617 * %Ap-2.21 * Vap-1.09 So if Ap% is 10 % with respect to the minimum combustion air, the speed of the primary air should be at least 93 m/s. If a lower speed is used, for example 50 m/s, a CO increase of 0.23% will be obtained. At a greater impulse a greater hold of the burner flame. It is important to remember that the kiln draft and the fuel granulometry displace the flame backwards, being the impulse the one wich holds it, therefore the flame is holded: - At a greater fuel amount - At a greater primary air % If well specific values might not be given for the different airs, since they depend on the burner type, values of 10% of radial air for gas and 30% for coke, with 5% of internal air and the rest of axial air, are very operatives. In kilns with a large diameter the radial level can become 50%. On which it is possible to be specific in its total primary air (expressed respecting the minimum air), not lowering it from the 10%, if well cold air is saved and therefore kiln specific consumption and emissions (Mainly NOx), the combustion is clearly disturbed, even in the low primary air kilns where the manufacturer offers the possibility to have 6% of primary air with respecto to the minimum one. The burner position also is very important, as in its penetration, as well as in its centering with respect to the kiln tube area. a) Penetration. The ideal is that the secondary air gets to the position where the burner point is, with a laminar flow (In order to keep the flame straight). That is why in grates kiln it is only needed a few penetration centimeters, while in the satellites ones it must be inserted from 4 to 5 meters.




Page 38

b) Centering. In grates kilns, because its proximity to the secondary air entrance, it will be seeked the position which, close to be centered, modifies the flame inclination tendency, so in this way the flame will not touch directly the refractory, or in the raw meal layer, remembering that while the secondary air gets hotter the flame tends upwards. If it is a satellites one, it is totally centered or at the most inclined 2% downwards to the material layer.

From the calculation point of view, the kiln balance tool, gives approximate information (With little additional data to the heat balance). The essential flame parameters: length, secondary air maximum position and its level; similarly the expected values for the secondary air, hot stage gas and maximum (with flame or without) temperatures at the calciner are calculated. Also, the tool calculates (see appendixes) the probable profiles for the gas, material and refractory hot side meter by meter, which are important at the time of selecting refractories by their Ts (softening temperature). The calculations are presented in the appendixes. KilnBalance.XLS Sheet: Temperature profiles Calculus sheet example

So the refractory hot side temperature can be calculated in function of the kiln shell temperature.



Calculus sheet example

KilnBalance.XLS


Page 39

Sheet: Hot Side Temperature

And with this, estimating the refractory thickness, estimate the coating thicknes, being very important when rings or thick coatings are presented which disturb the kiln operation. Calculus sheet example

KilnBalance.XLS

Sheet: Brick Thickness




Code DP-03-1

Page 40

2.3 Heat Balance The transmission mechanisms of generated heat by the combustion differ on the installation point which is being considered. Installation Section Transmission medium

Figure

Radiation  It is transmitted by electromagnetic waves generated by high temperatures.  It is transferred from the flame to raw meal, refractory, coating and from them to the free space.  The transmited energy is function of the emissivity and of the temperature raised to the fourth power. At higher emissivity higher transference (The flame is more radiant) so the radiation order is: Coke (≈1) Coal (≈ 0.95) Fuel oil (≈ 0.8) Gas (≈ 0.3)

Calc

Tower

Cool.

√

√

X

X

√

√

√

√

Flam e

Q = 4.88x10-8 *  * (T14 - T24) (Kcal/hm2)  = emitter body emissivity (01) T1 = emitter body temperature (°K) T2 = receiver body temperature (°K) °K = °C + 273

Convection  Convection flows by temperature difference.  Heat transferece from gas to the periphery.

Kiln

Gas

Q= hconv * (T1-T2) Kcal/hrm2 hconv = heat transference coefficient by convection Kcal/h m2 °C T1 = hot body temperature °C T2 = cold body temperature °C




Conduction  Contact between masses different temperatures.

Code DP-03-1

√

X

Page 41

X

X

at

 Transference from the hot side to the material (cooling the refractory), or from the material to the kiln shell or the the satellites through the refractory. Q = K * ΔT

With these mechanisms, their heat transmission models and approximations which facilitate the obtention of new needed data the heat balance is constituted (See appendix B part 2). Next the KilnBalance.xls tool screens are presented in the section where the data is captured (Heat Balance Sheet) and the distribution of heat inputs and outputs in the system is obtained. The second sheet (Balance Results) shows specific data of each section such as material and heat flows.

Calculus Sheet Example

KilnBalance.XLS

Sheet: Kiln Balance



Calculus Sheet Example

KilnBalance.XLS


Page 42

Sheet: Balance Results

The main balance objectives are listed next: 

Calculating the fuel consumption. (The fuel data does not harm the balance beyond its latent heat and therefore in 2 Kcal/kg)



Identifying opportunity areas mainly based in benchmarkings (diagnoses).



Being able to install it in the distributed control and therefore getting the specific kiln consumption online, with the purpose of verifying scales.



Supporting the people in charge of production to avoid the outphasing in the real inventories, mainly supporting the calculation for operations under low productions.



Getting fast flow levels and data in general which are need by the Engineering department.

Insert sheet with the diagnosis (Real vs Ideal)




Page 43

Part 3- Operation and Kiln Control at Regime The objective of this section is to show the concepts and calculations for a kiln control and also a recovery after a start up.

3.1 Kiln parts and their functions Before studying it, first on a given installation it is needed to raise, the “Bottle Necks” and for that first it is needed to describe what is being asked to each part.

3.1.1 Cyclones Tower The cyclones tower must reach and effective gas –material heat exchange. Therefore the gases temperature at the cold stages must be coherent with the number of total stages. The temperature demand for the hot stage and the draft level (excess oxigen), is described in the following empiric relation: Tgas (cold stage)  [0.66-(Number of stages*0.06)]*(1+O2exit/100)*Tgas (hot stage)

The heat recovery by the cyclone tower will depend on the cyclones and gases chambers sizing of the mentioned stages, and also if they have or not a vortex (or inmension tubes) and air lock flap valves which avoid the short circuit without preventing the constant raw meal flow. T°2

T° 1

Using the gases temperatures along the tower and calculating their differences anomalies or exchange shortage can be detected; so for hot stages, usually without vortex, difference in the order of 100° are found, while in the cold stages, prepared for the maximum separation efficiency, values of 180 to 250°C are found. Therefore, analizing the difference for the entire tower, this is between hot stage and cold ones, the heat recovery efficiency can be calculated throuhg the next expression proposed:

t  t f  % Re C  1.6  c   N 1  145 % Re C Normal  N

0.75

ReC  %ReC  %ReC Normal  5, Excellent  If , ReC is 0, Good  5, Bad 

where, tc y tf are the hot and cold stages temperatures (°C) (in the case of having several cold stages use the gases temperatures average), and N is the number of stages in the tower.



LE

LV


Page 44

The cold stage vortex, must reach a lower point than the gas-dust entrance to the cyclone:

LV  1.1 a 1.5  LE

On the other hand, the gas-dust entrance area or surface (m 2) to a cyclone is calculated as follows: 2

Entrance Surface  LE  PE  2  PE 

Flow Speed

Flow   1    PE     Speed 15  3600  2 15  3600  Normal design speed 15 m/s

1

2

m3 hr LE  2  PE

 in

d body  4  PE d vortex  0.5  d body

Lvortex  LE ( normal LV  1.1 to 1.5  LE , for cold stages) Lvortex  LE ( normal LV   0.5 to 0.7   LE , for hot stage)

The tower surface in function of the wished heat recovery can be calculated in an empiric way as:

S (m 2 )  2.6 10  5

   50  Tpd     Tgas hot stage  Tgas cold stage  1   O 2 21   

The sections where the exchange is more effective are the transitions from one stage to another (with speed 18 m/s), more than the cyclones bodies, so the point where the material fall in these gases ducts should be the lower possible and must be taken care in order to take the most advantage of the distance of these gas-dust exchange ducts, and taking care if the cyclone has an immersion tube (Vortex) or not. The load lost at the tower, measured as the difference of pressure between one stage entrance and the exit must be rational (in cold stages ±180mmca and in cold stages ±100mmca), pointing a normal sections sizing. From these the electrical consumption of the IDF will depend. Which must have the capacity to move:

   50  Tpd    1   O2     21 

 Nm

3

/ h  , for white clinker multiply by 1.54




Page 45

The dust level or percentage which come out with the kiln gases, which will depend on the pressure level generated at the cold stages exit, will be reflected as in the heat consumption as well as in the possible distorsion in the clinker desired modules, with respect to the tower feeding modules. The contribution of air by the tower cleaning air blasters does not represent a considerable volume (56 cannons of 100l discharging every 20 seconds just represent 18 m 3/h)

3.1.2 Calciner or fuel burning at the smoke chamber. a) Calciner The calciner existence or not and its disposition determine the preheater tower scheme, see page 45. In the calciner are essentials the following considerations: i.

The transversal section, since it will determine the fuel speed ascent (from 20 to 30 m/s), see picture in page 46.

ii.

The heigth along with the transversal section will determine the fuel residence time, it must be coherent with the fuel percentage which for design it has to be burnt in the calciner with little CO amount (< 300 PPM). This residence time is in the order of 1.5 to 3 seconds.

iii.

The relation between the velocity seal areas (a) or gases narrowing originated in the kiln and the tertiary duct (b), will be the one which delimits the air pass through these 2 routes (under the supposition of having the tertiary air 100% open). As greater (a) area vs. (b) area there will be less air tendency through tertiary and greater through the kiln. b In the previous case, along with a high calciner fuel % burning, the necessary air for this combustion will have to be provided through the kiln, which would affect the flame stretching and cooling it a lot, with a effective kiln meters loss. Therefore, it is very important to take care of the relation between the (a) and (b) areas to assure venting through the tertiary duct and which will be regulated through its gate. Normally they are designed to allow burning the 60% of the fuel even though they are often taken to other level and the tertiary air can be closed so the kiln tube is obligated to have oxigen levels of 1 % – 5 %. From the result of this balance, a decarbonation level can be consented which will be in syntony with the hot stage temperature. The decarbonation percentage can be estimated with the following formula:

% decarbonation  0.38  Tgas ( hot stage )  650




 Flow after the last fuel burning:

%  1.25   Calciner   100  ; [Nm3/Kg ck] Q2  O 1  2 Calciner Exit 21

2

 Q 2  TPD  1000   f  24  3600   V2   A2

; [m/s] V2 between 20 – 25 m/s

Where:

f  Factor 4

Nm 3 273   e 0.0001255H 3 273  T m

T = TCalciner; To be used in the f calculation.

Tertiary Air: Speed 25 – 30 m/s.

Q  Amin Calc  λ Calc

%  CEH   Calc  100    PCI Calc

; [Nm3/Kg ck]

λ Calc  1  0.09  O 2 Calc 3

Amin = Formulas in function of the fuel type; [Nm3/Kg Comb]

 Speed seal: Speed 25 – 30 m/s. Q = Q1

Smokes chamber: Speed 16 – 18 m/s. Q = Q1

 Flow after the main burner:

%  1.25   main   100  ; [Nm /Kg ck] Q1  O 2 Tube 1 21 3

11

 Q1  TPD  1000   f  24  3600  ;   V1  A1

[m/s] V2 between 12 – 14 m/s

TBack End = 1.25 THot Stage ; To be used in the f calculation.

Page 46




Page 47




Page 48

(continue iii.) The result of these formulas can differ from different installations, so it depends, besides on the ubication of the gas measurement thermocouple, on the material discharge height of the previous stage to the calciner and on the difficulty which the narrowing (a) offers to the kiln direct raw meal pass or to its ascent from the calciner to the hot stage.

iv.

If the calciner does not have a tertiary air, the fuel level which usually can be burnt will not exceed the 20% of the total, obtaining hot stage temperatures in the order of 830 – 840 °C with CO absence (< 300 ppm). If there is tertiary air the obtained temperatures will reach 910 °C.




Code DP-03-1

Page 49

v.

Even though it is obvious, it is important to mention that the production increase, by preparing more the raw meal which enters the kiln (high hot stage temperature), decreases the generated heat recovery since there are more losses by the escape gas, except additional stages are incorporated or the gas-dust transfer areas are increased.

vi.

The oxigen in excess at the calciners exit is between 2% - 3 %.

vii.

The material discharge to the calciner could be above or below or a combination of both, always having present the residence time to capture the heat at the ascent and the possible last stage short circuit passing directly to the kiln. Normally this discharge is handled in 70% above and 30% below, for the greater impact of the material short circuit.

viii.

The fuel vs. temperature regulation at the hot stage has to be the one which guarantees a maximum variation of  5°C for the mentioned temperature. For each 10°C increase in the material preparation exigence in the calciner 5 percent points will be decreased to the fuel at the main burner.

% CaloriesMain Burner

 THot Stage = 27    1000

  

6

b) Burning fuels at the smoke chamber. i.

The base is the same as the calciner without tertiary air, differentiating the transversal section and the length with lower residence times at one second. In theses cases usually the fuel for a CO < 300 ppm is not greater of 10 – 15 %. The O2 above the smokes chamber usually is above 3 %.

ii.

In the other hand, the gas temperature level at the hot stage will be limited by the volatile cycles. If the SO3 level in the hot stage material rises from a certain level (normally 6%) there are cyclones clogs risks, so the standard temperature must be lowered. A bad combustion at the calciner or smokes chamber, if not in other points, will cause high volatile cycles, so when decreasing this thermal demand at the hot stage will change the fuel level and therefore there will be a better combustion.

iii.

Whichever the system is, it is recommended several fuel entraces to distribute the heat better in all the calciner section or smokes chamber.




Page 50

3.1.3 Kiln Tube Towards the design and to evaluate our installation as a just kiln tube, the following formulas are presented:

CT  thermal load  (Gcal / hm 2 )  0.23  Tpd 0.2  ( main %) 0.34 where : main %  calories percentage in the main burner. CT  design estimated thermal load

Tpd  production level 

10.28  Di

3.84

0.06  ( main %) 0.72 where : Di is the internal armor diameter

(for white clinker multiply by 0.65)

CR  Specific refractory consumption (gr/ton)  Brick  Concrete  17.5  CT 2.2  100 DC  Normal campaign duration (months)  770  CT  2.64 Note1 : The usual relation Length/Diameter (L/D i ) is of 14 to 17 for kilns with preheater without calciner and of 10 to 15 with calciner.




Page 51

a) Main burner Regarding the fuel amount, this, if well is determined by the design (with or without calciner) and the raw meal burnability (as better the raw meal burnability, less necessity of main burner), should not generate a thermal load superior to 6 Gcal/h m2. Regarding the total air at the kiln tube, the excess O2 should not surpass the 4% because of everything seen in the calciner section 3.1.2-iii, however, it will not be very adjusted since it would generate reducing atmosphere (CO presence > 500 ppm) even when the oxigen is not close to zero, which can cause refractory problems, sulfur draining and lowering the production. A bad combustion might be due for an excess air as well as excess fuel, being possible in both cases to choke the flame. For this oxigen measurement the probe must be inside the tube, being able to pass the kiln seal, if not the seal must be in good shape and the kiln position (up/down) becomes critical and must be taken into consideration as a part of the measurement reliability. Assuring a good amount of primary air becomes more important as higher the excess oxigen which is desired to work in the kiln tube, since an excess air extends the flame, being a primary air increase the one which holds it. So in order to keep the flame in a specific position after increasing the kiln draft and fuel, must be accompanied with a primary air increase which keeps the stoichiometric primary percentage (minimum air % for combustion), being able to calculate the needed primary levels in function of the tower exit pressure (as a draft level indicator) once the adjustment values for the a and b coefficients have been found.

Primary Air  a  dp btower exit Usually the primary air is 10% above its respective stoichiometric air. Inferior values to the 10% increase the process thermal efficiency but raise the already mentioned risks, being necessary to balance the effects of this primary at each case. Finally regarding the proportion between primary airs, Axial (jet) and Radial (swirl), it is recommended little radial air if it is burnt with low O2 and vice versa, this is, more radial if it has a high excess air which can enlarge the flame. In general words (as seen in part 2 –Combustion-), a 70/30 distribution for Axial/Radial, is very recommended by burners manufacturers, who do not recommend to surpass the 40% of radial air, since increases the refractory risk and clinker quality, and also the SO3 drain. An exaggerated intensity difficults the coating formation since it melts it. Naturally all of these will be also modified for the raw meal burning easiness or difficulty. Therefore, a soft raw meal, allows having lower oxigens, with low radial (15% - 20%) and not having a flame too intense since it is not required.




Page 52

One way to increase the refractory campaign duration is to distribute the refractory harm through all the campaign, beginning with an axial/radial relation and finishing with other one. For example, if it is started with a 30 % radial (punishing the inferior transition), through time the campaign will be evolving to a lower radial until reaching, at the end of the campaign, a 10%. The same can be stated with the primary air percentage, but however, this never constitutes fixed rules, but, alternatives to situations of little campaing durations. In the other hand, in kilns where the maximum production capacity is already reached, and besides limited, by the air amount which can pass through the satellites or by the main fans capacities, or by the kiln diameter, could obtain a load increase through the liquid oxigen injection (12m3/ton of clinker), by means of a lance just made for that and placed under the burner, which intensifies the flame close to the material and away of the refractory. With this it is obtained to increase the efficiency from 5% to 7% but will not improve substantially the heat specific consumption, and the cost of the product is too expensive. b) Kiln drive. The kiln drive (speed and amperage) must allow reaching the speed required by the process, being this, a function of the raw meal preparation degree at the kiln tube entrance (hot stage temperature) and its burnability, playing an important role the fluxes level. In this way a raw meal: - Too prepared, will advance little in the kiln until reaching the viscosity which requires rotation, therefore it will demand a higher speed. The obstacles level at the kiln interior (coating, rings, etc.) increases this speed necessity as will be seen in the “control variables” section. - Too easy to burn, requires little residence time and therefore higher speed, even more when this burnability is accomplished based on the fluxes levels, on which case a lower speed favors the premature nodulation, with coating and/or rings in the kiln, and even the dangerous clinker balls formation for the burner. In the other hand, the motor has to have enough power to rotate the kiln at minimum RPM’s when it is loaded with weight (Raw meal + irregular coating). c) Draft Factor This factor evaluates the draft level in the kiln tube. Values above 5 indicates a overdraft in the kiln tube, which will be accompanied with gases speed increases worsening the heat transfer and high specific consumptions to compensate the flame cooling, also the production lost by effective tube disminution by flame displacement. The optimal levels are in the order of 4 to 4.5.

FDraft

Tpd CEH main comb. %  1  0.09  O2 Ki ln Tube    1000 100  24  2  Di 4




Page 53

Another way to express it, is directly as gas speed in kiln:

Q Tpd * 1000  3600 24 V (m / s )  (must be lower than 15 m ) s  2  De 4 3 where : Q  kiln tube flow expressed in m and evaluated kg clinker at back end temperatu re. Note : this flow can be found in the Kiln Balance tool.

3.1.4 Coolers The kiln discharge capacity must allow not only the material to go out, but also to allow introducing secondary air to the kiln tube, this secondary air must recover the maximum possible temperature from the clinker which goes out from the kiln, all of these is the coolers job.

3.1.4.1 Satellite (Planetary) Coolers The function of obtaining secondary air with enough amount and heat is compromised mainly in the satellites kilns. In these type of kilns, the lower the satellite number, the greater cross section area and greater production capacity. In this point, a thickness optimization in the refractory at the discharge zone allows balancing duration and production capacity. For the material filling in satellites the following must be fulfilled: The cross section area must be:

Tpd  70  2 N  Sat .  d satéllite 4 And the satellites volume must be:

Tpd 4 1.5 N  Sat .  LSat .  d Sat . And the total cross section area could be calculated from the kiln production:

Atotal

 Tpd   10   1000 

1.5

, from which Tpd can be solved

and where Atotal is the sum of the satellites shell areas in m2.




Page 54

The cooling in the satellites is very conditioned by the already exposed subjects, the relation between material layer and free air cross section, determining the maximum capacity and clinker temperature. The satellites area will determine the clinker and air pass and therefore conditions the satellite thermal load, and also the kiln production level, but not its cooling efficiency, which will depend on its length and on the internals design. In the cases where the satellites become jeopardized in its Thermal Load, they can be injected with water, and when this does not cool their hot part it can be considered the exterior water injection on their hot zone. Also the dust return to the flame will be increased by the reduction of cross section area increasing air speed through it. The normal exit temperature of the planetary cooler, usually surpass 200°C. In kilns with grates coolers the scheme is more complex. 3.1.4.2 Grates Coolers Starting from the concept of recovering the maximum clinker heat to the kiln (Secondary and Tertiary), the position of the tertiary air duct and the air regulation of the first grates section becomes critical, as much in flow as well as in grate design, and also in the clinker layer above them, translating all the effects in flow and residual air temperature and therefore in thermal inefficiency, and as it is common, the false air entrances (which in a cooler are understood as the air which has not have contact with the clinker layer) will increase these inefficiencies. An economic distribution is as following:




Page 55

For cooling the clinker, the grates needs too much air (ex: 2.4 Nm 3/kgclk <> 3 m3/kgclk) with respect to the one demanded by the kiln as secondary and tertiary air, an excedent of residual air is generated (ex: 1.3 Nm3/kgclk) which entails a lot of kcal/kg as losses (ex: 100 kcal/kg). To avoid this an excellent work of the first grates is needed, working with high layers (550 – 600 mmCA equivalent to 55 – 60 cm of layer) to capture the greater possible heat as secondary and tertiary air, and more speed for the last grates or compartments, therefore a smaller material layer (200mmCA equivalent to 20 cm of layer). When concentrating the material layer and the grates entrance, usually it is used a paulatine reduction arrangement for the static grate, where pressures of 100mmCA are obtained superior to the 1 st mobile control grate. The theroical speed of the first grate or control grate can be estimated in the following way: V (cycles

min

)

70  Tpd A   e  P  L  (100  slip)

where : A  Maximum cooler width in meters. P  Maximum operation pressure of the contol compartment (mmCA) L  Advance length of the mobile grate (displacement in meters) Slip  Material compression effect when pushed by the mobile grate, which makes them to be compressed instead of advancing in order to do it later. typical value of 15%  e  apparent density or liter weight (kg/liter)

In this way, for a kiln of 2800 tpd, with A=3.6m, P=450mmCA, L=0.12m,  e =1.25 kg/L, and slip of 15%, represents a control grate speed of 9.5 cycles/min. Being normal advancing values of 10 cycles/min. The next grates to the control one usually have speed increases of 30 and 60% with respect to the first one. Regarding the linear advancing speed of the material in the cooler, this can be calculated in a theorical way with the following expression:

V (m

min

)

0.7  Tpd P  e  A

where : A  maximum cooler wide in meters P  maximum operation pressure in the control compartment (mmCA)  e  apparent density or weight liter (kg/liter) Which for the previous example it is translated in V= 0.97 m/min, which for a grate with a length of 20m, would suppose 19 minutes of residency.




Page 56

Regarding the slopes, the static one usually is placed in 15° and the mobile ones below 10°. For a kiln operation togheter with the cooler, is important to recognize the optimal level of the material layer on the grates, translated as pressure in one or several compartments, this compartments pressure, is result of the clinker layer (consequence of the grates speed and clinker flow) and of the air level which enters the cooler through the fans of each compartment, and also the clinker granulometry, being the dusty clinker (or raw) the one which most difficults the air pass. A disturbance in the grate will interfere in the kiln by flow and by secondary air temperature, ussually the disturbances in the grate are caused by kiln inestabillity. A good tunning of the regulatory pressure loop at the hood with the residual IDF, will help to reestablish the stability by avoiding the cooling stabilization to interfere in the kiln tube. ¿How to know which is the optimal air level in a cooler? As it has been said, the economical and ideal level of operation at a grates cooler is of 1.75 Nm 3/kgClk, of which 1.1 Nm3/kgClk would be portioned between secondary and tertiary, and the rest (0.65 Nm3/kgClk) as residual, however, the previous one is the ideal and therefore should act in this sense considering: 

Secondary air temperature



Tertiary air temperature



Residual air temperature



Clinker temperature at the cooler exit



Each compartment pressure



Plates temperature

The secondary air temperature is an indicator of the heat exchange at the cooler (at least of the part which corresponds to the secondary air), and it can be said: that at higher secondary air temperature, better energy recovery (kcal) and our cooler is closer to the optimal point. In the other hand, and it is important to mention it, the clinker temperature at the cooler exit does not indicate a thermal efficiency, since this can be obtained through the last fans and not having a good heat recovery in the first grates as will be indicated by the secondary air temperature. The clinker heat recovery is given mainly by the layer thickness and the efficiency of the air which passes thorugh this one, that is why, the conventional grates coolers are handled with material layers of ~ 450 mm (Secondary Air T. ~950°C) and in coolers with fix grate technology and fans with higher pressure, with layers of ~ 650mm (Secondary Air T. ~1200°C). The efficiency of the air pass through the material layer on the grates will depend on: the flow and pressure which the fans can be capable to generate, as well as the plates techonoloy which direct more efficiently the air through themselves (with special grooving, valves, etc.) The cooler width is very determinative to obtain a good material layer and avoiding the air to pass, there where less clinker is. This width usually is 77% of the kiln diameter. The static grate % in a cooler usually is in the order of 9% of the total grate surface.




