THERMAL

ENGINEERING

PREDICTION OF THE SERVICE LIFE OF THE POURED CHECKERWORK OF A BLAST FURNACE HOT-BLAST STOVE L. N. Toritsyn

UDC 669.162.23

It is known that the service life of the checkerwork of a regenerative heat exchanger depends significantly upon the accumulation in it of slags and dust contained in the heat carriers and also upon the corrosion of the checkerwork material under the aciton of slags. Dust and slag have a significantly stronger influence on the characteristics of poured checkerwork than on checkerwork of normal designs such as channel ones [1-5]. In [3, 4] the results are presented of investigations of the interaction of high-temperature slag-containing combustion products with two forms of actual checkerwork, that of 20 mm diam. corundum spheres produced by the Experimental Plant of the Ukrainian Scientific-Research Institute for Refractories and that of mullite-corundum chips with an average size of about 20 mm produced by Semiluki Refractory Plant. Both forms of poured checkerwork have been recommended by the AllUnion Scientific-Research Institute for Metallurgical Thermal Engineering together with their collaborators in the work, the Ukrainian Scientific-Research Institute for Refractories and Eastern Refractory Institute, for use based on their thermal engineering and hydraulic characteristics, capabilities of use at high temperatures, and economic considerations. On the basis of analysis of calculations and actual methods of preparation of checkerwork in refractory plants it has been assumed that there are possibilities of production using the method of Eastern Refractory Institute of inexpensive checkerwork of mullite-corundum chips (200-300 rubles/ton in price before 1989) and of granulated corundum spheres (500-600 rubles/ton) using an effective and simple method developed in detail by the Ukrainian Scientific-Research Institute for Refractories [6]. The interaction of dust and slags with the checkerwork must lead to an increase in its hydraulic resistance [1-5]. During a certain time of service the resistance or the checkerwork reaches a limiting value based on the specific operating conditions of the heat exchanger and this determines the actual service life of the checkerwork. Therefore the hydraulic resistance of the heat exchanger is a criterion of