Page 57

Not matter the pressures, the distance between the material layer and cooler roof or arch (“dove chest”) has to be taken care of, which will allow the secondary flow to have a speed of ~ 12 m/s. That is why, always will be necessary to place the tertiary air in the hood, as well as a correct sizing. The set of secondary air, tertiary air and residual air must guarantee the evacuation of all the air injected to the grates, since in a cooler with positive pressure by lack of injected air evacuation, generates a heat atmosphere which prevents the cooling of the clinker and all the mechanical elements of the cooler. The desired is that the grate design allows to obtain the clinker temperature below T amb + 90°C by means of an air provision 1.7 to 1.9 Nm 3/kg Clk compatible with a specific flow by grates square meter of 0.8 a 0.9 Nm 3/s m2. That mentioned flow will be distributed between the different fans, being good to begin with values of 0.8 for the first fans and 1.2 for the last ones, even in some cases, defects in the layer or design force to inverse this efficiency relation, being very important to evaluate the real temperatures of the plates in order not to cool them, keeping them at a normal working temperature of 60 to 80°C. Average values of more than 1.5 Nm 3/s m2 must force to reconsider the design or current grate work. Towards the design the fans will be oversized to generate 2 m 3/s ٠ m2 for the static grate and 1.3 m3/s ٠ m2 for the rest, as well as in pressures, the pressure working at regime will be oversized of a 20 to 30%. As is described in the following table, greater grate areas with respect to the necessary one worsen its efficiency. The plates surface of plant 2 is bigger than the required for production tonnage, in other words, using the design equation proposed of 0.025*TPD (equivalent to 40 tpd/m2plates), in the plant 2 case would be of 83.7 m 2, this is, there are 49.3 m2 more of the recommended for the kiln production level, so this cooler will demand a greater air consumption (kcal/kg and kWh/ton) given the plates surface, causing with that mentioned oversizing greater false air entrances and heat transfer inefficiency to the air, by wrong clinker distribution on it. Kiln

TPD

Nm3/kg clk

Surfplates

Tempclinker

Tempresidual

Nm3/s m2

M2/T

Plant 1

2856

1.72

69.8

125

280

0.81

0.024

Plant 2

3350

2.56

133.7

125

280

0.74

0.039

Plant 3

2700

1.77

65.6

185

385

0.84

0.024

Plant 4

2784

1.86

75.47

200

420

0.80

0.027

Plant 5

1800

2.20

73

130

350

0.63

0.04

Plant 6

1920

2.43

60.5

140

248

0.89

0.031

Plant 7

7100

2.36

200

110

268

0.97

0.028

Data compiled from the plant balances.

On the other hand, in plants 3 and 4 can be observed that even with more adjusted oversizing of plates surface vs. clinker production, high levels of residual temperature are obtained, which shows a thermal inefficiency in the cooler, which forces to review the work distribution of the fans.




Page 58

Regarding the clinker crusher it is recommended the one with rollers vs. the one with hammers, since they have more capacity, higher working temperature, less dust generation, longer duration and work at a lower speed.




As it has to be mentioned, is fundamental that the injected air leaves through channels in the plates and not by defects, this is, by places not designed to canalize the air and where it does not recover heat. In this sense, is fundamental to take care of the false air entrances mainly by the material discharge of the dust hoppers which go through the plates. A high demand of the clinker temperature may cause greater heat losses by residual air with an unfavorable global balance, besides of increasing the milling energy because of colder clinker. That is why a good cooling efficiency must no be confused with a good thermal efficiency.

Yield or Efficiency (%) 

Code DP-03-1

Page 59

at ar

as clk E

p

Clinker

clk S

a

( clkE   a )  ( at   as   ar   p   clk S )



clk E

 a



*100

Replacing the obtained results of the “KilnBalance.xls” tool in the formula of a cooler efficiency or yield percentage we get:

% Ef 

 366  16  (7  38  50  4  24) *100  67.8% (366  16)




Page 60

 clk E was obtained through the formula: Cp * T° (Clinker temp. at the cooler entrance, supossed 1400°C), the CP has been calculated based on the values exposed in the Appendix B part II. Note: the




Page 61

Finally, the exterior surface of the cooler can be estimated (by effects of the heat losses), as:

External Surf. (m 2 )  2  Tpd 0.68

3.1.5 Production level Finishing this descriptive section of the demanded to each installation part, the productive level of the installation will depend on: - The capacity to add air Flow, evaluated as pressure drop at the cyclones tower. P  P2  P1 - The burning degree (fixed by the excess O 2) for that air flow, to allow clinkering can be expressed by the following relation: Burning degree =

 21  O2   100 21

- Therefore a way to express the Gcal/h which has been given for the clinkering would be: P  Burning Degree

- The raw meal burnability, expressed as burning chemical hardness:

D  0.476  LSF  10MS  Therefore the efficiency can be expressed in the following way: c

 21  O 2  d P (Tpd )  a  P b      LSF  10  MS  21   Where : a, b, c y d are adjustment coefficients to find for each installation being of the following order : b  0.7, c  0.3 y d  2.5

Just as an informative help the relations between production and thermal load are proposed next:



Tpd 

10.28  Di

3.84

0.06   %main.

CT (Gcal / hm 2 ) 

0.72

Tpd  1000 24  10  6  2  De 4

CEH main. 

Where : De  effective kiln diameter without considering 2   (cm) the refractory. De  Di 100 main %.  calories % in main burner CEH  kcal/kg in the kiln tube


Page 62




Code DP-03-1

Page 63

3.1.6 Operational Utilization The operational utilization in a kiln can be qualified, given its level, as following:

3.1.7

Eletrical

Power

< 88%

Bad

88 – 90

Regular

90 – 92

Good

92 – 94

Very Good

Consumption

Is very important to verify the electrical power consumption in > 94 Excellent the kiln global system, however, this must be sectioned to racionalize the participation of each part in the kiln specific (kWh/ton), the sections by which the equipments set must be subdivided in the calcination process are shown next.

Normal in

Normal in

kW

KWh/ton

Homogenization and raw meal feeding

0.12·Tpd

2.88

Fuel feeding and primary airs

0.14·Tpd

3.36

Dust Transp. + Filter + VTF

0.18·Tpd

4.32

Clinker Transp.

0.04·Tpd

0.96

Section

Kiln: 

Main motor

0.14·Tpd

3.36



Main IDF

0.44·Tpd

10.56

1.06·Tpd

25.44

Total for satellites kilns 

Residual IDF – in grates

0.05·Tpd

1.2



Fans and grates impulse

0.10·Tpd

2.4



Cooling – Grates filters.

0.10·Tpd

2.4

1.31·Tpd

31.44

Total for kilns with grates cooler

It is very important to have the power installed information of each machine which integrates the kiln installation, so consumed power values which do not coincide can be discriminated.




Page 64

3.2 Control Variables The kiln control, at the margin of the desired raw meal and fuel maintence flows, and the raw meal burnability (modules, mineralogy and granulometry), is based in the rawl meal design and its flow, the heat adjustment and its distrubtion (excess O 2) and the material residence time. For all the previous the points to be immediately adjusted are: 



Heat: o

Raw meal flow (TPH)

o

Total fuel flow (TPH)

o

Draft (rpm IDF or dampers %)

Residence: o

Kiln turn speed (main drive RPM)

Evidently everything is influenced by the desired cooling conditions. The basic instrumentation for the operation adjustment is as follows:

Tower instrumentation

Cooler instrumentation • Plates temperatures.

• Gas and material temperature in all the cyclones and if possible at the smokes chamber and/or back end. • Pressure measurement in all cyclones cones and in all gas exits of stages, mainly the cold and the hot ones, as well as in the smokes chamber and calciner exit.

• Speed of the different grates • Compartments pressure • Flows of each fan • Clinker temp. at the cooler exit and secondary, tertiary and residual airs temp.

• O2 and CO analyzer after each combustion point. NOx analyzer, at least after the last combustion point.

• Clinker crusher temperature. Residual IDF’s draft indication, could be variable (rpm) or dampers (%).

• IDF’s draft indication, could be variable speed (rpm) or dampers (%).

• Pressure differential at residual filter or signals at gravel and electrostatic filter.

• Raw meal and fuel flows indication. • Tertiary air gate % indication.

Kiln instrumentation • Kiln displacement • Kiln speed and amperage. • Shell temperature continuous scanning. • Burner signals: Fuel, Primary air % (axial, radial).




Page 65

The kiln control consists in answering the following questions: 3.2.1 ¿What speed does the kiln must have? The Sullivan equation, establishes for the residence time the following expression:

t r [min] 

1.77  L  θ  F P  De  N

Where, L = Kiln length from the burner point to the entrance cone (meters) N = Kiln speed (RPM) θ = Natural application slope of material, Clinker = 40 °, Raw Meal at 800°C = 6 °, being in the states set inside the kiln in the order of θ = 15°. De = Kiln effective diameter – bricks interior – (meters)

De  Di -

2   (cm) , where, 100

 = Bricks thickness in cm.

and Di is the shell interior diameter. P = Kiln slope in degrees P (Degrees) = ARCTAN (p (%) / 100) Note: it is observed that the t r does not depend on the production level.

Kiln Diameters

Inclination %

Φ (m)

p (%)

< 2.8

5

2.8 – 3.4

4

> 3.4

3

F = Obstacles factor or coatinging factor: 1 1.5 2

Completely coating free Very coated Ringed

F  1  0.01  ecrust , thickness “e” in cm.




Page 66

The filling degree (%) must not surpass the 13%, in order not to generate uncooked nucleus, even it depends on the kiln slope.

Re commended Filling Degree (%)  18  2  p(%)

  Tpd Filling Degree (%)  3.2   3   De  N  p (%)  Equaling both equations and considering two kilns where just the diameter differs, it can be found why they have very different speeds, for example: Tpd=2050, p=3%, brick thickness = 20cm, with diameters (Di) H1= 5.05 and H2=4.55, it is obtained that the Speed (H) for the Kiln 1 should be of 1.8 rpm and for the Kiln 2 is of 2.5 rpm. If the same distance between the flame and the load can be kept. With all the previous information, the residence time expression is reduced to:

tr 

6.9  L  F P  De  N

; [1]

As there is a kiln part which is traversed by the raw meal without the necessity of turning, until its transformation into liquid phase makes it viscous (θ raises), is very important not to use a faster speed than the required one, since in that case, with just 0.1 RPM faster, will go from clinkering to not doing so (increasing the free lime). Therefore, the residence time which is needed by the material which enters to the kiln tube, with its preparation degree Gp, for clinkering it until a determined level G clk, will be dependent of these other concepts, and can be evaluated with the following expression:

G  t r (min)  1.2   P   100 

6

 GClk

; [2]

Where:

G P  Preparation Degree (%)  T

GClk 

minimum for clinkering

Tmat Hot Stage Tminimum for clinkering

 100 ,

=1450 Gray Clinker and 1500 White Clinker

C3S Free Lime is the clinkering degree (100=1) = relation between corrected Bogue’s C 3S C3S Free Lime  0

discounting the free lime and considerating it as zero.




Code DP-03-1

Page 67

With both residence time equations [1] and [2], equaling them and solving for the speed, it is gotten, for =15°, the following expression of speed necessity:

L  F  Tminimum for clinkering  N (rpm)  5.7    De    THot Stage Material 

6



C3S CL0 C3S CL

Thus, for example, a effective kiln length, from the burner point to the entrance cone, of 63 m, with an internal diameter of 4.55m and a brick thickness of 20 cm, with a slope of 3% ( =1.72°), without obstructions (30cm coating  F=1.3) and with a raw meal at hot stage has a temperature of 810°C, with a Bogue’s corrected Potencial C3S (CL=0) of 64 and a wished free lime of 1.5 with C 3S=58.7, and with a minimum clinkering temperature of 1433°C, needs a speed of:

N ( rpm)  5.7 

63  1.3 1433   4.15  1.72  810 

6

 64     2.3 ; t r  34 min  58.7 

while with calciner could reach 870°C (hot stage material), will need:

N (rpm)  3.6 ; t r  21 min If from the speed expression, the C3S is cleared, the quality vs. Kiln speed will arise:

C3SC L

L  F  Tminimum for clinkering   5.7    De    THot Stage Material 

6



C3S CL  0 N

and for example, with the first case if the speed would be increased in 0.2 rpm and therefore N=2.5 rpm, will be obtained: C3SCL = 54.6, instead of 58.7, result of a free lime greater than the 1.5 desired value because the faster kiln speed. All these proposed equations, suppose a certain clinkering degree, if not, this is, after a stop the average angle is not in the order of 15°, but it would be of 6° and also the material inside the kiln would not correspond to a hot stage material temeprature of 810°C but of 700°C, because of all these the required speed would be:

63  1.3 1433  N ( rpm )  5.7   4.15  1.72  700 

6

6  64    0.6rpm   58.7  15

Also, the resident material can be retained (kept) in the tube which is calculated as following:

Tons Resident 

Filling deg ree(%)    De2  1.1  L 100 4

where 1.1 is the average apparent density of the material along the tube.




Page 68

The kiln has symptons of going at an excessive speed when its behavior is instable, any disturbance (smokes chamber cleaning, etc.) affects it, so the kiln cools. Also the flame becomes closer to the burner, when decreasing the free section due to that less prepared raw meal and therefore presents less apparent density (occupies more space inside the kiln). On the contrary, low speeds with respect to the ones required are manifested in the kiln with flames too streched causing premature clinkering, with coatings and/or rings. Therefore, if well the filling degree must be taken care of, being this parameter which consider a certain Raw Meal/Speed synchronism, also it is true that the speed evaluated is the external one and will differ of the real one in the kiln interior depending on the raw meal preparation degree, and the obstacles height, that is why it is recommended not to bind the exterior speed with the level of raw meal fed since it does not depend mainly on the amount. Because of the previous, it is recommended to set the speeds individually (manual), mainly in kiln recoveries, for later, if it is wished calculating them by synchronism; but always deciding about the speed and with that calculating raw meal feeding, and not otherwise. There are cases on which the primordial conduct is about the kiln speed, following 2 cases are mentioned: 1st case: in order to solve a problem, as the adherences in the entrances of satellites kilns. In these cases it is essential lowering the speed momentarily; either by overheating or by overspeed, so this material which is trying to stick can be retained, cool the entrances zone, lowering the fuel so the flame moves away and the cooling zone is enlarged (letting to rise the O 2 = flame cooling) and adding or eliminating raw meal in function of the causing reason of the product stickiness. Once the entrances have been opened, the speed is adjusted to the necessary one.

2nd case: in order to solve problems of rear rings, the speed must be increased and with it, the raw meal will pass the obstacle but will advance too much kiln, leaving little clinkering length. In other words, the need of speed by rear rings is due mainly to a heat adjustment and speed problem, which ends in a hot kiln and little clinkering, and the difficulty to introduce air by the transversal aerea descrease.

Finally and as an illustrative way, the timing of each stage for kilns with or without calciner are given: Calciner Calciner

Time in Minutes

Tower decarb. %

Calcination Zone

Transition Zone

Sintering Zone

Total

With

95%

2

15

12

29

Without

40%

28

5

10

43

The low speed that a kiln needs without coating coming from a refractory repair, generates little turn and therefore difficults the “encounters” delay for the reactions, so it is important to coat as soon as possible protecting the bricks and favoring these encounters.




Page 69

3.2.2 ¿What draft level does the kiln must have? As mentioned previously, the differential pressure at the tower indicates or is related with air flow which is introduced to the system, considering constant raw meal flow, which affects the tower pressure levels. The measurement point of the lower part pressure must be above the habitual coating zone. This differential pressure can be related with kiln production as with introduced air to the system, which exhausted as an excess O2 %, gives a Kcal/hr amount, capable of clinkering certain TPH level. Therefore, if it is considered as a flow aproximation:

Q( Nm

3

h

escape gas) 

1.25 1000  Tpd  O 24 ; and adjusting to:   k  p, with p  dp2 - dp1 1- 2 21 density   (kg Nm 3 )  1.4  0.005  O2

Where: dp2 = Pressure at the tower higher point, ususally cold stages exit. dp1 = Pressure at the tower lower point. In satellites or grates kiln without tertiary air, the smokes chamber is used. In kilns with tertiary air the calciner pressure above the tertiary entrances is used. Anyhow, it will be placed above the coatinging area. In the other hand the efficiency can be expressed as: P (TPD) = f (Δ P, O2, Hardness), if the load losses at the stages are in the order of 110 mmca and for the cold stages of 220mmca, the normal as a load loss at the tower will be in the order of:

P  110   N stages  1  220 and if the pressure at the smokes chamber (dp 1) is in the order of 80-100 mmca, the pressure at the cold stage exit (dp2) will be of:

dp towe exit   P  100 mmca

If the dp1 rises, but the dp2 does not do it proportionally, the obstructions are increased in some section below point 1; and in this case ΔP decreases and therefore the production capacity, since it will force lowering the fuel flow to keep the excess oxigen.




Page 70

If the dp2 rises, but dp1 does not do it proportionally, caused by a draft increase, and with this ΔP increases, the production capacity will rise. But there is a limit, if the draft is continuously increased, the cold stages recollection efficiency decreases and the dust which is being taken out is increased, diminishing the effective raw meal which enters the kiln tube, and speeding up the gas at the tower difficulting the heat exchange to the material, trying to pull the flame, displacing the heat backwards (high temperature at cold stages) and decreasing it, except it is compensated with primary air, kiln effective meters, and therefore a production descent. The optimal draft must be found for the maximum kiln capacity, and fixing it as a concrete value of the pressure at the cold stages exit and it must be kept, compensanting the variation caused by other inline IDF’s (raw meal mills or coke IDF on gases line), something that is relatively easy, if the pressure at the kiln IDF exit is measured continuously, and is forced, through movements over a mix aspirator, to keep a certain range through a regulator loop. The disturbances that over the kiln “pull” cause the stops and starts of these mills are minimized. In kilns with calciner and tertiary air, the set of the last one must secure an adequate gas speed through the kiln tube, since the heat exchange will depend on the relative speed between gas and material. Therefore finding the regime dp for the kiln, for a O 2 level, the draft movements for other production levels are simple:

P1 (TPD) P2(TPD)

 O2 constant

dp 1 Draft1  , Draft = IDF RPM or Gate %. dp 2 Draft 2

Thus, if a level production is desired and it is in other one, the movement would be:

Draft New 

dp New  Draft Actual dp Actual

Finally, the fan power consumption will be in the same proportion.




Page 71

3.2.3 ¿What fuel level is the required one? Given a draft level (Air flow), and therefore a P, the fuel level will be given by the desired oxigen level, exhausting the air through combustion. Naturally this excess oxigen is referred exiting the calciner, usually at the tower exit, and therefore its result gives the total fuel and not the individual ones (burner + calciner). So the next statement will be the the oxigen level at the kiln tube and in total at the tower exit. The seeked oxigen levels for both cases are those which assure CO close to zero (<300 PPM at the tower exit and < 500 PPM at the kiln tube). At the exit tower the false air entrances must be taken into consideration, which must not raise the O2 more than 1 percent after the last burning fuel point. In kilns with calciner values of 4 % – 5 % at the exit are usually used (if the raw meal transport is neumatic it is incorporated 1% to the O 2) and of 2 % - 3 % in the kiln tube burning coke (with fuel oil the value could be lower). In the case of calciners are considered good values of 3 % - 4 % at the exit and 2 % - 3 % in the tube, since the main false air entrance, the kiln seal, does not affect too much, because of the posterior combustion. The O2 level can be adjusted by 2 ways: a) The fuel movement for O2 adjustment can be expressed as the following simple way:

TPH fuelCalculated 

TPH fuelCurrent , O2 = O2 standard – O2 current. 1  0.009 O 2

b) The draft movement for O2 adjustment can be expressed in a similar way:

Draft Calculated  Draft Current .(1  0.009 O 2 ) With all the already exposed, the kiln capacity can be expressed as follows (with empiric coefficients): 0.3   21  O2   P (TPD )  f  P 0.7 ,     21   

The fuel distribution will be made in function of: the installation, the desired O2 levels, the desired thermal load in the tube (in kilns with calciner and tertiary air approximately 4 Gcal/hr m2) and volatile evaporation problem, essentially sulfur, as it has already discussed at the beginning of this chapter.




Page 72

There is an important consideration to be fulfilled when it is operated with a low production: “For clinkering it is necessary a zone with enough temperature, but with a low production, this zone must be displaced backwards, since if not, the kiln is not shortened naturally, so it will ask for a greater raw meal feeding”. Therefore, with a low production, the excess O2 level must be greater in the kiln tube for the case of kilns without tertiary air and of less raw meal preparation for kilns with calicner (lower hot stage temperature), see section 4.2. Finally and trying to control the O2, the response time of O2 and CO response time by fuel movementes depends on the gas speed, so in the starting up conditions the “final” response will be seen slowly by the low gas speed, being able to take more time in these cases, compared with the time under regime conditions (normal kiln capacity around its nominal).

3.2.4 ¿What feeding level does it admit? Previously it has been seen that a certain introduced air flow, characterized by the differential pressure and exhausted by the combustion at a certain oxigen degree, provides a heat which allow negiotating a certain production level: 0.3   0.7  21  O2    P(TPD)  f P ,     21    

But will also depend on the raw meal hardness (its burnability). Therefore for a D1 hardness expressed as: D1 = LSF + 10 MS (Without considering minerology and granulometry) The kiln capacity can be expressed as:

P (Tpd )  a  P

0.7

 21  O2     21 

0.3

  LSF  10  MS 

 2.5

Being important to obtain the adjustment factor (a), and with it the efficiency variation can be known approximately for each draft change, excess O2 or raw meal hardness. Thus, for a kiln with a 2500 tpd capacity, if the clinker to be obtained is modified, and just that, from LSF=97 (and MS=2.5) to LSF=98, the efficiency loss will be of 2%.




Page 73

Towards the raw meal scale adjustment in order to calculate the clinker production, several considerations must be done: 1. The raw meal scale might not be the only material entrance which exits with the clinker, also it could be the dust which enters with the coke, caused by the fuel milling with kiln drying gases ( dust % 4-12%), depending on the dust concentration in the kiln gas. 2. Some of what enters at the tower superior part will not end entering the kiln tube, since it will exit with the escape gas caused by the higher or lower cold stages inefficiency, and very dependent on the draft level (dust recovery levels of 10% are habituals), and where does the recovered dust go (to the kiln or to the homogenization silo). 3. The weight relation (raw meal/clinker), will depend also on the carbonates level, humidities (free and/or combined), etc., sumarized by the lost on ignition. 4. Others: alkali by-pass, oxidants compounds, the recovered dust in the the filter returns to the tower without passing through the scale, etc. For all the previous, what it could be a weights relations raw meal/clinker in the order of 1.56 by a normal lost on ignition of 34%, will be transformed in the scale into a different relation, so if this physical clinker weighing is evaluated it becomes just a calculation factor “Scale Indication – clinker Tpd”, not confusing when declaring the raw meal consumptions, and from this to raw materials consumptions. Finally, and towards the raw meal flow control and possible decalibrations, the gas temepratures level of the cold stages helps a lot to discriminate scale situations. Therefore, a high temperature with respect to the habitual one may indicate, if nothing has been touched, a lower raw meal flow with respect to the one indicated by the scale. Likewise it is very important the seals condition, as in the material entrances as well as in the discharge, the first one supposes cold air which cools the zone but mainly asphyxiates the kiln, while the second one cools the flame.

3.3 Control Parameters In the instumentation there are signals to indicate the Operator that the kiln is in conditions for admitting one or other raw meal amount, or in general to diagnose and decide what the kiln “IS ASKING”. These parameters can not be considered individually, but as a whole; to seek the given values logic or ilogic by that mentioned instrumentation, which must be enough and reliable.




Code DP-03-1

Page 74

3.3.1 Torque or rotation moment (motor amperage)

Torque motor  976636  kW

rpm

 kg  m

Torque = f( Shell + Satellites + Coating + Burning + Raw Meal + Mech. Problems)

Variable Shell

Effect on the Torque Is constant (weight) and does not affect in Torque changes

Satellites

Is constant and does not affect in Torque changes excepting in entrances stickiness conditions which decompensate the satellites, by being different the clinker amount in them, in this case the “torque” strip is widened.

Mechanical Problems

For example, kiln contact with the still joint of the tower (entrance seal), by kiln up and lubrication failure generate aditional efforts which are translated in sudden Torque rises, and to be kept later in time.

Coating

The symmetrical coating part does not generate torque movements, but in the contrary, if it is irregular, cyclic rises and drops will appear in the torque, according the coating position.

When the coating falls, the torque rises, generated by the fall effort against the turn, and then lowering when it is incorporated to the raw meal flow, this makes that the present torque, on its average value, an increase for then lowering, consequence of the cooling which this material adds to the flow.

Burning When clinkering the material, it nodulizes, raising with the turn and dropping against the motor, consuming higher amperage. In the extreme case of fusion by overheating, the motor effort effect disappears and the torque suddenly drops. On the contrary if the material does not clinker, this is behaved as liquid keeping the level. Raw Meal

Sube el Par

Baja el Par

The amount of resident raw meal in the kiln affects with its weight.

By all the previous, it is very important the TORQUE – TIME slope determination, at short term in order to discern if the movement the Torque is having is by clinkering evolution or by other circumstances (mechanical problems or coatings falls, mainly).

amp/min :

Heating value

<<<

Coating Falls value

Mechanical Problems <<< value

In the other hand, the torque value makes reference to the kiln condition 30 minutes before (approximately), this is, what is evaluated is the last section (clinkering) so the conditions are being “announced” 20-30 minutes after the raw meal entered the kiln.




Page 75

3.3.2 NOx The NOx is formed at high temperatures so it can be an indicator of the kiln thermal level and therefore, the clinkering degree. The NOx may come from the nitrogene in the fuel or air, because it is very influenced by the excess air level, as will be seen in section 6 “Emissions”. Therefore it is a signal that must be taken care of because it can induce errors. The NOx control for environmental effects will be seen in the emissions section, however, the presence of yellow smokes in the chimney indicates high NOx and therefore a very hot kiln, and not necessarily detectable at the kiln peephole. The delay or inerce in the response is about 10 minutes.

3.3.3 O2 and pressures When the kiln gets cold and the generated clinker results with a low apparent density, this is, it occupies more space, reducing the free transversal area and limiting the secondary air pass, for these reasons the following happen:  The O2 level is decreased.  The pressures rise as at the cold stages as well as at the hood. For the cases of satellites kilns the larger occupied space by the clinker at the cooling section limits the access of suction of the air generated by the kiln tunnel.  The flame becomes instable, as stretching by the air speed at the kiln tube and by the raw meal dust interference, by which the kiln begins to blow causing pressure variations at the smokes chamber. When the kiln is heated the contrary occurs. This support guide to the diagnose has to be questioned, mainly to not be confused with a ring formation, for which, the continuous temperature monitoring at the kiln Shell (scanner) helps to clarify. Also it has to be clarified that the movements had not been due to start ups or stops of inline equipments with the kiln (coke or raw meal mills) since them, or their operation irregularities, might alter the pressures, oxigen levels, etc., in the kiln. A way to insulate the kiln from that mentioned installations is placing a pressure measurement at the kiln IDF exit and fixing a “safe level” for the mentioned pressure and controlling it through a inline IDF, it can be the kiln filter one or other which recollects the hot gases as: the inline raw meal or coke mill.




Page 76

3.3.4 TV camera, zone temperature and hot stage temperature. The flame pyrometry, even it does not give the real reading of this one, it is an excellent measurement of the kiln condition on its final section. To assure that this situation concerns the rest of the kiln, the hot stage gas temperature must be observed and controlled, since assuring a constant raw meal preparation value for the kiln, an excellent kiln control can be kept with the flame pyrometry (Zone temperature). The kiln control base is an excellent control for the raw meal preparation. Note .- The zone temperature may be substituted by the flame luxometry (Measurement of the light intensity in the flame).

3.3.5 Analytical Is the less precise, since depends on the raw meal stability. Free lime - As it depends on the entrance saturation, the LSF must be known and bound (of the clinker) so there will not be an error with the free lime and therefore discerning if this, is caused by the kiln condition or is caused by the high lime saturation in the raw meal. And towards the clinkering degree calculation (GK), here is proposed two ways to quantify it:

with :

 C3 S  a ) GK (%)  1    100 C3 S  

 CaO total  CL  b) GK (%)   100  CaO total  CL E 

C 3 S  evaluated by corrected Bogue For the cases in with th e LSF  100, will be used :    LSF  C 3 S  CaO L  1    CaOTotal    4.07 100     For the cases in with the LSF  100, will be used : C 3 S  CaO L  4.07

with : CL  Analyzed free lim e CL E  Stoichiometric free lim e  CaO Total  2.8  SiO  1.65  Al 2 O 3  0.35  Fe 2 O 3  0 If  0  CL E  0 Volatile – The kiln volatile retention or exit (SO3, K2O, Na2O), and therefore their levels in the clinker, might help to decide if the kiln is being heated or cooled. If it is being heated, these values begin to decrease by volatilization; if it is being cooled, the values tend to rise, but in the case of wrong raw meal burning it does not drain since the volatile just escape if it becomes clinker. The delay time of this control is of approximately 45 minutes to one hour, since the analysis time is added.




Page 77

3.3.6 Exist gas temperature The temperature level lowering or rising at the cold stages exit, may warn that the heat and raw meal are being decompensated, announcing that kiln will end cooling or heating. Naturally this does not apply when draft movements are made intentionally or standards for the calciner or any other movement with flame displacement.

3.3.7 Kiln view The flame view, as by TV as well as in field (kiln tunnel or hood) facilitates the understanding of the kiln situation, and complements the diagnostic.

3.3.8 Weight/Liter It is not a good control parameter since the clinker density depens on its modules, therefore a clinker well clinkered with a higher MS (less fluxes) weights less than other one. Ex. LSF=99

MA=1.7

MS=2.1, a normal value of the P/L =1474 MS=2.4, a normal value of the P/L =1261

...and in both cases it will be clinkered correctly. Therefore the white clinker P/L is lower than the gray one.

3.4 Control There are several ways to regulate a kiln and they are between this 2 regulation extremes: 1. Changing the fuel and draft, keeping the raw meal feeding and kiln speed constants. 2. Chaning the raw meal feeding and kiln speed, keeping fixed the fuel and draft. The control reality makes the operator to change more the raw meal and speed than the fuel and draft, since the second one creates problem in the O2 adjustment level. It is important not to forget that when it is said that the kiln is cold (by the control variables of the previous section), the kiln is not only cold but its tendency is to become colder, in other words, a kiln is cold when it has to be done cold kiln actions, since it can be hot at its final part (clinkering), but its tendency to become colder (the material that comes from the back) will take that in a short time to stop being hot. With this criteria the operator must see not only the current flame zone, but by tendency, how is it going to be in a time period. In the other hand, if well there is a excess O2 in the normal operation, for the combustion and volatile drain, when the kiln gets cold, this level drops and it is when some more fuel has to be introduced, as much as the little O2 reserve allows, so, burning well, the kiln is recovered, even it is assumed in that little time it has been forgotten a perfect sulfur drain.




Code DP-03-1

Page 78

The combinated action of fuel increase (without additional draft), lowers the raw meal and the speed in order to give residence, will allow to recover the desired condition. All of these is relatively easy if the tower is kept normalized, this is, if the kiln has a calciner, keeping a material preparation degree (gas temperature at hot stage as the best signal for this) and if it does not have a calciner, keeping a gas temperature overlevel at the cold stages exit by raw meal decrease, which will be fixed according the cooling degree and will allow to decompensate raw meal and heat, assuming that this thermal inefficiency is very important to recover the kiln. By the mentioned previously, it is very important to have a reliable O2 and CO indication in order not to work blinded, that is why it is recommended to keep analizers in parallel (2 or 3) taking gases from the same point but not from the same probe. The actions by hot kiln must be taken even more dynamically, since they can derive in clinker and coating fusion, ending in refractory fusion, or other previous problems, satellites entrances adherences, cooling zone coating, rings, etc. It must be always present that the kiln control is divided, and it has to be divided, in the tower control and the tube control, and to be very conscious that until the tower does not get stabilized at the desired level, the kiln tube can not be controlled, since with material of different preparation already placed in different kiln places will force to a very different operation, and can not operate well in contrary situations. Also, very different actions in a short time, wil affect the tower instability, except if there is a calciner, where benefits will appear. Draft actions have not been mentioned, excepting very aggressive cases, it is best not touching it, because of its flame length and O2 effect, as well as in the gas and raw meal speeds at the tower, and the level of not recovered dust at the cold stages will vary the effective material flow to the kiln tube. The draft decreases raise the possibility of a certain raw meal by-pass at the hot stage which comes from the previous stage. Summarizing, a model for the calculation of stepped adjustments is proposed next:

Draft New 

dp Exit ( desired ) dp Exit (current ) 0.7

Tph Raw Meal  H New

 dp (desired )   21 - O 2 Desired   H Current . Exit  .   dp Exit (current )   21 - O 2 Current 




Code DP-03-1

Page 79

And to obtain that O2 Desired, first the O2 level is calculated (hypothetical) which will be obtained when the current draft (dp) is modified by the dp (desired):

 dp Exit ( Current )  O2 Hypothetical  21   21  O2Current    dp Exit ( Desired )   And therefore, the fuel (tph) to exhaust it until O2Desired:

Fuel new  Fuel New

 21  O2 Desired     21  O2 Hypothetical 

0.8

And finally the kiln speed, supposing it will not vary in the raw meal preparation degree:

N ( rpm) New  N ( rpm) Current 

C 3 S Current F . L. C 3 S Desired F . L.

In case of having controlled the quality to its desired one and therefore C 3 S Desired F . L.  C 3 S Actual F . L. , will not imply the speed modification. If the raw meal preparation degree is not gotten, then: 6

N (rpm) Nuew  N ( rpm) Current

 Tgas hot stage ( Desired )  C 3 S Current F . L.     Tgas hot stage (Current )  C3 S Desired F . L.

Which represents if the hot stage current temperature is lower of the desired one the other movements will adjust it (or if it has a calciner, this will adjust it) and therefore, the speed will be rised accommodating that better raw meal preparation. The acting order is very important. But this previously exposed model should not be confused, which is for a rise or drop in steps, as a recovery after shut downs or regime lower conditions by different necessities; with the movements by the kiln condition, which will be called ADJUSTMENTS, on which, if the draft must be adjusted, this will be the one to be done. After that it will be the raw meal feeding movement, keeping the tower conditions (if there is a calciner, it must be adjusted). Finally with a small delay in time will be the fuel one and finally the speed adjustment. When these are not adjustments to reach the regime (nominal load) but, by the “kiln condition”, the order changes, must differentiating the adjustments in: TOWER and KILN TUBE. Both adjustments might be found, first the tower conditions must be secured if the kiln tube has to be stabilized. This is specially critic after a shut down. Once started, the 1 st objective is to normalize the tower, “sacrificing” the kiln tube as possible towards the tower benefits. When talking about adjustments by kiln thermal state, it is not talked about its final state or burner zone state, but of the kiln GLOBAL state. How to define a global state, if almost all the control parameter values are just talking about the final kiln section?




Page 80

A control parameter value, for example the Torque, is indicating how it is clinkering, but not how the material which is advancing at the back is doing. Therefore, this signal tendency in time is the one which will indicate it. So, it will be a set between the current signal and the tendency the one which will diagnose, towards adjustments if the kiln is cold, hot or well. The kiln would have a hot kiln torque but a strong tendency of torque decrease, which will conceptualize the kiln from global into cold, so the adjustments will be to heating it up. This is evaluated with rationalized values in the following way:

Global State 

(a  Current State )  (b  Tendency - at the chosen term-) ab

Where: a and b will be the weighing of the current state and its previous evolution. 3 examples are shown next: a) a=1 and b=0, not normal it would imply a kiln which by its instable behavior just interests the current value looking forward the desired value. b) a=0 and b=1, would be more irrational since it will not imply the current state which is equivalent to its desired value, and only interests the current value towards the desired value. c) a=0.5 and b=0.5, a normal weighing will be gotten which will be corrected according the current situation and the tendency.



* For:

Code DP-03-1

Page 81

** Considered time: short, medium, or long term or a combination of them.

V≤Vm; VR=-1

Vm < V < Vs1;


VR 

V  VS 1 VS1  Vm

Vs1 ≤ V ≤ Vs2; VR=0 Vs2 < V < VM_;

VR 

V  VS 2 VM  VS 2

V ≥ VM ; VR=+1

With the result of the global evaluated state for each control parameter and averaging according to their reliability, it is established a definitive global state, which will take to the adjustments of a very hot, hot, little hot, well, little cold, cold very cold kiln, deriving in adjustmens on: 

Fuel



Raw meal



Raw meal and speed



Fuel, raw meal and speed



Fuel, raw meal, speed and draft




Page 82

Always if possible the kiln speed will be kept, understanding that it has been found its optimal point for the raw meal and draft preparation degree, on which there is a good zonification. However, if by variation on the raw meal preparation degree (hot stage temperature) or by coating incorporation to the raw meal (coating fall), the speed should be touched, it will be the minimum essential in order to assure a minimum quality loss. Regarding the fuel, this should not risk the kiln in order to “save” its quality; having a bad raw meal preparation has remedy, melting the refractory, not without stopping. The draft, as it has been mentioned it is preferable not to touch it expecting an urgen necessity (CO presence), since adjusting the kiln to a low draft, will force to a new adjustmen regarding other draft, and stabilizing the tower is the priority and it is difficult. What adjustmens should be chosen? Before answering this question the consequences of each one should be analyzed, knowing that it is always important to think about the heat exchange.

Speed Adjustment

If there is a calciner, the TgasH.S. will be kept constant but with other fuel level.

Raw Meal Feeding Adjustment.

If there is a calciner, the TgasH.S. will be kept constant but with other fuel level.




Page 83

Fuel Adjustment (Without Draft Adjustmen)

Draft Adjustment (without fuel adjustment (O2))

If there is a calciner, the TgasH.S. will be kept constant but with other fuel level. Taking this into consideration, the adjustment or adjustmens more coherent for the kiln situation will be seeked, as long as it is not being conditioned with the presence of a CO level above normal, on which case will be more important to correct the CO presence than any other consideration, but if it can be done trying to make it compatible with the kiln state: 1. If the kiln is hot or very hot: fuel reduction, as the desired O 2 is incremented (O2) to lower the CO:

  1 Fuel New  FuelCurrent    1  0.09  O2  2. If the kiln is not hot or very hot: draft increase, as the desired O 2 is incremented (O2) to lower the CO:

1  0.09  (O2Current  O2 )  Draft New  Draft Current   1  0.09  O2Current  




Code DP-03-1

Page 84

Part 4- Kiln control and operation under special conditions The objective of this section is to show the concepts and calculations for the 2 more significative cases: operations related with a kiln shutdown and working under a forced operation with low production.

4.1 Cooling, heating and kiln start up. When a kiln stops or it is stopped, there are some questions that have to be raised, for which this section pretends to give a “calculated” answer. In order to plan a cooling, first the cooling behavior in time has to be modeled for the different draft conditions. From this study the kiln thermal state forecast will born for any cooling time, not mattering how it is done. Thus, a kiln that is going to be stopped for inventory or it is going to start soon, it requires to keep all possible heat. In the first case, not wasting energy while cooling it before time, and not harming unnecessarily mechanical parts or refractory. In the second case not losing calories that will have to be compensated later with fuel. Doing so, how do we know that the kiln is losing as little heat as possible? When does the primary air can be turned off to protect the burner? When does the kiln need to be rotated? All this questions have an answer when its behavior has been modeled Similarly, if it is needed to cool it as soon as possible in order to immobilize it, how does it have to be cooled without any damage? And when could it be immobilize? The cooling time depends on the urgence, and if this is the situation, the rule of cooling at the rate of 40°C per global hour could be applied, which leads to 0.02*π*D*L in hours, all of these measured at the hot stage. The tool handles the next turn frequency concept and the next immobilization criteria:





Minimum _ Turn  1 of turn each T 145 T minutes 3 where, T  hot stage Tgas or scanner Tmaximun Kiln can be stopped, when : hot stage Tgas  Tambient  40, and Shell kiln Tmax.  130C




Page 85

The available file in the “Repair protocol” application, helps to answer automatically this questions, choosing as a reference the kiln thermal state, the gas temperature of the hot stage, as being an available instrumentation in the distributed control (scadas) and the scanner which, in addition of having other consideration, do not have signals for the calculation. As a special warning it is recommended not to exceed 180°C between the ring temperature and the shell temperature. Calculus sheet example

GENERAL REPAIRS.XLS

Sheet: COOLING

Cooling curve without the main IDF (Induced Draft Fan) use, in this case the only consideration is: to have the dampers opened in order to favor this natural draft.



Calculus sheet example


GENERAL REPAIRS.XLS

Page 86

Sheet: COOLING

Cooling curve using the main IDF with a draft level expressed as an exit depression of the tower.

In addition, at any time during the cooling, it is required to know if it could start immediately or if it requires heating, and in this case, how many hours and the cost (fuel amount). For a kiln that comes from a repair, the heating hours are given by the amount of recfractory placed, if it is brick or concrete, and of the kiln span (diameter and length), and by this simple general form:

Heating hours  a    D  L 0.02 if only bricks were placed (or mostly bricks were placed ) with a   0.03 if a lot of concrete was placed D and L in meters. .... and if the installation is totally new, is the manufacturer who determines it. Finally, the internal inspections of the tube state will or will not corroborate that the plannings made have not harmed the kiln. The time to get inside the kiln depends on the temperature in which the people can work and if the work admits kiln turns due the heat.




Page 87

On the other hand, if it is needed to keep the kiln in a determined thermical level (letting it cooling until a given point and keeping it until the heating permission is given), how much fuel is needed to keep it in that state? To answer this question it is just needed to use the “Heating” file available in the “Repair protocol” application. Calculus sheet example

GENERAL REPAIRS.XLS

Sheet: COOLING

Important: first the installation temperature must be allowed to decrease in order to apply the fuel that will recover the losses; this will avoid keeping the kiln hot in a similar level as the operational one.

Thus, for example, if the kiln has to be stopped and it can not be heated until x hours, how cold would it be until that given hour? How much fuel would be needed to keep it in that level and how much time would be needed to heat it in order to be ready to produce again? In what it respects to the level of raw meal needed when starting up, as it has been told, it will depend on the kiln thermical level by the moment of taking the decision. In this section a calculation of the levels of each control variables is offered (efficiency, speed, draft and fuel) in function of the hot stage temperature without draft (this is the only way in which the hot stage level is not influenced by heat displacements and weakens its correspondence with the real thermical level in the kiln tube). This starting up level is called “establishment level”, because it establishes the safe plataform by which, the kiln, before or after, will grow.




Page 88




Page 89

It is importat to clarify, that when a kiln has just stopped is not good to apply maintance heating, since the kiln is ready to produce and the fuel needed only to keep the losses would not be as exact, deriving in heating something ready. In which it is needed to let the temperature low a little bit (around 600-650°C) and keep it hot over this safe level, waiting the moment to heat it and feed it. This safe level is a level that lets, if it is needed, do a direct start up without jeopardizing the refractory and the mechanical parts due the thermal load. The heating planning implies several questions:  How much fuel is needed to start?  At which refrigertation level of the flame (O2)?  How many hours and which ramp? According to the answers of these questions, a fuel amount will appear for all the heating, and therefore also a fuel level for the final, that can or can not be compatible with the maximum permissible thermal load, for an empty kiln with slow turn. The program proposes, according to the initial and final heating temperature, a refrigeration flame level (O 2 excess) giving the option to plan it by heating ramp, °C/hr, or by hours. Finally it will be the tube thermal state (defined by the scanner approach to the normal operation) the one that will determine if the kiln is or is not ready to receive raw meal. Hurrying the start up might be possible, as long as the fuel fed (in order to increase thermically the kiln and also to clinker the received raw material) is compatible with the thermal load (permissible < 6 Gcal/h m 2). The assistance program for this matter, having the understanding that the more a start up is hurried the less of raw material will be fed.

In the previous chart, the areas under the curve indicate the amount of energy required, so in order to minimize the used energy to heat up the installation, the heating time must be decreased to the exclusively necessary through a good programming. It is necessary to indicate that if a certain heating ramp is not beign watched, but the same fuel level is kept during all the heating process, although it will be ready, the flame will be placed in the same




Page 90

position all time, so it will be always affecting the same zone and will entail to the refractory destruction of that particular zone.




Page 91

So that is why a good ramp (both in fuel level and in O 2 excess level) will assure a progressive displacement of the flame to its maximum point, and a accurate heat distribution to all the installation. The primary air will be according to the fuel level in each period, being especially important in the first moments, having the flame with little fuel and not asphyxiating it. Calculus sheet example

GENERAL REPAIRS.XLS

Sheet: HEATING CURVE




Page 92

4.1.1 Start up. Establisment level and regime recovery Draft percentage (% dp):

 Tgas h. s .  %dp  3.3  10    100  800  %dp Draft : dp (mmca)  dp N  100 -6

3.7

T gas h.s.  hot stage temperature

 20 just after starting up ( without draft ). dp N  pressure , at cold stages exit, under kiln regime or " nominal" conditions, in mmca units.

Fuel percentage (% calories):

%calories  25   %dp Fuel : O2  O2 ( regime " no min al" value) 0.3

Fuel (T/h)  Fuel Normal 

%calories 100

Feed percentage (%Production):

%Production  8.5   %calories  %dp % Pr oduction Raw meal  Raw meal Normal  100 0.25

Speed percentage (% Speed):

% Speed  10 5   %calories  % Speed Speed  Speed Normal  100

3.5

Note: do not start up unless %dp = 30, so Tg hot stage= 570°C, that corroborates the calculation of the minimum starting up Tghot stage. But if it comes from a heating or ignition, a fix %dp = 30 will be applied, and the rest according to this until levels are stabilized and then change the value of %dp to the one if %dp > 30 with a hot kiln.




Page 93

The kiln recovery speed (ramp) until it reaches the production regime will depend on the starting up level and on the kiln state, always watching the mechanical aspects (dilatations, etc.). Here the next ramp is proposed in mmWC/h:

Draft (mmWC )

h



 dp N (at regime )  dpout (at start up) hre cov ery

4 with, h recovery  hours needed for re cov ery  1.2  1012  Tgas h.s .

(fast calculation = 0.6·shutdown hours) Example: dpregime = 650

Tg  680C

45  650  295 100  7 hours

dp Start up  hre cov ery

ramp  (650  295)  50 mmWC h 7 If a calciner is available, the exigency ramp for the hot stage gas will be:

C  Tgas h.s.at regime - Tgas h.s. without calciner  h hre cov ery

The next movements to every DRAFT change are formulated in Part 3 “Kiln at Regime”, of which the next metioned criteria must be taken into consideration:

P1 (TPD) P2(TPD)

 O2 constant

dP1 Draft1  , Draft = RPM FAN or % Damper. dP2 Draft 2

Draft New 

dPNew  Draft Present dPPresent

Verify the ignition resultant scanner, with the normal expected profile, in order to detect possible refractory anomalies or damages.



4.1.2 Start up abacus and kiln recovery


Page 94




Page 95

4.2 Kiln with a low production. By its design principle, a kiln is not desgined to work at a low efficiency respect to its nominal, taking care of the operation under a low regime trying no to harm it. Here the concepts are exposed. The only ways to produce at low regime respect to its nominal, is to move the flame (unbalalncing fuel versus draft), or if there is a calciner, lower the raw meal preparation. If not, the kiln will ask the corresponding raw meal to the installation, or the flame will not be able to clinker. a) with Calciner:

P1

Preparation Decarbonation ends

P2 <<< P1

(Production level)

Tc2 <<< Tc1 (Hot Stage Temp.)

P2

Preparation Decarbonation ends

b) with flame displacement: O 2 (1) Enough T. for clinker Effective Kiln P 2 <<< P O 2 (2) >>> O O 2 (2)

Effective Kiln

1 2 (1)




Page 96

In the (a) case it is possible to get the new production from a kiln (with calciner or precalciner) to a conventional kiln (without calciner), with a better heat specific consumption. In the (b) case it will always have a greater consumption and a greater kiln risk, because the operator, as not seeing the flame (because it has been moved away from the hood) it becomes hotter (where he does not see it). That is why he has to watch that higher O 2 and not the flame, as high or less production is wished, and in the fact that the clinker will be produced within specifications. The kiln regime can not be lower because of the risk mentioned previously, and because in (b) case, the kiln cooling zone could be as long, that is, it will become a clinker reheating zone of clinker already produced, with the risk of fusioning it, due to rock formation (mainly on the dumps: entrances or brim), and refractory fusion. For that reason it is highly recommended not to speed down a kiln that is operating under a low regime, or if it is required, to be as low as possible.

Production Operable Margin

1

8

O2 Excess

A kiln with a low production is more prone to generate undesirable volatile cicles, although of having high oxigen levels, since this do not guarantees the mentioned volatile control.




Page 97

A special case of opeating the kiln under a low regime is the presence of rings, where the kiln control operation becomes more complicated because of the need of speed, due the ring, to the margin of the needed for clinkering. This is agravated by avalanches generated due to raw meal retention / liberation and due to the raw meal spilled from the seal, which makes to raise the draft (lowering even more the production) and complicates even more the operation. When the ring, even big, is irregular, there is a margin to try bringing it down, but for this the speed does not have to be very high, since if it is, the gap where the raw meal is escaping only stays for a little time, this is, as we were talking about the dampers of a dynamic separator. Before declaring that the kiln is spilling from the entrance seal due a ring in a determined meter, it is good to make calculations if possible.

d

H Length L(m)

Inclination (%)

Will spill from the seal if : H  d   sen 0.45  (%)    L

Besides seeking the main causes that produced the ring and correcting them (SO 3 in the system, raw meal hardness that requires high fuel flow with a high sulfur, or low hardness, due low modules accompanied of high flux), in order to try bringing them down while working there are some alternatives to try together: 1. Lowering the sulfur addition switching fuel for 24 hours. 2. Making sure that drains, not to accumulate more, by which there is the need of clinkering. 3. Moving the flame in order to destabilize the ringed zone – thermal clash. 4. Hardening a little bit the raw meal so the lacking of flux will instabilize the ring. 5. Having no other compromise than clinker, draining the sulfur and bringing down the ring, without considerating the efficiency. 6. If it is needed, after 24 hours, suddenly stop the kiln, let them cool for 3 hours and then start again. If the ring has a regular thickness, this is, there are not gaps (the scanner shows it without a 360° temperature strip); there is no other solution than to stop, cool and bring it down. In the zones prone to generate rings is estimated, by the results obtained, to install basic refactory (spinel) without iron, instead of aliminous silica (acid), since the thermal conductivity of the basic one makes the hot side to get colder, decreasing the attaching tendency, and if it happens, it will frequently detach while functioning.




Page 98

Other recommendations to drain the sulfur (others than setting up a by-pass) are: 1. Search for raw materials with less sulfur or more alkali. The SO 3 introduced by raw materials affects less than the one that comes form the fuel, since in the first case it is going to be incorporated into the phases and in the second case it will become directly a gas. S=Coke→ SO3Coke

Contribution ratio 10:1

Raw meal SO3

Contribution ratio 1:1.6

Clinker Total SO3

5

12.5

1.3

0.8

1.3

1.3+1.3=2.6 without problems

7

17.5

1.8

0.3

0.5

1.8+0.5=2.3 with problems

2. Grinding very fine the coke a. Through 75μ ≥ 100 – 0.85% volatile b. Through 90μ ≥ 100 – 0.50% volatile and for the calciner, 2 more points. 3. Grinding more the raw meal, so the specific heat consumption will decrease, therefore the addition of sulfur from the fuel will decrease. 4. Assuring and oxidant atmosphere at the kiln entrance, in order to assurance the formation of sulfates, which will come out with the clinker. It is pursued to have a CO < 500 ppm, with O 2 ≥ 2 not being necessary to exaggerate the O2 level, since in extreme, it rarifies the flame (cooling it). 5. Centering well the burner so the clinker, under the flame, will vaporize the CaSO 4 as less as possible. 6. Placing more inside the burner, if the secondary air arrives without regularize itself at the burner point and directs the flame into a preferential direction. 7. Smooth flame, which obtains clinkering but not vaporizing the CaSO 4, by means of the total set of primary air, radial % and O2 excess. 8. Adjusting well the kiln speed to the residence need of the material, an excess in residence favors the volatilization, since is a time – temperature combination at that temperature. 9. If the filter dust goes enriched with SO 3, do not reintroduce it directly to the kiln, but take it to homogenization, and if possible: to hoppers for cement milling. 10. Assuring continuous and stable flows at the cyclones tower, adjusting well the unloading counterbalance valve. 11. Contributing with limestone to the coke grinding, in a 17% in order to get the vaporized SO 3. It has been tested and the results give a 40% efficacy. 12. Contributing with additives, based in micrometric magnesium, wich forms stable composites with vanadium (1600 ppm) and nikel (350 ppm) from the coke (wich favors SO 3 evaporation). The ratio

Mg

V

3

must be met; this implies 1 per thousand in coke proportionings, as a

flame additive proportioning.




Page 99

Inspecting the rings composition is fundamental in order to determine their origin. The sulfur in the coating along the kiln is on the order of:

% SO3 ( position)  0.15  position (meters) Finally, it is very important to indicate if high sulfur cycles are being handled, we have to watch very carefully the cyclone premature crammings (or bars) diagnostics and if they are produced, cleaning very well in order to leave no residues. The canons number and arrangement is fundamental for the continuous functioning (air pressure ≥ 6.5 kg/cm2), as their shot sequence so, what a canon cleans will fall where another one will clean later and not otherwise. Also is fundamental not to suffer fuel flow faults, without raw meal cuts, that will cool abruptly the tower, falling the already sticked (normally at the hot stage end) formed as a cram. The fast detection of an incipient cram is fundamental, since the easiness and mainly the risk to solve it, depend on the stored raw meal in the stage. For the automatic stage cramming detection (by menas of PLC logic), the pressure measurements in the cones and the differential measurement between the gas and the material temperatures can be used. A crossing between both temperatures, and a cone pressure elevation, are not enough to declare with assurance a cramming, since they depend on the instrumentation localization, and where the material is being held. This has to be complemented with temperature and pressures movements in the rest of the stages, so: A cramming at the hot stage will raise the pressures above it at the tower, and will raise the gas temperature of the previous stage, since the tube originated gas will not cool with the material that should come out from this hot stage. A cramming at the previous stage from the hot one, will raise mainly the gas temperature of the hot stage, since material is not falling, therefore, not cooling the gas which rises from the ascendant. In kilns with calciner, the temperature is controlled by the fuel of it, a strong difference between the fed fuel and the normal one will be watched. Crams at the colder stages than the last one, are anormal, but if they occure the same philosophy will apply. Briefing, any sustancial pressure (relative to the exit pressure, since the production level can change) and/or temperature movements, between gases and material and between stage gases, must warn of a “possible cram or clog”, warning the field personnel and acting with diligence on the verification, otherwise, to stop. The fast tower stabilization during a kiln start up is fundamental as well as the fuel cuts due to raw meal feeding failures. Although not frequent, other volatile cycles have to never be minimized, mainly Chlorine, because of its highly evaporation, as it is showed at the next table.




Code DP-03-1

Evaporation and volatile valves.

Page 100

K2 O

Na2O

Cl-

SO3

0.2 a 0.4

0.1 a 0.25

0.99

0.35 a 0.8

(VH)

0.15

0.4

0.05

0.15 ª 0.5

In Raw Mill

(VM)

0.36

0.56

0.35

0.17

In cooling towers

(VT)

1.00

1.00

1.00

1.00

In filter

(VF)

0.4

0.7

0.3

0.8

1.2 a 1.6 (*)

1.1 a 1.3 (*)

Evaporation (E): % evaporated (being 1 to 100%) that becomes gas. Depends on the flame conditions. Valves (V): % not condensed (being 1 to 100%), therefore escpaes from the condensating conditions zone. Depends on the conditions zone (Temperature and O2). KEV

K · E E

KE(1V)

K K(1E)

In Kiln + Precalciner

1 K 1E(1V)

1 Base de cálculo: 1 unidad de volatil en la alimentación

Flow (K): how many times the calculation base unit circulates. By=% by-pass (being 1 a 100%) Sup. m=1 (0%) by=0 (10%) by=0.1

66 9

1.5 a 3 (*)

(*) decrease a llitle bit. a0= (1-E)ּk

a4= (1-VH)ּa3

a8= (1-VM) ּa6

a1= Eּk

a5= VHּa3

a9= VMּa6

a2= byּa1

a6= mּa5

a10= a9 + a7

a3= (1-by)ּa1

a7= (1-m)ּa5

a11= (1-VF) ּa10

m= % kiln gases (being 1 a 100%) that goes to the raw mill. a12= VFּa10 BALANCE: 1= a0 + a2 + a12 Resulting in:

k

1 (1  E)  by  E  VF  ( VH )  E  (mVM  m  1)

In absolute values Ai = Aּai where A= % Volatile at the entrance set.




Example: chlorides balance.

 coke  ּ  r.meal 

Entrance = A = 0.021 + 

Chlorine data In raw meal feeding: 0.021 Limestone: Clay:

0.007 (80%)..... 0.0056

0.021 (19%)..... 0.0040

Iron: 0.031 ( 1%)..... 0.0003 Raw Meal: Dust : In Coke = 0.01

Code DP-03-1

0.0099 0.16

(93%)..... 0.0092 ( 7%).... 0.0112

Raw Meal 0.0204

In Clinker = 0.007

Page 101

0.01

= 0.021 + 0.06ּ 0.01 = 0.0216

by  0  KTeórico     66  m 1  Kreal ? Ao = A ּ ao = 0.0216ּ(1 - 0.99)ּ K Ao = 0.007/1.6 in % reference to the raw meal So Kreal = 20.4 << 66 ok Circulant Concentration = 0.0216 * 20.4 = 0.4% Cl By less evaporation or more Valve

4.3 Process protection interlocks. At the margin of the normal machinery sequential start up interlocks and the corresponding ones to the mechanincal and electrical protections, process protection interlocks must be defined, wich avoid functioning in anormal conditions, like:

a) Under CO presence and O2 absence: alarm levels are fixed for maximum CO and minimum O 2. b) Fuel feeding functioning without raw meal feeding: the feeding failure detection by temperature increment has to act over the fuel flow and/or over its shut down. c) Raw meal feeding functioning without fuel feeding: the significative decrease detection of temperature at the tower has to be solved immediately to avoid later clogs due to slices falls by an abrupt tower cooling. This situation may happen for fuel flow failures (an increment of O2 will show it) and for raw meal overfeeding (an increment at the cold stage pressure will show it). d) Functioning with an obstructed or clogged stage: the automatic detection of a stage clogging must warn the OPR (Operations Room) control (panel) and if the alarm continues after a while, the kiln must be stopped.




Page 102

Part 5- Refractory Kiln Reparations Protocol This section tries to establish a protocol for the kiln reparations attack, which entails plannings about: 

Cooling.



Activities.



Resources and materials.



Heating.

5.1. Refractory coating. Solid materials with a high fusion point and with mechanical and chemical benefits, and low heat transmission, which protect and insulate the structural component, the steel, from the high temperatures that the process demands.

5.1.1 Benefits  Mechanical resistance (N/mm2)  Chemcial resistance (to the operation process)  Abrasion resistance.  Refractory strength (temperature and conductivity)  Volume stability. Thermal changes stability wich are interpretated in specifications of:  Chemical composition  Mechanical resistance (N/mm2)  Ta Refractavility (softening temperature, in °C)  Density and porosity (% volume)  Thermal conductivity (W/mK or kCal/m h °C)  Thermal expansion (% vs temperature °C)  Temperature changes resistance, elasticity (cycles) 5.1.2 Types Conformed

= Bricks or prefabricated pieces

Not conformed

= Concretes and mortar (they are made in situ)

The advantages and disadvantages of each type (what and where to use them) are summarized as following:



Characteristic Preparation Properties

Bricks Already formed Constants

Mounting Storage Drying need Aplication zones Joints (possible infiltrations attack) Structural characteristic Demolition

Difficult Good Little Simple geometry 1/Break (bad) weak: a piece fails = the rest fails Simple


Page 103

Concretes In place According to preparation and aging Fast with projection technics Caducity Much Complicated geometry 1/zone (expansion) (good) Strong (monolithic) Expensive

That is why one or another method will be applied, according to this and other possible questions. Except the kiln tube and the first satellite meters, the rest usually is mounted with concrete according to high efficient projection technics (few resistance losses). 5.1.3 Refractory Brick Reviewing the fusion temperatures for different oxides in nature: Oxide ThO2 HfO2 MgO ZrO2 CaO UO2 SrO2 Cr2O3 Al2O3 BaO SiO2

Fusion temperature °C 3200 3050 2800 2700 2580 2500 2430 2400 2020 1923 1470

Fe2O3

1400

Comment

Al2O3 () = corundum Goes from tridymite (870° C) to cristobalite. At 1710° C obtains melted silica

We find that when eliminating the few abundants (Thorium, Hafnium, Uranium, Strontium and Barium) just left: MgO, ZrO2, CaO, Cr2O3, Al2O3, Fe2O3, SiO2, which will be the components of the different refractories, which have to be added to others from the carbide family, as the Silica Carbide (SiC, with a fusion point of 3,000 °C), which comes from the steel industry, or nitrates as the Si 3N4. Notice that SiC has to be used where it is known that the vitrification temperature will be reached (> 1000 °C). Fulfilling the different requeriments in the clinkering operation, two distinctions are made:  Basic materials: thermal and chemical resistance is seeked.  Neutral or acid materials: Resistance to abrasion and/or mechanical resistance are seeked.




Code DP-03-1

Page 104

Bricks behavior according to their composition Spinels % MgO

% SiO2

D ol o m it es

% Al2O3 MgO

Al2O3

Basic

Al2O3 Neutral

SiO2 Acid

5.1.3.1 Basic bircks (MgO base) Since the MgO is very abundant in nature (MgCO 3), and its refractavility is higher, as its chemical stability (attacks) is higher than the CaO (the CaCO 3 is more abundant than the MgCO 3), the magnesium is the base of the basic bricks, called that way because of its MgO basic behavior. The basic bricks technology begun with natural o synthetics dolomites (MgO.CaO, product of the carbonates baking), but since this type of bricks presents little elasticity*, elastic refractories were begun to add, in order to resist ovalness and deformations problems in the kilns. Next these are related and written down the problems they entail, but before it has to be indicated that this elements are introduced in one of the more stable forms that the nature presents: the spinel structure AB2O4 (in the MgAl2O4 nature).

MgCO3

T MgO 

T = 700 – 1000: Calcination = no refractory T = 1500 – 2300: Sintering = refractory T > 2800° C: Electrofused = high quality refractory The dolomites eutectic has a fusion point at 2370°C

Nomenclature %MgO

%CaO

Magnesium

+1%

>94%

Dolomite/magnesia 40%

57%

Magnesia/Dolomite 58%

38%

*Note: By temperature cycles or by mechanical problems, as being a rigid brick it breaks.

Basicity ratio (B/A) = (MgO + CaO + Fe2O3 + Na2O) / (SiO2 + Al2O3 + TiO2) B/A > 1 B/A < 1

 Basic

 Acid / Neutral




Page 105

ELASTIFIER Magnesia –spinel: MgO + MgAl2O4 2800°C

2135°C

(MgO . Al2O3)

Magnesia – chromite: MgO + MgCr2O4 2180°C

(MgO . Cr2O3)

Al2O3 attackable by alcali – watch corrosion Al2O3 attackable by clk liquid phase – watch corrosion Cr (III) attackable by alcali – watch corrosion Note: not recommended by hexavalent chrome toxicity.

Magnesia – ferrocrhomite: MgO + FeCr2O4

Cr (III) attackable by alcali – watch corrosion

Iron as a coatinginess.

Fe2O3 attackable by REDOX – watch corrosion

(FeO . Cr2O3)

Magnesia – hercynite: MgO + FeAl2O4 (FeO . Al2O3) Iron as a coatinginess. At the contact zone C 2F y C2AF are formed, wich are the higher heat transmission, the zone cools and more coating is added, and without spinel structure.

Magnesia – zirconite: MgO + ZrO2 MgO

Fe2O3 attackable by REDOX – watch corrosion Al2O3 attackable by alcali and liquid phase – watch corrosion

ZrO2 attackable by SO3 – watch corrosion

2370°C The ZrO2, as addition < 5%, contributes refractivity, clinker attack resistance, mechanical resistance and REDOX resistnace.

Magnesia – galaxite: MgO + MnAl2O4 (MnO. Al2O3) Corrosion resistance, REDOX stability.

Other combinations less used by less refractivity are: 

Magnoferrite: MgO Fe2O3 (1710°C)



Enstatite: MgO SiO2 (1550°C)



Fosterite: 2MgO SiO2 (1890°C)



And other with Zn, Cu, V, Ti, Ge, etcetera.

Must consider that: 

In MgAl2O4 spinels the CaO/SiO2 ratio must be between 1.4 (if less it becomes a silicate with a low fusion point) y 1.87 (if more it corrodes).



The ZrO2 is presented in two forms, the Rutile type (with temperatures from the ambient one to 1100°C) and the Fluorite type (for temperatures from 1100 to 2700°C), of very different volume and that is why it is added so little.

This magnesia (MgO) combinations are used and less dolomites (MgO.CaO) because they are delicated and attackable. By delicated it is understood the need to conserve them in a dry place mainly because of the magnesia and calcite hidratation: MgO + H2O

 Mg(OH)2 : Slow reaction. Volume increase causes destruction CaO + H2O  Ca(OH)2 : Fast reaction



Attackable by: SO2 (watch corrosion): CaO + SO2




Page 106

CaSO4 + CaS

SO3 (watch corrosion): CaO    C2S (that is why it coats well), however: SiO 2clin ker

C2S(of clínker) + MgO + SO3 27% increase

 CaSO

4

+ M2S (phospherite) :

 compact  do not flexes  spalling (it breaks)

The more used combinations are as follows, ordered by price, refractivity, plasticity and REDOX and chemical resistance: MgO with:

Cr2O3 Mg Al Spinel

ZrO2

Besides the chemical component, there is the granulometric one, getting higher resistances to the clinker attack with large grains. Although this subject will be explained at the formats section, the ones used in basic bricks used to be VDZ (german standarization). The basic bricks properties will be explained with the rest. Note: joining between the different grains may be done with direct bond* or through chemical bonds. The direct bonding is preferable since the chemical bond is obtained through cementant composes (C2S belite type), and therefore they result more vulnerable to the chemical attack, as we saw: C2S + MgO + SO3  Ca SO4 + M2S (phosphorite) 5.1.3.2 Acid or neutral bricks: Al2O3 SiO2 If the bricks are MgO mainly because of the magnesium nature abundance, the base of the acid or neutral bricks are the alumina and the silicon, they are aluminosilicic or silicoaluminous, depending on which predomines, the neutral component (Al) or the acid one (Si), which has a low refractivility but gives high mechanical benefits, it is used where high thermal benefist are not required. The habitual elaboration of this brick type is as follows (by 5% in Al 2O3 content): Type High alumina

%Al2O3 > 70

Medium alumina

50 a 65

Low alumina

40 a 45

Silicoaluminous

< 35

As higher alumina higher the refractivility (Ta), higher the resistance, higher the fragility (less elasticity), higher the chemical attack and higher conductivity (less insulating strength).

In the other hand the alumina might be, ordered by price and quality (refractivility and mechanical resistance):  Corundum base (Al2O3)  Andalusite base (Al2O3·SiO2)  Base Bauxita (Al2O3·2H2O)  Bauxite base The most recommendable because of its cost and resistance ratio is the andalusite.




Page 107

5.1.3.3 Carbon usage The carbon, except by its oxidability, is the best refractory since it is practically infusible (it begins at 4330 °C) and chemically stable, being the base for the steel carbides, and as we have seen at the already used silica carbide in a 30% of high alumina brick. Despite as it has been told, is not checmially stable under oxigen presence, since it oxides into CO 2 and more when 700 °C are surprassed. That is why the carbon only can be used if it is included as an matrix element in order to obtain more benefits and in low temperature and low O 2 level zones. The carbon (graphite) when getting in touch with O 2 becomes CO2, leaving holes for mechanical resistance losses and making it susceptible to the chemical agents attack. 5.1.4 Bricks formats The ISO and VDZ formats differ on how they divide the circumference in order to define the pieces size (a), leaving the wedge (b) depending of the thickness (h) wished and the kiln diameter (). The ISO format is also called /3 since it is normalized (a+b)/2 = 71.5. Although is more comfortable the /3 format because it constitutes bigger pieces (and therefore less pieces by ring), it is left for the acid or neutral bricks because of their low expandibility, while for the basic ones (more expandibles) it is better small bricks in (a) to be able of inserting between them mortar joints or an arranging of natural joints, if the circular contour is tried to be built better, the VDZ format should be used. The length (L) in both cases is normalized to 200 mm (198 mm), 300 mm can be used to cut the closing line to the size. The thickness (h) will depend on the kiln diameter with the next strip:  (m)

h (mm)*

< 2.8

160

2.8 – 3.5

160 / 180

3.5 – 4.2

180 / 200

4.2 – 4.5

200

4.5 – 5.0

200 / 220

5.0 – 5.5

220

5.5 – 5.8

220 / 250

> 5.8

250

*Note: except for other retention rings thickness. The nomenclature for the Bh formats, where:  B: is written if it is VDZ format, if it is ISO its omited.  : kiln diameter expressed in “whole” meters.  h: thickness expressed in cm. For example, B620 talks about a brick expressed in VDZ format, for a 6 m kiln diameter and a (h) of 200 mm (in other words, 20 cm).




Page 108

In order to mount a circumference with a straight base (a) and shut it perfectly, form combinations are realized for a greater and smaller kiln that the one being calculated, tending to +1, for example, for a 4.55 m kiln the 3 and 6 m (or the 2 and 6) kiln formats are used, it would be if they are mounted with a 20 cm thickness: 

B620 andd B320 VDZ formats + BP-20 y BP+20 shutting pieces.



620 y 320 ISO formats + P-20 y P+20 shutting pieces.

For a well mounting, it is very important that the supplier takes care off his molds state in order to assurance that all the pieces arrive in the dimentions tolerance, as well as marked in the hot size so, the “normal” pieces (with a little wedge) will not be applied other way around and therefore the kiln when rotating will “spit” the bricks. With the “fast” one (greater wedge) does not exist that problem, but a good piece marking allows to recount the number of each type that the kiln “has asked”, and in that way knowing what is their real size relation for future orders, besides of being a good indirect measure of the shell deformation. The pieces, could be smooth or grooved, allowing the last ones (because of the cavity they form), in the clinkering zone, the clinker penetration, working as a coating subjection base, but if it is wrong placed in a kiln zone where is not possible to coat then a lot of heat will be transmited to the kiln shell.

a

They can have, or not, a cardboard joint of 1 to 2 mm in its axial side (figure a), and also they can have or not a radial plate of 1 mm in thickness in order to favor coating (figure b), considering that in cold zones, an isulator can be placed between brick and shell.

b

Mortar or “in bone” mounting The mortar usage as a radial joint, from the point of view of piece over piece settlement is better than without mortar (in bone), but the placing time rises in 20%. 5.1.5 Refractory brick placing systems. Three types are basically distinguished: Tensions

Glue

Circle segments or form

By sections, the brick is fixed to the shell and it is rotated.

Is the safer and the more used.




Page 109

5.1.5.1 Brick placing. 0.- If it is a demolished zone, a fastening must be weld to the not demolished brick in order to avoid moving while rotating. 1.- The kiln shell is cleaned of dust and husks. 2.- If the shell is convex (deformed) it has to be filled up with refractory. 3.- The base line is traced for the lines starting up.

4.- Guide lines are raised (parallels to base line traced) to 1.5 m of the base line. For reference on the advances at the fronts it may be used each shell section weld lines The recommended alternative to place the bricks is the intercalated way (seen from the top):

Kiln axis

Weld

5.- There must not be outphasing in the normal bricks / fast advance bricks ratio, since if too many fast bricks are placed together, steps are produced there and in the normal ones zone that was pretended to compensate, that is the importance of not outphasing.

Steps



Code DP-03-1

Page 110

Parallel sides

6.- The pieces are seated well (with rubber hammer) against the kiln and between them, therefore their sides always stay parallels and will never rock against each other. Due to this, if there are rivets or dents, they must be rectified with refractory mortar.

Mortar Last line

7.- The closing line must not have less piece length than L/2, so in order to gain consistence they will be alternated like this:

8.- The rings locking must be made tensing first with a jack and then shutting with the locking bricks and some brick cut for that effect, then ending with iron plates of 190 x 190 x 2 mm sharpened at on edge (never more than two in the same joint). It is advised that a closings line will not be generated, in other words, that the ring locks never coincide.


Old brick

30 mm 3 mm 30 mm 200 mm 10 mm

Metallic piece pieza metálica

5.1.6 Refractory life – campaign The normal wear measured in grs/ton clinker in the kiln tube might be expressed as a function of the kiln diameter: grs/ton clk = 28 x 2 Being estimated by G. Hotz the refractory life days or campaign in: days = 5840 / 2 It can be reduced by a wrong refractory election according to zones requesting, or by operation conditions, and to enlarge it for the same reasons.




Page 111

At this line is fundamental, in each repair, to write down the revestiment residual thickness, meter by meter, and to evaluate the wear in cm/million of clinker tons, understanding that a brick (at the basic zone) with thicknes < 5.6 cm does not have mechanical resistance and therefore it will be a hot spot reason, not being that residual above 6 cm. In this way if the expected wear of a refractory zone is known (X cm/million of tons) and the useful thickness is thickness - 6 cm, how many tons each zone will resist could be calculated and therefore:





Which the campaign would be (tons



Where will the failure appear



How compensated are the zones to select zones with whole number campaigns.



Where does the refractory quality has to be improved (or search for specifications) vs. zone problem.

months)

The heating (and drying) and cooling (by programmed shuts or not) will also determine, in the first case, the campaign, and in the second one, the brick that can be left. All of this will be seen in the Kiln repairs protocol. The accurate coating formation will protect the refractory (see coatinging indices according clinker chemistry). 5.1.7 Hot Spots When a kiln shell hot spot appears three questions arise: Where is it? Because if it is in a zone where there is no habitual clinker coating it will be difficult to cover (must stop the kiln). If it is in a support zone is particularly dangerous and must stop. What size is it? Since if it is too large, even it is in a stable coating zone, it will be very difficult to cover it and the kiln must stop. What is the hot spot intensity? Since this will indicate at which temperature the shell is:

ShellTemp. (°C)

Intensity

380 to 420

Fans are required

420 to 480

Little visible at night

480 to 600

Dark red color

> 600

Live red color and shell deformation




The main factors that benefit or harm the refractory (and with it the hot spot possibility) are:

Heating For totally new refractory or too much concrete: 0.03**D*L

Maximum 42 hours.

For parcial repaired refractory or little concrete: 0.02**D*L

Maximum 30 hours.

Page 112




Page 113

Kiln Control ¿to stop or not? In order to take the decision of stopping or not the kiln when a hot spot appears or there is risk of happening, it is needed to ask:

*Note: A zone that has a lower temperature than the next ones, shows the later failure of refractory and coating falling that protects it (this zone show less temperature by having a thicker coating, in other words, has a thiner brick)

*

5.2. Refractory Concrete Just as the conventionals concretes, they are composed of cement + fine aggregate + thick aggregate + placing water, but what differs from each other is the seeked refractivity of the following components:  



Cement



Aluminous cement

Aggregates  Aluminas (bauxites, andalusites, corindium) and other dense components as in silicoaluminous bricks (ZrO2, SiC, Cr2O3, TiO2). The thicker part surpasses 120 microns and the thiner part or matrix is below, being the most importat part of the concrete benefits. Also additives are used in concrete, like elastifiers and plasticizers, and also for a fast setting.




Code DP-03-1

Page 114

Taking care of the chemical bonding (the most attackable by the clinker liquid phase and other agents) the concretes are classified as:

Clasification

Cement %

CaO %

< 3%

< 1%

Low cement content

3 to 10%

1 to 2.5%

Conventional or Normal

10 to 30%

> 3%

Ultralow in cement

*Note: There are concretes without cement base, like the non hydraulic setting, in which the water is substituted by colloidal solutions, which flocculate thanks to additives forming structures that bond the arid, simulating a setting letting the temperature to be the one which will make the ceramic bond.

Taking care of the concrete rheology, these are classified as: 

Normal or conventional: they need a great ammount of kneading water to get the required consistence, thus, that water loss (or evaporation) will rest resistances.



Thixotropic*: they need less water as already having this behavior and acquiring the consistence with agitation and/or vibration (as the paints).



Autopours: Materials that with a little more water get consistencies, thanks to additives, very soft (that last) which allow placing without the need of vibrating in order not to leave spaces without filling up in the mold or form



Compactibles: they are refractory mortar doughs with plastic consistency, which can be molded “by hand”.

Taking care of the way they will be applied several technics are distinguished: Technic

Benefits % (Resistance, Porosity, etc)

Rebond losses %

Vibration

100%

0%

Gunite: The dry concrete is pumped and water and a fixing additive (accelerator) are added at the toe, which causes segregations.

50%

20%

Shotcreting: The already kneaded concrete is pumped (with sufficient water) adding at the point a flocculant (being, more than an accelerator, a rheology modifier, in other words, of its elasticity, viscosity and fluidity).

80 a 90%

7 a 10%

90%

0%

Autopour Dry concrete

Shotcreting:

High pressure air

The old concrete is fixed. Efficiency of 3 to 10 ton/hr and reaches heights of 60m.

Water Pump Mixer

Flocculant

Isolating time < 2 minutes Aging time = 2 hours




Page 115

The % kneading water influence in hot mechanical resistances, besides the drying problem during the kiln heating, can be expressed as: R (N/mm2) = 210 * Ha-0.98 * T0.23 Where: Ha = Water % needed to pour it. T = Temperature in °C at which it will work Therefore, a concrete that needs a 4.75% of water and another of 3.5% will have, at 1100 °C resistances of 228 y 308 N/mm2 respectively. Other expression in base of the initial conditions (0 subindex) is: R1 = R0 * [1 – {0.0004 * (T0 – T1)} – {0.18 * (Ha1 – Ha0)}] In the vibrated concretes there must be selected the needle vibrator type that will be used in order to get a specific influence zone (Zi): Zi (m) = 0.02 * fc0.5 Where: fc = centrifuge force, in kg (transmitted force from the vibrator to the dough) Since normally it is seeked that Zi = 1m we have that fc = 2100 kg, which gives an expression to a specific needle diameter of the vibrator  (in mm):  = (33.33 * fc)0.4 or: fc = 0.03 * 2.5 That result in an 87 mm diameter for fc = 2100 kg value just metioned. When vibration is at a mold that does not allow internal vibration, contact vibrators are used, which are anchor to the mold. In these, higher influence zones are seked (1.5 m), so higher forces (6000 kg) are required too. When evaluating a concrete price it is importat to check its density, since the volume that will be placed will take one or another material ammount, this is, it must be compared: 1 * price1 vs 2 * price2 5.2.1 Refractory concrete drying The following aspect must be taken into consideration: 1. Matrix: aluminous cement + fine refractory aggregates. 2. Kneading water for rheaology (physical water) and for cement hydration (chemical water).




Page 116

3. By thermogravimety (weight vs. temperature) it will be observed until 200 °C there is only physical water loss, but from 200 to 350 °C the chemical water from the cement hydrates, as much as the instables as the stables, and from 325 to 350 °C that chemical water is lost very quickly and might chip when trying to leave. 4. When heating, the heat transmission is not the same along the thickness of the poured concrete, but in the first 5 to 10 cm (hot side) they are being heated and dried, while in the cold side is still cold and nothing is happening. When that first layer has completed its drying, the heat goes through the next and so on. With all these, what happens when a concrete is heated rapidly? The first hot layer losses its physical water and the chemcial (200 to 350 °C) and because it has escape, more difficulties are not presented, since there is enough evacuation channels for the rapidly released chemical water. When the heat is transferred to the next layer, the physical water is lost well, since its eliminating speed is moderated, but when that layer reaches temperatures between 325 and 350 °C, the chemical water is realeased at higher speed than the first layer channels are able to evacuate, generating a high water pressure and, except having an escape at the cold side (shell), will reach such a pressure that will explode that 5 to 10 cm of the hot side. Because of all the previous, there is no problem when heating as long the 200 °C are not reached, but from 200 to 350 °C, we must be ver careful. The most recommendable is to have a smooth heating and to facilitate the water evaporation generating channels when puncturing the dough during its pouring, and even during kneading adding polipropinele fiber, wich when heating, they will be destroyed leaving channels. A way of knowing that the drying has been finished along the thickness is to take an shell pyrometric reading, if it surpasses 110 °C all the water has been eliminated, as physical as much as chemical.

5.3. Refractory Specifications In order to take care of the diffent exigencies that the kiln operation will make to the refractory, this must have specific properties that are summarized and quantified as technical specifications, according to them the technician will choose the optimal refractory, for the zone characteristics to be installed.

5.3.1 Chemical request – chemical wear The refractory attackability by chemical agents or REDOX conditions are denominated as refractory corriosion (which can derive from shell corrosion) and implies the next specifications:


Chapter III: Clinker 

Chemical and mineralogical composition



Density and porosity


Page 117

Sulphate or alkali corrosion Depending on the imbalance or decompensation (MSO 3) between alkali (Na y K) and sulfur (S) and chlorides (Cl-) an attack from the remainder will be produced. That is how the sulphate module is defined MSO 3: MSO3 = (SO3/80) / [ K2O/94 + Na2O/62 – 2*Cl-/35.8 ] Which express the compensation between alkali and SO 3 and Cl- in order to give alkaline sulphates and chlorides, SO3 or alkali being able to remain but not chlorides, since the little level of this one is given by needs of the own process. MSO3

Formed

Remains

Corrosion

>1

K2SO4, Na2SO4, KCl, NaCl

SO3

=1

Ídem

Nothing

No (see note*)

<1

Ídem

K2Oex, Na2Oex

By alkali

excess

By sulphate SO2/SO3

*Note: There are no corrosion problems of these types if MSO 3 is between 0.83 and 1.0. It is admissible 0.83 since the clinker might be part of the excess as NaC8A3 y KC23S12, reducing quality.

The alkaline sulphates and alkalineterreous and alkaline chlorides corrosion, happens in very different zones, due to the different volatilies, therefore even a material is prone to be attacked, for example alkaline chlorides, if where it is placed is not where it condesantes, it will not be affected.

in gas The attack of these compounds is not a chemical type but mechanical: 

NaCl, KCl, Ca Cl2 are infiltrated in the holes and harden the refractory.



Ca SO4, Na2SO4 y K2SO4 as an incoatingation or ring constitute a risk for the refractory when they fall.




Page 118

a) SO2/SO3 corrosion while infiltrating the refracory:



MgO + SO3 + 2 C2S

CaSO4 + M2S

The MgO comes from the basic refractory and the C2S is the clinker belite. The MgO part that forms some of the spinels is more resistant by the chemical stability inherent to the spinels.

Volume increase of 27% with respect to SO3+CaO occupying all the pores space, hardening and by temperature changes the attacked thickness is chipped: spalling by sulfur (Note: M2S = 2MgOSiO2).

CaO + SO2



CaO + SO3

CaSO4 + CaS



CaSO4

Therefore it is observerd the close relation between chemical composition and porosity, as a measurement of attack impermeability. MgO (f.p. 2800 °C) + SO3 + 2C2S (f.p. 2130 °C)*



CaSO4 + C3MS2 (f.p. 1575 °C) + CMS (f.p. 1498 °C)

*Note: The belite from silicoaluminous bricks or basic ones chemical bonded by belite.

Because of all the previous, the following will be seeked where is anticipated the presence of SO 2/SO3: 1. Do not use silicoaluminous refractory (C2S presence) 2. A low level of CaO: CaO + SO 3 protect them)



CaSO4 (not dolomites, except of stable coating zones which will

3. A high spinel percentage 4. Low porosity 5. High flexibility (high spinel or with zirconate) b) Excess alkali corrosion when infiltrating to the refractory. This kind of corrosion appears at the high transition zone: MgCr2O4 + K2O (ó Na2O)



Chromite (Cr3+)

K2CrO4 (ó Na2CrO4) Cr6+, yellowish green and toxic. Volume increase and hardening the attacked thickness, chipping while changing temperatures. Spalling by alkaline chrome attack.

Al2O3 + SiO2 + K2O Silicoaluminous refractory



Feldespars K2O Al2O3 x SiO2 Of greater volume, causes hardening, chipping and spalling by alkaline attack.

Therefore, where excess alkali is observed, the use of chromite will be avoided (magnesia – chromite or ferrochromite) as well as silicoaluminous materials with high porosity, being recommendable the spinels use without chrome.




Page 119

On silicoaluminous birkcs, as higher its silica content higher its attack protection, but lower its Ta. Corrosion by Chlorides The alkaline chlorides with chrome from the basic bricks form chromates which harden them and cause spalling. Since this situation happens at the transition zones, the use of this bricks type will be limit to the clinkering zone, avoiding the use of high belite content bricks. REDOX Corrosion The reducing atmosphere, by bad combustion (fix carbon without burning) or by bad burner direction towards the refractory, causes the iron 3+ to iron2+ reduction with volume reduction, then by oxidant conditions, to become again iron3+ with a 20% volume raise, causing redox spalling. Fe2O3



FeO



Fe2O3

If this kind of operation happens or it is anticipated, refractories will be seeked, basic or not, iron free or with a little iron. If coatinging is wanted, the ones with iron in a spinel structure will be preferred since they are more stable. Corrosion by CO2 (activating corrosion by sulphates) Previously it has been shown that: CaO (dolomite) + SO2



CaSO4 + CaS

With organic compounds infiltrations: CO + CaSO4



CaS + CO2

Now reacting as: CaO (dolomite) + CO2



CaCO3

Which causes a volume increase = Spalling. Corrosion by clinker liquid phase. Eutectic Reaction. Al2O3 + CaO free clk Alumina base refractories, placed at the exit zone and having attachments due clinker melting (usually by a very hot kiln due hard to burn raw meal) The alumina base spinels, because of its spinel structure, present higher resitance to these attacks than the silicoalumous ones.

T 1400 C   

C12A7 Having a low fusion point (1365 °C) causing corrosion by eutectic (fusion  sinking or licked refractory) This attacks kind is called “spinel corrosion” if this one is the one attacked.




Page 120

Under SO2/SO3 presence: Al2O3 + CaOfree* + SO3



C4A3.SO3 (fusion point = 1590 °C)

*Note: it can be C3S or C2S if there is overheating.

This kind of reaction also can happen to the chrome, forming calcic chromates with a low fusion point. The solution is the zirconites usage, which is affected by the reaction; they form calcic zirconate with a high refractivility. The refractories properties calculations are seen at the appendix E.

5.3.2 Thermal stress – thermal wear Under the heat, the refractory must resist its level and its changes; that is why the refractivility and the cycle’s specifications are incumbent on the refractory. In the other hand it must avoid the heat reaches the structural shell, incumbent on the thermal conductivity and, for last, the volume stability with temperature: the thermal expansion, even it is not a decisive element, it is essential its previous knowledge so, during mounting, placing the expansion joints which allow creating free spaces, which will be occupied when working temperature will be applied. The thermal expansion, as it ends being, or not, a tensions problem it will be seen at mechanical wear In order to count the refractivility specification it is resorted to express it by terms of maximum working temperature, by which it is resorted to the material test under a 2 kg/cm 2 pressure, and to a temperatures scaling until it softens (Ta = softening temperature). Ta is used in °C to express the material refractivility. As the refractivility is an indirect measurement of the material fusion point, it has the previously described order: Basic refractories High alumina Med. and low alumina Silicoaluminous

> 1650 °C 1450 to 1650 1350 to 1450 < 1350

(alumina content > 70%) (medium alumina of 50 to 65%, low alumina 40 to 50%) (alumina content < 40%)

This factor is determinant, but not unique, when selecting the kiln material and zone. The abruptal changes resistance is determined with the cooling-heating cycles test, which shows the material resistance at temperature changes, but now in almost all the refractories the levels are very similar (100 cycles), even the silicoaluminous, but where those cycles are truly created, at the very high heat zones and, mainly, at the flame zone, therefore the resistant cycles importance is more at the basic bricks Finally, the thermal conductivity (inverse to the insulating strength) is inherent to the composition; therefore the most insulatings are the ones with higher %SiO 2, then the ones with higher Al2O3 and, finally, the ones with more MgO. In other words, the insulating order from higher to lower is:  Silicoaluminous  Low alumina  Medium alumina  High alumina  Basics




Page 121

This order is disturbed with the inclusion of silicon carbide (SiC); its conducting strength makes the concretes or bricks which contains it, the most conductives. The importance of the refractory conductivity knowledge is double: 

On one hand in order to anticipate possible cracks at the kiln shell if it has been placed at an inadequated heat zone.



On the other hand, in order to judge hot spots at the shell and being able to define the still remaining thickness. Therefore a hot spot at the zone of basic may mean nothing, it can be occasional, while if it appears at the zone of silicoaluminous (very insulating) it almost invites to think in a hot spot (refractory failure).

A coating loss makes the temperature to rise to 500 °C right away at its side that faces fire, while it is only prepared to rise at 50 °C/hr. The thermal wears are observed by thermal spalling in which it refers to cycles and laminations referring to: 

High thermal load: as an example by raw meal failures or by rings, pulsant flame (irregular fuel feeding), wrong burner orientation, high flame intensity or flame way too open, and wrong raw meal design.



Inadequate material (Ta) for the zone in which it has been placed.

Having the support of the length and flame temperature calculation, the anticipated profile can be calculated and therefore adequate refractories can be placed by zone, for example:

And in a therocial way, being  the internal kiln shell diameter, it can be zoned in the next way:



1.5 

2.5 

7.5 

9.5 


Page 122

11.5 

Kiln with satellite cooler R. Meal 1

0.5 

1

0.5 

5

1.5 

2

6.5 

2

8.5 

Resto Rest

(Approximated total kiln length == 1616 (longitud aproximada total horno

)

10.5 

Kiln with grates cooler 0.5  Exit

1 Low tran.

5 Chamber (sintering)

2 High transition

2

(longitud aprox. totaltotal kiln horno length==14.5 14.5 ) Resto (Approximated Rest

Safety zone

Calcination zone

Note: Normally the first two kiln supports are 1 and 5.

This zoning is only illustrative; the physical and scanner inspection will determine the meters which limit each zone and the most appropriated materials for each one. It must be taken into consideration that any change at the burner disposition or its airs, fuels or raw meal, production levels, etcetera, will modify the burning conditions and, with it, will make necessary a new zoning valoration. The specific brick for the clinkering chamber will be limited to the exclusive zone that has been demonstrated or seen in the kiln which constitutes that mentioned chamber (stable coating zone). Consider as a premise that alumina means safety and transition means basic. The decision of where to limit each zone will also depend on the cost of the refractories that have to be placed, and the possibilities of extending a zone respect the other, for the different operation general conditions (feasible).

5.3.3 Mechanical stress – mechanical wear These wears can be summarized in: Wear by displacements By thermal expansion it only applies to the basic bricks, which dilate a 2% at working temperature, since the silicoalumous hardly present a 0.5% of expansion or, even, some may contract. Axial

20 cm Expansion joint (2 mm Cardboard) between lines (only the basic one)

Basic brick with a expansion, which means 3.4 mm 1.7% Shell at 350 °C with 0.012 mm/m °C and 5 rings/m




Page 123

The shell dilates 0.84 mm/20 cm, so the brick dilation of 3.4 mm is absorbed: By the cardboard joint:

2.0 mm

By shell dilation:

0.9 mm

Rest (3.4 – 2.0 – 0.9):

0.5 mm

Those resting 0.5 mm the own brick must absorb (with its elasticity) occupying pore spaces and by the natural joint (grain). In this way, the brick stresses are normal, but if the heating (for example after a shut down) is too fast, the brick expands but not the shell, by which: 

The pressure between lines increase (stress compression) and/or,



There is masonry displacement on the shell, with shearing on the cold side and/or,



The hot side dilates but not the cold one, causing spalling of the hot side (side facing fire) Considering now a radial expansion of 0.12 mm, this is compensated by the joint, by basic mortar layer which only “wets” the side, or by “in bone”, since the grains of the side generate between the two pieces a natural joint of these mm.

The compression spallings that can be generated in a same ring pieces, by pressure, have their origin in the assembly ring closure. It can be produced by: 

Excessive pressure on the jack (> 30 N/mm2) in order to place the last pieces, and/or



Closing plates in excess, which expand over a few bricks. Never place more than four plates of 2 mm by ring, and never grouped in a small zone. The plates must be put by the hot side in order not to harm the zone between pieces.

Also a joint excess is very dangerous, since the expansions will not fill up the spaces left by them, causing masonry displacements.

Axial joint excess This happens when ring pieces advance and occupies spaces, “locking it”. The rings must work in an independent way in order to “migrate” invidually, as the rolling hoops do. To avoid this problem the brick must be mounted in staggered formation.




Page 124

In this way, the ring will be occupying and distributing the spaces independently, with a little shearing on the cold side and with a ring advance or retrocession. It must be taken into consideration, also, that a heating that has been aborted will burn the cardboard, but will create and axial joint excess. Radial joint excess In a same ring the pieces will try to reocuppy the holes or spaces, which can be translated that they try to separate from the shell or that they try to rotate, in the first case the whole line could fall.

These cases are more frequent in low expansion bricks (silicoaluminous) on which the closure has been made with low pressure or the assembly of each piece has not be hammered to seat it well with the previous one. Push by inclination The own kiln inclination causes the lines to try resting on the previous one, when this accumulated push arrives to the previous line the retention ring (or material retention ring) since its bigger size will not be the one that yields by push. The line which suffers from this push must be very resistant (high alumina with SiC or elastic spinel), and if the push gets to become too big, how to save the retention ring must be restated (using conrete) and in a last case more retention rings must be placed.




Code DP-03-1

Page 125

Ovalness wear The kiln ovalness at the rolling zones will create mechanical tensions on the same type lines, not in intensity (more perceivable), than the described in “radial” pressure but all over a zone. The maximum ovalness W (%) that can be permitted is calculated as: W (%) < 0.1 *  (expressing the diameter in m) This means that if a kiln has a diameter () of 4.55 m, the maximum permissible ovalness is of 0.45% (0.1 * 4.55). The excessive ovalness may cause “spalling” and/or spirals rotating, even more if the assembly was with a wrong tighten between spirals or at the spirals closure. Wear by migration failure The rolling hoop and the kiln tube rotate at the same angular speed but different linear speed, and this linear speed difference of turn is described as displacement, measured in mm/turn. In this way the kiln rotates independently to the ring and only seats on its inferior part. The normal kiln displacement (dh) can be calculated as: dh (mm/turn) > 2 *  (expressing the diameter in m) This means that if a kiln has a diameter () of 4.55 m, the normal displacement that must be expected is of 9 mm/turn (2 * 4.55). The migration failure is presented (also called ring strangling) when the displacement = 0, this is, the ring and the kiln are adjusted or srtongly tightened, causing torsion on the kiln tube with brick destruction. 5.3.4 Kiln tube corrision It can be originated mainly by two causes: 

During stops, by condensation and water absorption.



During operation, by high temperatures.

In the other hand, the corrosion also is presented with greater facility in kilns with calciner (tertiary air usage) than in those that do not have. With calciner (tertiary ari)

10 to 20 m

Coating zone

Without calciner

Greater corrosion

10 to 20 m

Coating zone corrosion Greater




Page 126

The corrosion products are: Oxides and iron sulfides

Hematite (Fe2O3) y Magnetite (Fe3O4) Pyrite (S2Fe) y Pyrrhotite (Fe1-xS)

Iron chlorides

FeCl2

Alkali (alkaline chlorides)

KCl FeCl2 y NaCl FeCl2

The analytical of the scales of corrosion throw the next proportions: Fe2O3

60 to 90%

S=

2 to 15%

-

2 to 20%

Cl

K2O

0 to 7%

The mechanism by which the O2 enters is through refractory cracks, joints and porosities. When it reaches the shell it creates a very dense Fe2O3 layer that acts even as a corrosion tranquilizer, causing a little penetration. The SO2 oxides into SO3 consuming the refractory and joint with the shell O 2, condensanting on the refractory and leaving the zone next to the shell under reducing conditions. The SO 2 is condensated into Ca and Mg sulphated salts giving Fe 3O4 + FeS2, in a porouser layer that not reflects so much, causing the penetration to be from notable to moderate. The Cl2 (g) is the real accelerating agent of corrosion, and four cases are distinguished: 1) Over the shell (Fe0)

2) Over an already existent corrosion layer.

3) Through KCl and SO2 (gas)

4) Combined with K2O from the liquid phase

If the KCl from the raw meal reaches 1000 to 1050 °C it evaporates and never gets to the shell, but it gets to the cyclones tower, which is why, in kilns without tertiary air the volatilization is more premature than those with calciner. This corrosion occurs at the tower; while in the kilns with tertiary air occurs at the entrance zone.

It corresponds to a hot corrosion. The K2O and Na2O from clinker liquid phase are merged by the brick. Above 340°C at the shell the hot corrosion begins.




Page 127

If there is a high corrosion there are ways to minimize the problem (special paints, sacrifice plates, etc.) but what really is important is to identify the Chlorine origin and to avoid its presence Shell corrosion In the kiln shell penetrate: K2O, KCl, SO2 y SO3, O2 y H2O, with the following reactions:

Protection mechanisms a) Oxidating atmosphere, so Fe2O3 will be formed and corrosion lowered and all the SO 2 will become SO3 consuming O2 at the zone between the bricks and the shell. b) Contributing with the minimal possible amount of Cl and S, by entrances or placing a by-pass. c) Impregnating the shell with impermeable to the gas materials and heat resistant. d) Lowering the burning temperature in order to increase the chloride drain, improving the burning coefficient or raw meal burnability or adding mineralizers.




Page 128

5.4. Ovalness and kilns alignment The nomenclature presented in the annexed diagram will facilitate the exposition of ideas in this document. If the tube kiln temperatures before (T1) and after (T2) the rolling hoop are too different (due to variation on the coating or coat thickness) it is produced what it is called thermal crankshaft. The tube kiln temperature is measured with a radiation pyrometer The rolling ring temperature is measured by a contact thermometer or gun (the emisivity of the material to which the temperature will be measured must be verified). Ovalness calculation: Absolute W (in mm) = (A– B)/2 Relative W = 100 * Wab/(mm) % Max admissible = (m)/10 % Critical = 1.2 * Admissible B

A

Displacement estimation: Normal (in mm) = 2 * (in m) Warning maximum (in mm) = 4 * (in m) Warning minimum (in mm) = 0.5 * (in m) The kiln tube ovalness is measured indirectly, based on the rolling hoop displacement (the rotation movement between the kiln tube and the rolling hoop, in a kiln turn), and it is measured as the period length of the drawn curve (see instrument). Therefore, for a 4.55 m kiln the displacement must be around 9.1 mm (not being worrisome between 3 and 20 mm). Also, the maximum ovalness for this kiln would be of 0.46% and the crictical of 0.54%.

Among the possible causes of a limited displacement (< 3 mm) are, a wrong lubrication or kiln tube constriction (strangulation).




Page 129

The strangulation is presented when the tube temperature under the hoop is too high because to the coat or coating lost, expanding the tube until lockin it to the hoop. Ther permanent deformation by ovalness can be corrected changing the shell sections. An alternative for the displacement measurement is the measurement with chalk, consisting in drawing two marks overlapped at the shell and the hoop, and to measure the distance between them through a specific revolution times, so the displacement will be: Distance between marks / kiln turns. 5.4.1 Kiln alignment A recommendation to avoid kiln alignment problems is that the supports must be always in line, whitout horizontal or vertical divergences, which give irregular loads on the supports and tensions, as in the kiln tube as in its rollings, which are transmitted to the refractory coat affecting its duration and performance. An amplified view of these divergences on the kiln would be:

The base for a divergences adequated measurement consists on the equipment usage as the theodolite with laser beam, and trangulations that will allow calculating the needed position values.

These problems can be diagnosed by the temperature of the rollers bearings or by clear spaces between roller and rim (rolling hoop).




Page 130

5.5. Prolonged shut down of an installation (greater than 2 months) While shutting down 

Programmed shut down leaving the installation empty (material silos, air slides and covered conveyors since the uncover conveyors are left with material so it will protect them), as elevators and bucket transporters or chain ones.



Cleaning the IDF’s (induced draft fans) vanes of the kiln due a thermal clash, since they are hot.



It is recommended that while shutting down the kiln stays in an “up” position (the tube must be moved towards the raw meal feeding), so the kiln closure would be as sealed as possible.



Important: The shut down must be help for kiln alignment studies, shell thickness, turbines balance, etc.

During the shut down 

Covering the kiln supports with a phosphoric acid layer in order to avoid rusting. Stopping cooling.



Covering those motors that will not be revised, with plastics, and keep their heating resistance.



Pressurizing the transformers and circuit breakers that will be out of service, taking an oil sample in order to compare it with the one just before starting again.



Take away the T.V. camera, pyrometers and infrareds.



Avoiding dust and rodents in the MCC’s (Motor Control Centers), control room and other installations that will be out of service, backing up and saving the PLC’s memories.



Emptying well all the fuel conducts.



Closing well all the manholes, sluices, kiln outlets, etc., as the satellite entrances.



Loosening the transporter conveyours.



Keeping the filter with heating resistances (avoiding condensations) and starting up the fans and stricking systems every 15 days.



Rotating the kiln 2 or 3 turns every 15 days, always leaving it at 120° from the last position, avoiding it “goes down”.



Moving the transporters whitout material for some hours every 15 days.



Recirculate the fuel oil for 2 to 3 hours every 15 days.



Measuring the ground voltage in big motors.



Disconnecting the gas analyzers.



Calibrating the proportioner scales. If they have dolomitical material cover them with special oils (ask the supplier) or leave the coating on them.

Previous to the start up 

Tests without loads must be done through all the installation. Make sure that the air sliders will not have coatings on its canvas and start up the fans.



The T.V. must be reconnected, analyzers, pyrometers, infrareds, etc.




5.6. General repairs protocol. a) In order to make them we must have previous information: 1. Wear consumptions in cm/million of clinker tons., for:  Kiln tube: meter by linear meter for the brick  Satellite tube: according to zones (entrance, reel, elbow, screen, brick)  Grates cooler: vault, ceiling, walls, stools, arcs.

And in habitual durations in clinker tons for:  Entrances  Entrance cone  Exit cone  Back end And in last thicknesses and shell temperatures for:  Smoke chamber  Cyclones

Page 131




Page 132

2. Refractories drawing of all the kiln parts, with it maximum, last and current duration, in clinker tons, and the placed materials (qualities).

3. Information about the thicknesses, originals for the new ones placed in the last repair, as the thicknesses left in the kiln zones that were not repaired and also those that were brought down (in order to know its habitual wear).




Page 133

4. Coatings disposition and habitual rings, coupling them to the programmed scans with the scanner.

5. Scanning day by day during the campaign, using the last data to locate the maximum temperatures that had been observed at each kiln point, therefore getting and idea of the possible refractory harms by zones (considering for how long it has been placed at each zone). The comparison of the scanner data from the beginning of the campaign with the current data has a special importance.

6. Every information of events in campaign, as design problems (raw meal modules, fuel changes), operational (low production, airs, etc.) or mechanical (ovalness due excessive migrations, hoops and rollers heating, etc.)




Page 134

7. Information taken from inspection at some occasional shut down (for example the inspection of the entrances and satellites at a short shutdown, or rings visualization, burners, arcs, ceiling, ducts, gates, etc.) 8. Information of the refractory ammount, by qualities, and a catalogue of the different options.

b) With this information the kiln situation is estimated (thicknesses wanted) and at which points more refractory must be replaced. If the kiln has not shown refractory failures, when and where a failure will come out can be estimated, which results very important in order to go ahead of the material necessities, labor and machinery to confront the future repair. Therefore the steps are: 1. Estimating zones to repair and ammounts. 2. Comparing those necessities with the present inventory and generating the corresponding orders. 3. Defining when it will shut down and warning with anticipation the contractors for demolition and mounting. c) With all the previous establish a repair plan. The plan will contemplate the activities sequence and with what resources it will be done, also their durations. Especially it must be specified when the inspections at the cooler, tube and tower, will be done, since the main modification of the plan depend on them. Also it has to be specified how the cooling will be done (annexed to the plan), and what securities must be followed (annexed to the plan). The heating (after the repair) will be planned later.




Page 135

As the kiln repair is not only of refractory, but usually also mechanical (mainly at general repairs), it must be defined in the plan, by Production when the Maintenance department would be able to work with motors and conduits, since can affect the ventilation in the demolition works. In an inverse way, the plan will contemplate the mechanical necessities for demolition and mounting (for example, taking out the burner in order to let the mini-excavators get into the kiln tube).




Page 136

The master plan must include the total detail of activities, resources and coordination between areas for a complete repair, trying to eliminate those concepts that in a particual case will not apply, so addition activities to the already listed (and their consequences) will not be considered. d) From the inspections made (visuals, measurements, etcetera), the kiln real situation is defined, as far as zones to be repaired, materials (refractories) and resources (labor and machinery) necessities. When talking about refractory material necessities is not only about the own refractory material (bricks by qualities and positions, concretes and mortars), but of everything surrounding them: anchorings, cardboards, wedging plates, isulators and all the elements to make the repair: scaffolds, cutting discs, neumatical guns and hammers, pointers, mixers, contact and needle vibrators, wheelbarrows, buckets, beaters, entrance molds, cabins, necks, wooden sticks for the segments, hoists, cutter, etcetera.




Page 137

e) Repair following and its costs. f) Registering in the linning plans for the new assemblies and, in general, preparing the necessary documentation (points 1 to 4 from section a) for future repairs, specially the inventory recount. g) It is concluded defining the kiln starting up process in base on the type and ammount of new placed refractory.

5.7 Refractory consumptions and campaign durations The refractory duration, to the margin of mechanical defects and operational irregularities, depends mainly on the thermal load, the prolonged shut downs (mainly greater than 24 hours) and the silica module being used. IN THE KILN TUBE – Predominantly brick a) Silica module and therefore coating thickness C (g/ton) = 200 * | 2.5 – MS | MS>> 2.5, unprotected refractory MS<<2.5, big thickness coating, high weight, coating falling with bricks. Example: MS = 2.8, C = 60 g/ton. b) Shut downs - Greater than 24 hours. C (g/ton) = 20 * NP>24; Where NP>24 is the num. of stops greaters than 24 h. Example: NP>24 = 3, C = 60 g/ton - Smaller, counted as an usage % excepting the shut down time due these shut downs greaters than 24 hours.. C (g/ton) = 7*(100 - % usage) Example: usage % = 90 %, C = 70 g/ton. c) Thermal load (CT); is the main C (g/ton) = 17.5 * CT2.2 Example: CT= 5.0, C = 604 g/ton Note: These consumptions are evaluated as a mean of 2 years in order to compare them with the real average consumptions (biannual).




Page 138

IN THE REST – Predominantly concrete. The refractory consumption at the cyclones towers and the cooler is close to 100 g/ton TOTAL The refractory total consumption by produced clinker ton, may be expressed as: C (g/ton) = 17.5 * CT2.2 + 200 * | 2.5 – MS | + 20 * NP>24 + 7*(100 - % usage) + 100 CAMPAIGN DURATION The campaign duration may be expressed in function of the most important factor, the thermal load (CT) due to the main burner. Duration (months) ≈ 770 * CT

-2.64

Remebering other expressions already given: C (g/ton) = 28 x 2 Duration (days) = 5840 / 2




Page 139

Part 6- Emissions Although in this section the emissions to the atmosphere subject will be intensely seen, also a global view of the points that must be taken care of when pollution is being considered, remembering that the applicable norm must be fulfilled at each case.

a) Pollution signs in materials: ppm in: As (Arsenic) Be (Beryllium) Cd (Cadmium) Cr (Chrome) Hg (Mercury) Pb (Plumb) Se (Selenium) Te (Tellurium) Tl (Thallium) V (Vanadium) Ni (Nickel)

Raw Meal 10 – 50 <1 0.5 – 2 20 – 50 <1 5 – 10 <1 5 – 10 <1 40 – 60 10 – 20

Clinker 10 – 50 <1 <1 50 – 75 <1 5 – 10 <1 5 – 10 <1 60 – 80 20 – 40

Dust in filter 5 – 20 <1 5 – 10 20 – 40 1–5 100 – 500 5 – 10 10 – 20 20 – 100 20 – 40 10 – 20

By the resident time and the reached temperature at the cement kilns, the fixation percentage of these materials to the clinker is superior to 99.5% (superior to the incinerators), excepting for the mercury (Hg) which can be fixated to the clinker in a 60%. b) Poured to the water: In the final   



pouring water to the affluents must be evaluated: Temperature pH Hardness (content in CaCO3 and MgCO3, evaluated as CaCO3 equivalent to CaO) i. A french degree = 10 mg CaCO3 equivalent / liter ii. A german degree = 10 mg CaO equivalent / liter iii. A amercian degree = 1 mg CaCO3 equivalent / liter iv. 1° french = 0.56° german = 10° american DBO and DQO




Page 140

DBO is the biological demand of oxigen. Are the mg O 2/liter that the bacterial poblation uses in a time (normalized to 5 days, this is, DBO5) in order to degrade the organic material (biodegradable). DQO is the chemical demand of oxigen. Are the mg O 2/liter that through chemical process (dichromates) are consumed to oxidate the total organic material (biodegradable or not). Therefore: DQO – DBO = not biodegradable organic (to control in order to keep life in waters). Measurement of how much the TOC (total organic carbon) has been displaced as a measurement of the not biodegradable one, since the TOC (CO2 measurement that is lost when burning the organic material) does not give the oxigen necessity of biodegradating. c) Atmospheric pollution. Emissions: These emissions are divided in:  Particuls (dust): that affects the picking up of solar radiation and the raining formation.  Gases: CO, CO2, SO2, SO3, NOx that affect the global temperature and the acid rain formation. Added to these are the hydrofluoric acid and the hydrochlorate acid.  Smokes: heavy metals, dioxins, furans, etcetera, which affects the human being since they are toxics.  Vapors: affecting as smells. The bad smell, even it does not affect the health, it is considerated a polluting agent. The human being breathes approximately 20 m 3 air/day, and we must realize that nature produce pollution by itself:  Particles emissions: by volcanos, meteorites, fires, etc. The nature emits 3 millions tons of dust by year.  CO2 emission: the living beings emit 1,500 millions tons by day of CO 2; because of their breathing (just the human poblation emits 2,000,000 m3/day of CO2.) Despite these, the caring that the industry must have with the pollution emissions does not have to be minimized, due the effects that are listed next: Global heating. Greenhouse effect The greenhouse effect is inevitable, as we will se next, for the planet life, but It has to be controlled under certain limits. Thanks to the atmosphere, and in particular to the ozone (O3), CO2 y H2O, just the half of the solar radiation (emited as ultraviolet) comes to the Earth, being absorbed mainly by the seas, that will be giving off the heat gradually and as an infrared way. This infrared radiation is retained by the gases (CO2 and H2O) so a mean earthly temperature could be maintained of approximately 15°C, when whitout this greenhouse effect it would be –18°C.




Code DP-03-1

Page 141

The ozone layer disappearance (O3) is more bonded to its consumption (to give O 2) by absortion of solar radiation at the high Sun energization cycles, although the sprays actions also affects to its disminution. The current overheating of the Earth it is bonded to the CO 2 increases in the atmosphere (mainly by fuels) and by the apparition of this in compounds of a higher greenhouse effect as the methane (CH 4) and the CFC (thousand times more powerfull than the CO 2 in greenhouse effect). While the global temperature has changed a lot in its history through long periodos (thousands of years) due to variation on its orbit, now the heating is very noticible, around 0.6°C/100 years, causing (whitout really knowing the cause) defrostings at polar zones, with the risk of water levels increase and creating deserts. Year

ppm CO2

TEarth (°C)

 Sea level (m)

1850 (reference)

295

0

0

1900

297

0.02

0.00

1950

308

0.10

0.00

2000

350

0.46

0.25

Increasing 0.1°C by each increment of 10 ppm in the CO 2 at the atmosphere. Although in nature and in the industry the main contributor of CO2 is the carbon combustion (fuels), for the case of clinker fabrication it is the decarbonation, therefore the “Ca” contribution by other mean but the carbonate is well received, as technically as the pollution aspect. Fuel contribution Type

Kg CO2/Gcal

Fuel oil

19.3

Gas

14.0

Coal

24.0

Coke

25.0

Alternate CAS

≈25

Alternate CAL

≈25

Carbonate contribution bye the raw meal Approximated method in function of the lost by ignition or by fire. (LxI or Lf)

LxI 100  LxI

 kgCO2   kgClk  

Example: Kiln that produces 2100 tpd of clinker consuming 790 kcal/kg, using 100% coke and feeding raw meal with a LxI = 34%.

2100  1000  790  1659 1000000 kgCO 2 / day  41.475; Tpd CO2  41

Gcal / day 

34 kgCO 2  0.515 2100  1081 kgClk 100  34 Total 1122 TpdCO 2



Acid rain: NOX and SOX




Page 142

HNO3 y H2SO4.

NOX: NO, NO2 N2O (in kiln gas 90%NO and 10%NO2)... evaluated as NO. Assuming that in the used raw materials the nitrogen is almost insignificant, the nitrogen contributors are, by importance order: 

The combustive agent (air): approximately 79% of the atmospheric air is molecular Nitrogen (N 2.)



The fuel: nitrogen oxides (NO, N2O, NO2), ammonia (NH3) and cyanided compounds (CN-) are present in fuels. Thus the coal has around 1.5% of N, being the order from higher to fewer contributors: coke, coals, fuel oil, and gas.

Therefore there are two kinds of NOX: Thermal or Zeldovich NOX, due the air N2, high temperatures, the permanency time at this temperatures, oxigen excess and residency time: At a very high temperature N2 + O2

 2NO (90% NO y 10% NO2) T °C

ppm NO

1500

 1,000  10,000  20,000  100,000

2000 3300 4100

Note: As they also discompose, and at a higher speed are formed, the values down a lot in respect of the table ones, which are formation concentrations (  200°C of flame   400 ppm.) Respect the residence time, in the kiln is of the order of 1 sec.

ppm

1800°C

1600°C

1,000 1400°C 100

Residence time at each temperature 1 sec

10 sec




Page 143

The thermal NOX is affected mainly by the excess of O 2. An increment of 0.5% in O 2 increases more of 100 ppm of NOX. The water injection by the burner makes that the steam expels the O 2 at the bottom of the flame; lowering the NOX (it is enough 700 liters/hr, which represent 5 kcals/kg.) So the NOX will be influenced by the flame form, being the flame extension (long or short) and its intensity (strong or weak) which with the excess O 2 determine the NOX level at the kiln tube (the burners of low primary air help not contributing so much O 2 with the fuel). Outside the tube, at the calciner there is no temperature for the formation of thermal NO X but there is for the formation of the fuel one. Fuel NOX, it has a less importance than the one in the main burner, but fundamental in the calciner, being this one a point where the escape gas NO X is tried to lower by “low NOX designs” in which a little amount of sub-stoichiometric fuel is injected in order to even consuming the excess O 2 from the kiln tube, it stays with high CO concentrations which increase when reducing NO X to N2, for then find that the rest of the fuel and tertiary air are enough for a complete combustion. 1000 1200  C NOX + CO     N2 + CO2

70% comb.

Note: the fact of having a calciner will lower the NOX in the escape gas needing less combustion at the main (which generates thermal NOX).

30% comb.

If there is no reduction calciner, also it has been reduced the NO X to N2 by ammonia contribution (NH3) in a tower zone where there are temperatures of 900 to 1000°C. By each 100 mg of NH 3 that is being contributed by Nm3 of gas, it can lower 7 mg NOX/Nm3. The treatments against the Nox emissions can be of two kinds mainly: Catalizador 4NH3+4NO+O2 4N2+6H2O 1. Ammonia reduction in escape gas 350°C it is reduced in a 90% the NOx level proportioning:

NH 3  1.05  1.10 NO

, no more since the emission of NH3 rises

2. Urea addition CO(NH2)2 at smokes chamber, since it needs from 800 to 1000°C to reduce the NOx, since at lower temperatures the urea oxidizes generating a higher NOx emission. The presence of a sulphuer cycle can cause ammonium bisulphate (NH 4HSO3) as incoatingations. 3. Adding water to the flame, reducing its temperature. However the installation is, the NOx level in the escape gas must not be confused with the fixed on by norm, since the last one is fixed for a exit chimney gas, where it has already been dilution by other currents (false air, mill gas, etc.) no hay que confundir el nivel de NO X en el gas de escape con el fijado por la normativa, pues este último lo fija para gas de salida de chimenea, donde ya ha habido dilución por otras corrientes (aire falso, gas de molino, etcétera) that is why the norm forces to reference at 10% of O2.




Page 144

For example:

Conversion to 10% O2: f10 = ppm NOX

10%O2

21  10 21  10 = = 1.375 21  O 2 21  13 = f10 * ppm NOX ( to compare with norm)

SOX: SO2 and SO3 Given the clinkering process, in which mainly calcic carbonate is used, is too difficult to have important concentrations of SO2 and/or SO3, since these are captured by the lime, to become at the mass currents sulphates (CaSO4) and sulfites (CaSO3). That is why the SO2 and/or SO3 emissions are ussualy ZERO, excepting that sulfides (S=) or sulfurs (S) are being contributed by the raw meal, that for being fed at the beginning of the tower, it does not have temperature conditions and time to become sulphates or sulfites.

If SO2 in escape gas is present may force to inject free lime as a calcic hydroxid (approximately 2 to 14 g. Ca(OH2) / kg clinker) so the SO2 will be captured by the CaO and become sulfite.




Page 145

Just as the NOX, the evaluation of SO2 / SO3 it is made on the chimney gas, correcting at 10% O2. Heavy metals. The heavy metals concentrations at the chimney gas (expressed at 11% O2) are of the next magnitude order, being normed the sume levels of them in groups: Metal Cd (Cadmium) Hg (Mercury) Tl (Thallium)

mg/Nm3 < a... 0.0001 0.001 0.003

Ni (Nickel) As (Arsenic) Se (Selenium) Co (Cobalt)

0.001 0.005 0.005 0.0005

Pb (Plumb) Cr (Chrome) Cu (Copper) Zn (Zinc) Sb (Antimony) Mn (Manganese) V (Vanadium)

0.008 0.02 0.002 0.05 0.005 0.01 0.01

Limitation of mg/Nm3 according european norm

Cd + Hg + Tl < 0.2

Ni + As + Se + Co < 1.0

Pb + Cr + Cu + Zn + Sb + Mn + V < 5 Dioxins y furans Are denominated dioxins a big family of derivatives of polichlorinedibenzene-para-dioxin: Cl

Cl

O O

O O

And are denominated furans a big family of derivatives of polichlorinedibenzenefuran:

Cl

Cl

O O

O

Both, dioxins and furans, are evaluated altogheter and are expressen in base to factors of international equivalence (I-TEQ), being the amounts so small that they are expressend in nanograms (1 ng = 10 -9 g = 10-6 mg) or in picograms (1 pg = 10 -12 g = 10-9 mg.) These cancerigenic products are found in dirt, water and atmosphere, due to forest fires, volcanical eruptions, photolityc reaction, even in the blood and grease of the human beings, being the feeding ingestion process the one the contributes more, around 120 pg I-TEQ/day, the OMS establishes a tolerable value of 1 to 4 pg by each kg of corporal weight and day.




Page 146

Respecting the emissions caused by human activity, more of the 75% are produced by the metals fabrication and processing, the metallurgy, followed by a 15% of energy production, where are included the incinerators even these destroy more than they emit, the 5% of domestic heating and the 2% or urban traffic, being just the 0.3% the contribution of other industrial thermal processes on which the cement sector will be. With the PCB’s elimination (chlorinebenzenes in transformators) only the kiln combustion throws dioxins and furans, but in a minimum concentration ( 0.005 ng I-TEQ/Nm 3 at 10% O2) respecting the normative limit (depends on the country but it is around 0.1 ng/Nm 3) so the cement kiln can not be considered as a polluting agent by dioxins and furans, but otherwise, because of the elevated working temperatures (1800°C when to destroy them it is required 800 to 1000°C), the resident times (higher than 2 seconds) as well as the good combustion and the oxigen excess, are excellent places for the destruction of toxical and dangerous residues, even more than the incinerators, on which these conditions are not reached Particules emission. The particules in suspension cause distrosions on the solar radiation pick up, on the rain cycles and the live beings breathing. Naturally the dust displacement by the own Nature is huge, even more than the human factor, establishing, by the norm, particules emission levels and chimneys dimensions in order to avoid the concentration on the urban poblations. The allowed levels (in mg/Nm 3) by each country differ a lot, and even more in kiln chimneys where the exigent is not only limited to punctual emissions by to accumulative periods, through calculations that is considered by the process idiosyncrasy. On the same way there are different formularions for the minimum chimney height,which assures its pollution dispersing work, being one of them the following (must be calculate the height H for each pollution agent, so the result will be the final H for the chimney focal point ): H=

A * Q*F 3 n * CM V * T

Where: H

= Chimney height (m)

A

= Climatic conditions of the place = 70 * Ic where Ic = climatic coefficient (with values from the last 10 years) =

T1  2 * t 80  Tm h

where: T1 =

T maximum at a warm period – T minimum at a cold period (°C)

t = T warmest month – T coldest month (°C) Tm = T year mean (if its lower than 10°C, then 10°C is considered) h = relative mean humidity of June, July, August and September, taken at 07, 13 y 18 hours. Q

= Maximum flow of the mentioned pollution agent, expressed en kg/hr.

F

= Sedimentation coefficient, for gases F = 1, for particules F = 2.

CM

= Maximum admissible concentration for the mentioned pollution agent, in mg/Nm 3, as:




Page 147

CM = C reference – C background where: C reference = Reference value contributed by the Administration, Ex.: particules=0.3 mg/Nm 3 C backgroudn = Background concentration (if there is any) at the locality (mg/Nm 3) n

= Chimney number, included the one that is being calculated, located at the inferior horizontal distance of 2*H of the chimney location.

V

= Gases emited flow, in m3/hr.

T

= Chimney gas T – annual mean T of the air at the place (°C)

Besides the height the chimney must have a minimum convection impulse, in order to fulfill: T2 > 188 *

v2 * s H2

Where: T2

= gas chimney T – mean T of the maximums of the warmest month (°C)

v

= Chimney gases speed (m/seg)

H

= Chimney height (m)

s

= Chimney minimun interior section (m 2)

The big industrial installations, which emit great amount of contaminants to the atmosphere, required also deep studies based in diffusion models, or even scale models where the emission volumes, topographic and climatic conditions are reproduced. To finalize, it is advisable to discriminate the information given by the environmental authorities about flows (by which the pollution agents will be divided), since they can have direct measurement mistakes that easily are detected with simple O2 balances, like this one:




Page 148

Note: to convert ppm to mg/Nm3 it is just neede to multiply by the gas density in kg/Nm 3, so: gas * ppm pollution agent = mg pollution agent/Nm3

gas

These balances allow verifying flows handled by fans and relate them with pressures, having by this way a fast flow estimate with the panel instrumentation. With the gas altitude and temperature, its pass will be obtained in m3/hr.




Page 149

Appendix A Part I. Potential composition statement by Bogue

General Strategy for the stoichiometric statement: 1. Knowing which compounds (phases) will be formed. 2. In what order those mentioned compounds will be formed. 3. Reactions statements: Example: x A + y B  AxBy 3.1 From the product AxBy will be forme as many amount as the minority allows. ¿Which one will be the minority? It will be A, if A/B is lower than

x  PmA ; otherwise B will be the minority. y  PmB

3.2 Knowing the minority, for example B, formations and consumptions are calculated. xA

+

yB



AxBy

xPmA

+

yPmB

=

PmAxBy

Aconsumed

Aconsumed= Bexistent

Bexistent

AxByformed= Bexistent

AxByformed




Code DP-03-1

Page 150

3.3 And what is left for the following reactions: Brest = Bexistent – Bconsumed = 0 x  PmA  Bexistent y  PmB

Arest = Aexistent – Aconsumed = Aexistent -

Portland Type Clinker Case MA > 0.64. C4AF formation. As being the Fe2O3 the minority one with respect to Al2O3 (MA > 0.64), C4AF will be formed as much as the Fe2O3 level allows. 4 CaO + Al2O3 + Fe2O3 = C4AF 4 PmCaO + PmAl2O3 + PmFe2O3 = PmC4AF 4 (56) + 101.96 + 159.77 = 485.66 CaO(Consumed) + Al2O3(Consumed C4AF) + Fe2O3 = C4AF(Formed) C4AF(Formed) =

1  Pm C4AF 1  PmFe2O3

Al2O3(Consumed C4AF) =

* Fe 2 O 3 =

1  Pm Al2O3 1  PmFe2O3

485.66 159.77

* Fe 2 O 3 =

* Fe 2 O 3

101.96 159.77

= 3.04 Fe2O3

* Fe 2 O 3

= 0.64 Fe2O3

Al2O3(Rest C4AF) = Al2O3 - Al2O3(Consumed C4AF) = Al2O3 - 0.64 Fe2O3 CaO(Consumed C4AF) =

4  Pm CaO 1  PmFe2O3

* Fe 2 O 3 =

4  56 159.77

* Fe 2 O 3 =

CaO(Rest C4AF) = CaO – CaO(Consumed C4AF) = CaO - 1.4 Fe2O3

(1)

(2) (3)

1.4 Fe2O3

(4) (5)




Page 151

C3A formation. It will be formed as much as the Al 2O3 % allows (Rest), which will be the minority with respect to CaO. 3 CaO + Al2O3 = C3A 3 56 + 101.96 = 269.96 CaO(Consumed C3A) + Al2O3(Rest C4AF) = C3A(Formrf) C3A(Formed) =

1  Pm C3A 1  Pm Al2O3

* Al 2 O 3(Resta C4AF) = 2.65 * Al2O3(Rest C4AF)

= 2.65 * (Al2O3 - 0.64 Fe2O3) = 2.65 Al2O3 – 1.69 Fe2O3 CaO(Consumed C3A) =

3  Pm CaO 1  Pm Al2O3

(6)

* Al 2 O 3(Rest C4AF) = 1.65 * Al2O3(Rest C4AF)

= 1.65 * (Al2O3 - 0.64 Fe2O3) = 1.65 Al2O3 – 1.05 Fe2O3

(7)

CaO(Rest C3A) = CaO – CaO(Consumed C4AF) – CaO(Consumed C3A) = CaO – 1.4 Fe2O3 – (1.65 Al2O3 – 1.05 Fe2O3 ) = CaO – 1.65 Al2O3 – 0.35 Fe2O3

(8)

C2S formation. It will be formed as much as the minority compound allows, in this case the SiO 2. 2 CaO + SiO2 = C2S 2 (56) + 60.09 = 172.09 CaO(Consumed C2S) + SiO2 = C2S(Formed) C2S(Formed) =

1  Pm C2S 1  Pm SiO2

CaO(Consumed C2S) =

* SiO 2 = 2.86 SiO2

2  Pm CaO 1  Pm SiO2

* SiO 2 = 1.86 SiO2

(9)

(10)

CaO(Rest C2S) = CaO - CaO(Consumed C4AF) – CaO(Consumed C3A) – CaO(Consumed C2S) = CaO – 1.65 Al2O3 – 0.35 Fe2O3 – 1.86 SiO2

(11)

C3S formation. It will be formed as much as the minority compound allows, which can be the rest C 2S or the CaO. C2S + CaO(Rest C2S) = C3S 172.09 + 56 = 228.09 ¿Which one is the minority?



C2S mols =

CaO mols=

C 2 S (Formed)

=

MwC2S CaO (Rest) MwCaO

=


(9)

(12)

172.09

(11) 56

Page 152

(13)

a) Minority C2S (12) < (13) (Very strange) C2S + CaO = C3S 172.09 + 56 = 228.09 C2S(Formed) + CaO(Consumed C3S) = C3S(Formed) C3S(Formed) =

1  Mw C3S 1  MwC2S

* C 2 S (Formed) = 1.32 * C2S(Formed) = 1.32 * 2.86 SiO2

= 3.8 SiO2 CaO(Consumed C3S) =

(14)

1  MwCaO 1  MwC2S

* C 2 S (Formed) = 0.93 SiO2

(15)

CaO(Final Rest) = CaO - CaO(Consumed C4AF)– CaO(Consumed C3A) – CaO(Consumed C2S)– CaO(Consumed C3S) = CaO – 1.65 Al2O3 – 0.35 Fe2O3 – 2.79 SiO2

(16)

b) Minority CaO (12) > (13) (Normal) C2S + CaO = C3S 172.09 + 56 = 228.09 C2S(Formed) + CaO(Rest) = C3S(Formed) C3S(Formed) =

1  MwC3S 1  MwCaO

* CaO (Rest C2S) = 4.07 * (equation 11)

= 4.07 CaO – 6.7 Al2O3 – 1.43 Fe2O3 – 7.6 SiO2 C2S(Consumed) =

1  MwC2S 1  MwCaO

(17)

* CaO (Rest C2S) = 3.07 * (equation 11)

= 3.07 CaO – 5.06 Al2O3 – 1.08 Fe2O3 – 5.70 SiO2

(18)

C2S(Rest) = C2S(Formed) – C2S(Consumed) = 8.56 SiO2 - 3.07 CaO + 5.06 Al2O3 + 1.08 Fe2O3

(19)

Since the kiln process is statistical of encounters between compounds, CaO (L) will always be formed, therefore the Bogue expressions are corrected, using (CaO – CaO (L)) and the calcium sulphates, which are formed with the remainding SO 3 of its combination with alkali (K y Na), consuming again CaO and the Bogue expressions are again corrected.




Page 153

SO3 + K2O = K2SO4 80 + 94.2 = 174.2 K2SO4 = (174.2/94.2) * K2O = 1.85 K2O SO3 + Na2O = Na2SO4 80 + 62 = 142 Na2SO4 = (142/62) * Na2O = 2.29 * Na2O SO3(Rest) = SO3 – (80/94.2) * K2O – (80/62) * Na2O = SO3 – 0.85 * K2O – 1.3 * Na2O SO3 + CaO = CaSO4 80 + 56 = 136 CaSO4(Formed) = (136/80) * SO3(Rest) = 1.7 SO3(Rest) = (1.7*1) SO3 - (1.7*0.85) K2O - (1.7*1.3) Na2O = 1.7 SO3 – 1.44 K2O – 2.2 Na2O CaO(Consumed SO3) = (56/80)*SO3(Rest) = (56/80) SO3–(56/80)*(0.85) K2O– (56/80)*(1.3) Na2O = 0.7 SO3 – 0.6 K2O – 0.9 Na2O = 0.7 * (SO3 - 0.85 K2O – 1.3 Na2O) Finally correcting the equation 17 free lime and consumed lime by SO3 we get the following form: C3S = 4.07[CaO – CaO(L) – CaO(Consumed SO3)] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 C3S = 4.07[CaO – CaO(L) – 0.7 (SO3 - 0.6 K2O - 0.9 Na2O)] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 That algon with the MgO (Up to 2%), givien as magnesium equivalence with respect to the calcium at the phases (C3S + C2MgS + CMg2S + Mg3S), of 0.75, the C3S Bogue expression is: C3S = 4.07[CaO – CaO(L) + 0.75 MgO – 0.7 SO 3 + 0.476 K2O + 0.72 Na2O] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 c) Both equal (12) = (13) C2S + CaO = C3S 56*2 + 60.09 + 56 = 228.09 C3S = (228.09/60.09) * SiO2 =3.8 SiO2 CaOconsumed C3S = (168/60.09) SiO2 = 2.8 SiO2 C3A = 2.65 Al2O3 – 1.69 Fe2O3 C4AF = 3.04 Fe2O3 C2S = 0 CaO(L) = CaO – CaORestC3A – CaOConsumed C3S = 0 CaO(L) = CaO – 2.8 SiO2 – 1.65 Al2O3 – 0.35 Fe2O3 = 0 CaOOptimal = 2.8 SiO2 + 1.65 Al2O3 + 0.35 Fe2O3




Page 154

Defining saturation degree (SAT.D) as the optimal CaO approximation: CaO

SAT.D = CaO

 100 

Optimal

CaO  100 2.8 SiO 2  1.65 Al 2 O 3  0.35 Fe 2 O 3

As not being pure phases, the founded coefficients, based in experience, change giving place to the LSF commonly known. LSF =

CaO  100 2.8 SiO 2  1.18 Al2 O 3  0.65 Fe 2 O 3

MA Case < 0.64 With the same statement but beginning with the C 4AF until the Al2O3 is consumed and then the C2F unitl the Fe2O3 does, the following Bogue expressions are defined: C4AF = 4.77 Al2O3 C2F = 1.7 Fe2O3 – 2.67 Al2O3 C2S = 2.867 SiO2 – 0.754 C3S C3S = 4.071 CaO – 7.6 SiO2 – 4.48 Al2O3 – 2.86 Fe2O3 Note: If C3S<0; C3S = 5.8 SiO2 and C2S is recalculated. For liquid phase percentage the expressions are as following: FL1338 °C = 6.1 Fe2O3 + MgO + K2O + Na2O + SO3 For MA>1.38 = 8.2 Al2O3 – 5.22 Fe2O3 + MgO* + K2O + Na2O + SO3; For MA<1.38 FL1400 °C =2.95 Al2O3 + 2.2 Fe2O3 + MgO* + K2O + Na2O + SO3; For MA>0.64 FL1450 °C =3.05 Al2O3 + 2.25 Fe2O3 + MgO* + K2O + Na2O + SO3; For MA>0.64 Note: MgO* = MgO up to 2%. Expressed as phases FL1450 °C = 1.13 C3A + 0.75 C4AF + 0.6(CaSO4 + K2SO4 + Na2SO4) Respect to the minimum clinker temperature T(°C) = 1300 + 4.51 C3S – 3.74 C3A – 12.64 C4AF Sulphates Module MS 

(SO 3 /80)

 (K 2 O/94)  (Na 2 O/62)  (2Cl /35.5)

Equivalent Alkali Equivalent Na2O = Na2O + 0.659 K2O Equivalent K2O = K2O + 1.52 Na2O




Page 155

Reaction heat (ΔH) Kcal/Kgclinker ΔHZur Strasse = 7.65 CaO + 6.48 MgO + 4.11 Al2O3 – 0.59 Fe2O3 – 5.12 SiO2 ΔHFLS = 7.65 CaO + 6.48 MgO + 2.22 Al2O3 – 0.597 Fe2O3 – 5.12 SiO2 – 10(K2O + Na2O) + 11.6 H2Oraw meal Note : Oxides in clinker base Coatingability coefficient (AW) AW = C4AF + C3A + 0.2 C2S + 2 Fe2O3 Burning FactorAptitud de cocción (BF) BF = LSF + 10 MS – 3(MgO + K2O + Na2O)

A.2 – “Tentative” statement for potential composition with CaF 2 intervention, applied to clinkers of MA > 0.64 and supossed fluorestelladite composition (3C2S•3CaSO4•CaF2) K2SO4 formation K2O + SO3 = K2SO4 94.2 + 80 = 174.2 K2SO4 = 1.85 K2O SO3(consumed)= (80/94.2) K2O = 0.85 K2O SO3(rest)= SO3 – 0.85 K2O Na2SO4 formation Na2O + SO3 = Na2SO4 62 + 80 = 142 Na2SO4 = 2.29 Na2O SO3(consumed)= (80/62) Na2O = 1.29 Na2O SO3(rest)= SO3 – 0.85 K2O – 1.29 Na2O CaSO4 formation CaO + SO3 = CaSO4 56 + 80 = 136 CaSO4 = 1.7 SO3(Rest) = 1.7 (SO3-– 0.85 K2O – 1.29 Na2O) = 1.7 SO3– 1.445 K2O – 2.2 Na2O CaO(Consumed) = 0.7 SO3 – 0.6 K2O – 0.9 Na2O CaO(Rest) = CaO – (0.7 SO3 – 0.6 K2O – 0.9 Na2O)




Page 156

C4AF formation 4 CaO + Al2O3 + Fe2O3 = C4AF 4 (56) + 102 + 160 = 486 C4AF = (486/160) Fe2O3 = 3.04 Fe2O3 CaO(Consumed) = [(4*56)/160] Fe2O3 = 1.4 Fe2O3 Al2O3(Consumed) = 0.638 Fe2O3 CaO(Rest) = CaO – 0.7 SO3 – 0.6 K2O – 0.9 Na2O – 1.4 Fe2O3 Al2O3(Rest) = Al2O3 – 0.638 Fe2O3 C3A formation 3 CaO + Al2O3 = C3A 3 (56) + 102 = 270 C3A = 2.647 Al2O3(Rest) =2.647 * (Al2O3 – 0.638 Fe2O3) = 2.647 * Al2O3 – 1.69 * Fe2O3 CaO(Rest) = CaO – 1.65 Al2O3 – 0.35 Fe2O3 – (0.7 SO3 – 0.6 K2O – 0.9 Na2O) C2S formation 2 CaO + SiO2 = C2S 2 (56) + 60.09 = 172.09 C2S = 2.86 SiO2 CaO(Rest) = CaO – 1.86 SiO2 – 1.65 Al2O3 – 0.35 Fe2O3 – (0.7 SO3 – 0.6 K2O – 0.9 Na2O) (3C2S ∙ 3CaSO4 ∙ CaF2) formation: A little amount of CaF2 (Molcular weight 78) is added with products which are inevitably formed as C 2S (Molecular weight 172) and the CaSO 4 (Molecular weight 136) in order to give a big amount of a 3C2S∙3CaSO4∙CaF2 flux (Molecular weight 1002) 3C2S + 3CaSO4 + CaF2 = 3C2S ∙ 3CaSO4 ∙ CaF2 3 (172) + 3 (136) + 78 = 1002 a) Minority determination CaF2 = (78/408) * CaSO4 = 0.19 CaSO4 = 0.19 (1.7 SO3 – 1.445 K2O – 2.2 Na2O) CaF2 = (0.323 SO3 – 0.274 K2O – 0.418 Na2O) F





CaF2 2.05

= 0.1583 SO3 – 0.1346 K2O – 0.2 Na2O




Page 157

b) If CaF2 is the minority F- < 0.1583 SO3 – 0.1346 K2O – 0.2 Na2O; F





CaF2 2.05

In this case, 3C2S∙3CaSO4∙CaF2 = 12.86 CaF2= 26.37 FCaSO4 (Consumed) = (408/1002) * 3C2S∙3CaSO4∙CaF2 = 0.40 * 3C2S∙3CaSO4∙CaF2 CaSO4 (Consumed) = 0.4 * 26.37 F- = 10.73 FCaSO4 (Rest) = 1.7 SO3 – 1.445 K2O – 2.2 Na2O – 10.73 FCaF2 (Rest) = 0 C2S(Consumed) = (516/1002)* 3C2S∙3CaSO4∙CaF2 = 0.515 * 3C2S∙3CaSO4∙CaF2 C2S(Rest) = 2.86 SiO2 - 0.515 * 3C2S∙3CaSO4∙CaF2 c) If CaSO4 is the minority F- > 0.1583 SO3 – 0.1346 K2O – 0.2 Na2O; F





CaF2 2.05

In this case, 3C2S∙3CaSO4∙CaF2 = (1002/408) CaSO4 = 2.45 CaSO4 3C2S∙3CaSO4∙CaF2 = 4.165 SO3 – 3.54 K2O – 5.39 Na2O CaSO4 (Rest) = 0 C2S(Consumed) = (516/1002)* 3C2S∙3CaSO4∙CaF2 = 0.515 (3C2S∙3CaSO4∙CaF2) C2S(Rest) = 2.86 SiO2 - 0.515 * 3C2S∙3CaSO4∙CaF2 Note: CaF2 in balance with CaSO4, si CaF2 = 0.32 SO3 – 0.27 K2O – 0.4 Na2O CaO + C2S = C3S 56

+ 172 = 228

C2S(Consumed) = (172/228) * C3S = 0.754 C3S C3S formation: Same corrected Bogue expression, given that the C 3S is not affected by the C2S consumption by the mineralized pair, lefting always as an excess. C3S = 4.07[CaO – CaO (L) +0.75 MgO* – 0.7 SO3 + 0.476 K2O + 0.72 Na2O] – 7.6 SiO2 – 6.7 Al2O3 – 1.43 Fe2O3 Remaining C2S: Different to the corrected Bogue C2S = 2.86 SiO2 – 0.754 C3S – 0.515 (3C2S∙3CaSO4∙CaF2)




Page 158

In the empirical behavior, this 3C 2S∙3CaSO4∙CaF2 phase is behaved as the C3A but in an lower extension, so in order to include it effect in the modules, minimum sintering temperature, etc., the equivalent in work Al2O3 can be expressed: Al 2 O 3 Equivalent in work 

3C 2 S  3CaSO 4  CaF2 2.65

So, in a tentative way, until the phase’s bibliography using fluorite come out, the modules can be corrected as the kiln would feel them: CaO  100

LSF =

3C 2 S  3CaSO 4  CaF2   2.8 SiO 2  1.18 Al2 O 3     0.65 Fe 2 O 3 2.65

SiO 2 MS =

MA =

Al2 O 3  Al2 O 3 

3C 2 S  3CaSO 4  CaF2 2.65

 Fe 2 O 3

3C 2 S  3CaSO 4  CaF2 2.65 Fe 2 O 3

Clinkering T (°C) = 1300 + 4.51 C3S – 3.74 C3A – 12.64 C4AF – 12 (3C2S∙3CaSO4∙CaF2) Coatingability coefficient (AW) AW = C4AF + C3A + 0.2 C2S + 2 Fe2O3 + 3C2S∙3CaSO4∙CaF2 Burning factor (BF) (With the new modules) BF = LSF + 10 MS – 3(MgO + K2O + Na2O) CaO30T°C = Already seen formulas but using modules calculated with 3C 2S∙3CaSO4∙CaF2. Liquid phase % at 1450 °C FL1450 °C =3.05 [Al2O3 + *

MgO = MgO up to 2 %

3C 2 S  3CaSO 4  CaF2 2.65

] + 2.25 Fe2O3 + MgO* + K2O + Na2O + SO3




Code DP-03-1

Page 159

Example: Mineralizing white clinker 1 (Normal)

2 (Mineralized)

CaO

67.8

68

SiO2

23.1

24

Al2O3

6.2

2.3

Fe2O3

0.28

0.28

MgO

0.83

0.40

SO3

0.89

3.5

K2O

0.1

0.1

Na2O

0.1

0.1

TiO2

0.05

0.05

F-

0

0.85

PxI

0.5

0.4

Sum

99.9

99.9

CaO(L)

0.5

0.5

New Formulas

LSF

93.9

97

89.2

MS

3.56

9.3

3.09

MA

22.1

8.21

26.7

C3S

54.3

67.1

67.1

C2S

25.3

18.2

11.2

C3A

16

5.6

(5.6 +13.7)

C4AF

0.9

0.9

0.9

FL1338°C

3.6

5.8

5.8

FL1450°C

21.15

11.6

27.2

Coating Coeff.

22.4

10.7

23

BF Factor

125.2

186

118

Clinkering T

1475

1571

1406

Resistance 2D

33.3

26.6

Lowers when C3A lowers

Resistance 7D

57

58.2

Equal for same C3S + C3A

Resistance 28D

67

75.1

Raises for higher C3S

Real results in white clinker kiln. The 2 did not behave as corresponded to the modules and traditional calculations, but as the new expressions gaining 10% of production, lowering the specific heat consumption in a 15%, and being a clinker with lower benefits at early ages (with a longer setting), the result at late ages was better and with a more pulverizable clinker.

by 3C2S∙3CaSO4∙CaF2 effect




Page 160

Appendix B Part I. Calibration of FRX and DRX equipments Adjustment of process equipment (FRX) to process control (Automatic or Manual): 1. Simple: Calibration method. 2. Reliable: Handling by menus “Without Error” 1. Method of FRX Calibration (by melted pearl)  SIMPLE A.

Strategy:

a) Having the main 9 channels (Ca, Si, Al, Fe, Mg, S, K, Na, Ti) and if there is just a little of anyone in the raw materials, for example the Ti, a fix value can be left and not use the channel. If there is one very correlated with other, for example the Na with the K, this relation can be fixed, not needing that channel, and materials with sodium can be used as a flux (borate) and as antiadherent (iodide) while making the pearls. b) Working with high dilutions samples vs. flux so:  Melts well  “Linearize” the curves which cause the interelementary matrix effects.  Keep lineal the losses of volatile. The most effective is the dilution at 10%, this is, 0.5 gr of sample: 5 gr of flux. c) Minimizing the number of calibrators, the best is to have only 2 calibrators:  Calibrator for raw meal: built with all the materials which form the raw meal.  Calibrator for Clinker and Cement: built with all the materials which form the different cements. d) Rationalized Patterns. They must meet that the sum of the 9 oxides + the lost lost of ignition, gives a value very close to 100, in order to rationalize them definitely to 100%. e) The straight line must be natural. The compounds with “extreme” elements indicate how the natural slope is. Concentration (element %)

Pivot

Previous Strategy. Family curves, using minimum squares resulting in not natural straigth lines.

Realtive Intensity (I)




Code DP-03-1

Page 161

B. Calibration method In both straight lines (for raw meal and cement) and for all the oxides, a point on the line will be THE TARGET PRODUCT (raw meal/clinker) of the Plant, this is, the straight lines WILL PIVOT always on the raw meal analysis, or clinker, that is wished to always be obtained at the mill or kiln. In this way the analysis will very precise since it has accomplished what was looked for, that target raw meal and clinker. The other point with which the straight line is built (each one) will depend on the element, having the material which offers the higher range with respect to the product (raw meal, clinker). Thus: FOR THE RAW MEAL STRAIGHT LINE: Material (PATTERNS)

Oxides Al

Fe

Mg

Reference Raw Meal

X

X

X

X

X

High Limestone

X

Silica Corrector

*

S

Na X

K

Ti

X

X

X

(X)

(X)

(X)

(X) X

Alumina Corrector*

(X) X

Iron Corrector

X

Gypsum

(X)

LOI LOI = 0.786·CaO

Si

+1.08·MgO-0.1·SO3

Ca

(*) Usually the same material (clays.....) (X) the one that is more appropriate When calibrated, all these materials are analyzed by the calibrator and the “rationalized” analyticals are given as definitives (“OPERATIONAL ANALYTICALS”) FOR THE CLINKER AND CEMENT STRAIGHT LINE: Material (PATTERNS)

Oxides Si

Al

Fe

Mg

S

Na

K

Ti

Reference Clinker

X

X

X

X

X

X

X

X

X

Puzolane

X

X

X

Gypsum

X

X X

Others of support: (*) Iron Mineral

X

(*) High Alkali Mineral (*) Even not being own materials of cements.

X

X

LOI 0.5 flow for clinker + LOI due to additions, seek relation LOI - CaO

Ca




Code DP-03-1

Page 162

Adjustment to procces equipmet (DRX) to process control (Automatic or Manual) 1. Simple 2. Reliable 1. DRX simple calibration method (non purist explanation, but for automatization practical use).

Impulsesde los Contaje impulso en counting at Máximo Pico Maximum peak donde it where lo is found

Example: For the C3S at and angle of 2Θ = 41.15 41.05

41.15

41.25

1. 1st Straight Line: drawn with  and  For corrected Bogue: C 3S

1



PIVOT: By dilution C 3S  : C 3 S · 0.4

 Clinker (the closest wished)  Dilutin the clinker  With a 60% of the

2

mayority in cements, grinding All togheter until finesse equal to



Impulses (counts)

2. Certification of cement primary patterns The two more extreme cements made are taken (in clinker %) and are introduced to the straight line, and the resulting values, for the phase, constitute the PRIMARY PATTERNS. The “monitor” intensity will allow the auto correction by progressive exhaustion of the tube by time.

3.

Analysis: the samples to be analyzed will be introcued in the “Primitive Straight Line” (straight line made with the primary patterns), and its values will be corrected by the exhaustion factor (derives from the tube, this is the relation between the counts of the day monitor and the ones when the equipment was calibrated). C 3 S CEMENT  C 3 S By Curve  f exhaustion

f exhaustion 

I monitor ( at calibration) I monitor (today)




Page 163

Note: It is very recommendable to eliminate the background noises mainly by additions, that is why it is important to find a free peaks point which reperesents the background (example: angle 17) to disconnect that background noise at the peak intensity which represents the phase that is going to be evaluated.

I Error situation (by background noise): I(34.2°)80/20 >>> 0.8 *100 I(34.2°)

I(34.2°)100 I(34.2°)80/20

((80%Clk / 20% Poz)

(100%Clk)

angle C3S (34.2°)

I

Corrected situation (by background noise): I(34.2°)80/20 - I(17°)80/20 = 0.8 * (I(34.2°)100 - I(17°)100 ) I(34.2°)100 I(34.2°)80/20

((80%Clk / 20% Poz)

(100%Clk)

angle Background (17°)

C3S (34.2°)




Code DP-03-1

Page 164

And so on for the other angles (phases). The selected peaks must have a enough curvature for the detection (and counting) of the maximum point. Very intense and narrows peaks, are not interesting, facing stability and reliability. In the other hand, the peaks (2 Θ) selected for the phases must be free of interferences by the additions and the gypsum, that is why it is recommended the following angles (for tubes with a copper anode): C3S : 34.2 Y 41.2 C3A : 33.2 C4AF: 33.8 C2S : 30.9 For the free lime (DRX vs. chemical method), the primitive straight line is built with kiln clinkers, taking andvantage of kiln situations of bad burning, for the high point and the objective clinker as the low one. Cx (free lime): 37.3 VERY IMPORTANT NOTE: through time, the equipment will displace the diffractograph, that is why it has to be accompanied with a displacement of the angle that is going to be measured.

41.1

41.2

41.3

41.1

41.2

41.3

41.2

41.3

41.4

“ALSO IT MUST BE TAKEN CARE OF THE PATTERNS RENOVATION” Other commentaries: In case of cements added with limestone, the CaCO 3 quantification by DRX (calibrating previously intensity vs. lost of ignition) allows having LOI of the produced cement without needing the adquisition of other fast evaluation equipment (for example: LECO). Note: CaCO3 at 43.1 / 47.4 / 48.5 / 39.4 angles. For the gypsum case, what refers to quantify the dihydrated, hemhydrated and anhydrite percentage, DRX constitutes a good support for antincipating (before grinding or during) or antincipating for false setting problems. Dihydrated: 11.7 Hemihydrated: 14.7 Anhydrite: 25.8.




Page 165




Code DP-03-1

Page 166

Appendix A Pare II. Combustion Calculations for the stoichiometric air Table 1 FUEL

AIR

CO2

ELEMENT

WEIGHT %

C

86.0

9.818

H2

10.5

3.605

S

3.5

0.1502

TOTAL

100

13.5732

H 2O

SO2

3.154

7.542

10.696

2.765

3.71

0.07

0.1152

0.1852

0.07

10.422

14.59

0.945

3.154

0.945

CARBON COMBUSTION C + O2 Kg C + Kgmol C

CO2

32 Kg O2 Kgmol

44 Kg CO2 Kgmol CO2

Dividing between 12 for reducing in carbon base, we have: 1 +

Total Stoich. gas

TABLE CONSTRUCTION

12

N2

2.667 Kg O2

3.667 Kg CO2

Kg C

Kg C

Multiplying by the percentage in weight of the carbon in the fuel, we find: O.86 Kg C Kgcomb

X 3.667 Kg CO2 = 3.154 Kg CO2 Kg C

Kg comb




Code DP-03-1

Using air instead of O2: O2 +

N2

AIR

% MASS

0.233

0.767

1

O2 BASE

1

3.29

4.2918

C + 2.66 (4.2918) AIR

3.66CO2 + 2.66(3.2918) N2

C + 11.4168 AIR

3.66 CO2 + 8.77 N2

Substituting: 11.4168 X 0.86 = 9.818 Kg air Kg comb. 8.77X 0.86 = 7.542 Kg N2 Kg comb.

HYDROGEN COMBUSTION H2 + 1/2 O2 2

Kg H2 + Kgmol H2

H2O

16 Kg O2

18 Kg H2O

Kgmol O2

Kgmol H2O

Dividing between 2 to reduce in hydrogen base, we get: 1 +

8 Kg O2 Kg H2

9

Kg H2O Kg H2

Page 167




Code DP-03-1

Multiplying by the percentage in weight of hydrogen in the fuel, we find: O.105 Kg H2

X 9.0 Kg H2O = 0.945 Kg H2O

Kgcomb

Kg H2

Kg comb

Using air instead of O2:

O2 +

N2

AIR

% MASS

0.233

0.767

1

O2 BASE

1

3.29

4.2918

H2 + 8 (4.2918) AIR H2 + 34.334 AIR

9 H2O+ 8(3.2918) N2 9 H2O + 26.334 N2

Substituting: 34.334 X 0.105 = 3.605 Kg air Kg comb.

26.334 X 0.105 = 2.765 Kg N2 Kg comb.

SULFUR COMBUSTION S + O2 32

Kg S + Kgmol S

32 Kg O2 Kgmol O2

Dividing between 32 to reduce in sulfur base, we have:

SO2 64 Kg SO2 Kgmol SO2

Page 168


Chapter III: Clinker 1 +

1 Kg O2

2


Page 169

Kg SO2

Kg S

Kg S

Multiplying by the percentage in weight of sulfur in the fuel, we find: 0.035 Kg S

X 2.0 Kg SO2 = 0.07

Kgcomb

Kg S

Kg SO2 Kg comb

Using air instead of O2: O2 +

N2

AIR

% MASS

0.233

0.767

1

O2 BASE

1

3.29

4.2918

S + 1 (4.2918) AIR S + 4.2918 AIR

2 SO2+ 1(3.2918) N2 2 SO2 + 3.2918 N2

Substituting: 4.2918 X 0.035 = 0.1502 Kg air Kg comb. 3.2918 X 0.035 = 0.1152 Kg N2 Kg comb. EXPLANATION OF THE COLUMNS IN TABLE 2 The total of the 3rd column represents the stoichiometric air amount needed to burn 1 kg of fuel oil. While burning the fuel oil transforms totally into gases (CO 2, H2O, etc.), 1 Kg of fuel oil produces 1 Kg of gases. The Kg of gases are added to the 13.57 Kg of stoichiometric air, finally, the total amount of stoichiometric combustion gases (GT) is: table 1, column Total.

 Kg  GT  13.57  1kg fuel  14.57  Gases   kg Fuels 

Appendix B Part II. Heat Balance The heat balance is based on the heat flows of the the different components that leave and enter the system starting off the base Incomes = Outcomes. Next it is presented a table with the calorific capacities of the main compounds based on the next formula:




Page 170

Cp = a + b * 10-6 * T + c * 10-9 * T2 (Kcal/kg °C) (T in °C) A

b

C

CO2

0.196

118

-43

H2O

0.443

39

28

N2

0.244

22

0

O2

0.218

30

0

Air

0.237

23

0

Escape Gas (T in °K)

0.306

58

7.96

Raw Meal

0.206

101

-37

Clinker

0.186

54

0

Coal/Coke

0.262

390

0

Note: The temperatures are always in °C unless other is indicated. Heat losses QLosses = QClinker + QCooler + QTower + QKiln + QDuct 3° + QEscape Gas+ QExcess Air + QDust + QReaction + QCO + QResidual Air Clinker Latent Heat (Kcal/hr) QClinker = mClinker*CpClinker*TClinker Radiation and Convection (use T in Kelvin degrees) A = exterior Areas which will be evaluated in m2. Temperatures in Kelvin degrees T = T °C + 273 Radiation (Kcal/hr m2): hRadiation = 0.000000042*(TSuface4 – TAmbient4) Convection (Kcal/hr m2): hConvection = 80.33*(( TSurface + TAmbient)/2)-0.724 *( TSurface - TAmbient)1.33 Cooler (Kcal/hr) QCooler = ACooler * (hRadiation + hConvection) Cyclones Tower (Kcal/hr) QTower = ATower * (hRadiation + hConvection)




Page 171

Kiln Shell (Kcal/hr) QKiln = AKiln * (hRadiation + hConvection) Tertiary Air Duct (Kcal/hr) QDuct 3° = ADuct 3° * (hRadiation + hConvection) Latent Heat of: Escape Gas (Kcal/hr) (Including Excess Air) QEscape Gas = mEscape Gas*[CpEscape Gas*(1-(%Excess Air/100)) + CpAir*(%Excess Air/100)]*TEscape Gas mEscape Gas (Kg/hr)= ρEscape Gas * VEscape Gas ρEscape Gas (Kg/ Nm3)= 1.4 – (0.005*%O2) (Tower Exit) VEscape Gas (Nm3/hr)= 1.25 * mClinker / (1 - %Excess Air / 100) mClinker in Kg/hr %Excess Air = %O2 (Tower

Exit)

/ 0.21

Excess Air QExcess Air = mExcess Air*CpAir*TEscape Gas (Kcal/hr) mExcess Air = ρAir * VExcess Air ρAir = 1.29 kg/Nm3 -air normal densityVExcess Air = VEscape Gas * %Excess Air /100 Escape Gas (Kcal/hr) (Without Excess Air) QEscape Gas without excess air = QEscape Gas - QExcess Air Dust Latent Heat not recovered by Cold Stages (Kcal/hr) QDust = mDust*CpDust*TDust MDust = mClinker * (HC/Ck) * (1+ %Dust/100) * (%Dust/100) %Dust= % of raw meal that escapes to the cold stages Where (HC/Ck) is the conversion from raw meal to clinker stoichiometrically. Heat reaction (Kcal/hr) QReaction=(7.646*%CaO+6.48*%MgO+2.22*%Al2O3–5.116*%SiO2–0.59*%Fe2O3–10*%K2O+Na2O *mClinker Residual Air Latent Heat(Kcal/hr) QResidual Air = mResidual Air CpAir*TResidual Air mResidual Air = ρAir * VResidual Air VResidual Air = VTotal Air - VTertiary Air + Secondary VTertiary Air + Secondary = VExcess Air + VStoichiometric Air VExcess Air = VEscape Gas * %Excess Air /100

+11.6*%H2O)




Page 172

VStoichiometric Air = 1.09 * CEH / 1000 Note: CEH is the specific kiln consumption iterating from 800 Kcal/kg for gray clinker and 1500 Kcal/kg for white clinker. CO presence at the precalciner exit (Kcal/hr) QCO = 32 * %CO * mClinker Heat Incomes Raw Meal Latent Heat (Kcal/hr) QRaw Meal = mRaw Meal*CpRaw Meal*TRaw Meal mRaw Meal = (TPDclinker * 1000 / 24) * 1.56 Note: 1.56 for raw meal of habitual LOI. Fuel Latent Heat (Kcal/hr) QFuel = mFuel*CpFuel*TFuel mFuel = (CEH/PCI dry base) * (TPDclinker * 1000 / 24) , Note: CEH (Kcal/kg) to iterate or a approximated value of the installation, since the contribution of Q Fuel to the balance is too small. Total Air Latent Heat (Kcal/hr) QTotal Air = mTotal Air*CpTotal Air*TTotal Air MTotal Air = ρTotal Air * VTotal Air Return Dust to the Cyclones Tower Latent Heat (Kcal/hr) QReturn Dust = mReturn Dust*CpReturn Dust*TReturn Dust Combustion Heat (Kcal/hr) Qcombustion = QLosses – QRaw Meal – QFuel – QTotal Air – QReturn Dust Calculation of the relation Nm3 a m3 Nm3 = m3 * (273 / (273+T)) * 13.6 * 760 * e-0.0001255 * H /10333 Where, T in Celsius degrees H in meters over the sea level




Appendix C Part II. Flame Calculations Minimum air (Nm3/Kg fuel) AminSolid = 1.01*PCI/1000 + 0.5 AminLiquid = 0.85*PCI/1000 + 2 AminGas = 0.875*PCI/1000 Stoichiometric Relation λ = 1 + 0.09 * %O2 Kiln Mass Flow of Stoichiometric Air Kg/s Mst = Amin * ρAir * (%Fuel QP / 100) * (CEH / PCI )* (TPHClinker / 3.6) Mass Flow of Primary Air Kg/s Mpa = Mst * (%

Primary Air

/ 100)

Mass Flow of Secondary Air Kg/s Msa = λ * Mst - (%

Primary Air

/ 100) * Mst

Primary Air Speed m/s vpa = f * 236 * %

Primary Air

-0.4

; where f = 1.5 for modern burners and 1 for the rest.

Primary Air Flow m3/s Vpa = Mpa / ρPrimary Air Diameter Relation Burner (Dq) / Kiln(Di)

Vpa Dq/Di  2 *

  v pa

 Di  0.4

Thring – Newby coefficient

 M pa  M sa θ  Dq/Di    M pa 

  ρ pa   ρ   sa

  

0.5

, Maximum value = 0.94

Fuel Mass (Kg/s) Mb = (%

Fuel QP

/ 100) * (CEH / PCI )* (TPHClinker / 3.6)

Burner Push (Newtons -Kgm/s2-) G = (Mpa * vpa) +[23*(Mb + Mc)] Mc ashes mass in Kg/s = (% Ashes /100) * Mb

Page 173



Flame length


  M pa  M b    M st  M b     M b  (Thring - Newby coefficient) L  7.95   0.06   0.5 1     G  ρ Flame    θ  

ρFlame = 0.181 (Kg/m3) Maximum Flame Position = Burner Position + 0.42 * L (Empirical) Note: the maximum temperature position of the flame.

Final Position of the Flame = Burner Position + L Burner Position = Distance from the dump to the burner point. Nota: in satellites kiln the dump is the entrances middle point.

Page 174




Appendix D Parte II. System Temperatures Calculation Secondary – Tertiary Air Temperature ΔQ

Clinker

= ΔQ

ΔQ

Clinker

= (TPD/24) * CpCk * (TIncome – TOutcome)

ΔQ

Air

Air

= Msa * Cp * ( T3°+ 2° – TAmbient)

Where; TIncome = Theroical sintering temperature – 50 * Burner position T3°+ 2° = (ΔQ

Clinker

+ Msa * CpAir * TAmbient )/( Msa * Cpsa)

T2° = T3°+ 2° + 100 T3° = T3°+ 2° - 100 Main Flame Temperature PCI (=) Kcal/kg ΔQ

Fuel

= mFuel * Cp Fuel * TFuel

ΔQ

Secondary

ΔQ

Primary Air

Air = (λKiln * Amin - Amin*(% = Amin*(%

TFlame  1.1 

Primary Air

 PCI  ΔQ

Fuel

Primary Air

/ 100)) * ρAir * CpAir * T2°

/ 100) * ρAir * CpAir * TAmbient

 ΔQ Secondary Air  ΔQ Primary Air 

 %   0.393  1.5  0.01  Primary Air   100  

    VoR       

Calciner Flame Temperature TCalciner Flame = (ΔQ

Kiln Gases +

ΔQ Tertiary Air + ΔQFuel) / (mEscape Gas * CpEscape Gas)

Calciner Temperature TCalciner = (ΔQ

Kiln Gases +

ΔQ Tertiary Air - ΔQ

Raw Meal

)/ (mEscape Gas * CpEscape Gas)

Hot Stage Temperature THot Stage = 153 * T0.25 Where: ΔQ

Raw Meal

= mRaw Meal * CpRaw Meal * (TCalciner Flame– TPrevious Stage)

TPrevious Stage = 700 + 2 * %Calciner Fuel ΔQ

Kiln Gases

= VKiln Gases * ρ Kiln Gases * Cp Kiln Gases * Tback end * mclinker

ΔQ Tertiary Air = λCalciner * ρair normal AminCalciner * CEH * %KcalCalciner * Cpair * T3° ΔQ

Fuel

= CEH * mclinker * %KcalCalciner

Page 175




Code DP-03-1

Page 176

Appendix E Part II. Thermal Conductivity and Insulation Thermal Conductivity The conductivity can be expressed in different units, and according it is distinguished by the letters W and K: Conductivity W, is expressed in units W/m K: W = Watt K = Kelvin degrees temperature (by nomenclature, the ° is not written) m = meters Conductivity k, kCal h m °C

is expressed in units kCal/h m °C: = kilo Calories = hours = meters = Temperature in Celsius degrees

The approximate relation between both expressions is: 1 W/m K

 0.86 kCal/ h m °C

The conductivity calculation at 1000 °C for a refractory with different oxides proportions is: K

1000 °C

Oxide I MgO Al2O3 SiO2 Cr2O3 CaO Fe2O3 ZrO2 SiC Others

=  (%Xi/100) * Ki Wi

1000

Ki

1000

]

1000

W/m K

kCal/h m °C

3.1 2.3 0.0 1.2 0.4 0.0 0.0 10.4 0.0

2.7 2.0 0.0 1.0 0.3 0.0 0.0 9.0 0.0

The thermal conductivity at any other temperature (expressed in °C) is obtained by:  1000  T C  KT C  K1000C  0.73   1000   Refractivity Ta (°C)

 % xi  Ta     Ta j  100 



Oxide i


Page 177

Tai °C

The Tai temperature must not be although there is some relation. *Note: If in the composition a due to other elements not taken for that shortage. Cold Resistance (N/mm2)

 % xi  R   Ri  100 

TiO2 SiC ZrO2 Cr2O3 MgO Al2O3 SiO2 CaO Fe2O3 Others*

7000 2000 1750 1750 1750 1700 1200 825 750 1000

Oxide i

confused with the fusion point, 100% of oxides is not comepleted considerated, the 1000 °C value is

Ri N/mm2

For concretes the same equation multiplied by 1.55 to obtain the Density(g/cm3)

 % xi     i  100 

Cr2O3 ZrO2 SiC Al2O3 MgO Fe2O3 SiO2 CaO

190 170 150 80 45 30 2 2

will be used, but the result will be resistance value at 1000°C.

i

Oxide i

g/cm3

Dilatation (%) It is mainly presented in low zero in those of high density. It factor affects the dimensions (L)

 % xi  D1000C     Di 1000C  100 

SiC ZrO2 MgO Al2O3 Cr2O3 CaO Fe2O3 TiO2 SiO2 Oxide i

At other temperatures: D

T°C

=D

1000°C

+ 0.0014 * (T°C –

Cycles Cycles =  Xi * Cycles i ]

4 4 3 3 3 3 3 3 1 Di

density oxides, and it is practically must be considered that this as the (h).

1000°C

%

ZrO2 MgO CaO SiO2 Al2O3 Fe2O3 i Cr2OOxide 3 SiC MgO TiO2 Al2O3 Cr2O3 ZrO2 TiO2 SiC Rest

80 1.6 1.6 1.6 0 0 Cycles 0 0 110 0 80 80 80 80 80 0

1000)

i




Page 178

Placing of isolators in smokes chamber.

Ta 15 °C

Ts 300 °C

Ta 15 °C

Ts 300 °C

12 mm 200 mm

200 mm

The isolator plate its placed easily (the placing cost does not rise) The heat losses by convection and radiation (Kcal/hr m2) are described with the following formulas:

hconvection

 T  Tair   80.33 *  s  2  

0.724



4 hradiation  4.2 x10 8 * Ts4  Tair

*  Ts  Tair 

1.33



Example for shell temperature of 300 °C:

 300  25  hconvection  80.33 *   2  



0.724

*  300  25

hradiation  4.2 x10 8 *  300  273  15  273 4

4

1.33

 1831.6

  4238.6

q Kcal  1831.6  4238.6  6070.2 A hr  m 2 With shell temperature of 200 °C:

 200  25  hconvection  80.33 *   2  



0.724

*  200  25

hradiation  4.2 x10 8 *  200  273  15  273 4

4

1.33

 1127 .3

  1813.3




Page 179

q Kcal  1127 .3  1813.3  2940.6 A hr  m 2 Total saved energy:

q Kcal   6070.2  2940.6   3129.6 A hr  m 2 Coke PCI = 8200 Kcal/kg coke, coke savings = 0.38 kg coque/hr m 2 Coke cost = 0.65 $/kg, Savings in $ = 0.25 $/hr m2, So a 100 m2 section where isolation is installed will save each year (7,800 operation hours) $200,000.00 in radiation losses. For a kiln of 2000 TPD of clinker with a heat consumption of 820 Kcal/kg of clinker:

Kcal    2000TPD * 1000 * 820 Kg Kgcoke    * 8200  916 8200 day  Specific consumption =  Kcal/kg of      816.2 2000TPD * 1000

ck The saving will be of 3.8 Kcal/kg of clinker.

Chapter 3 Clinker

Recommend Documents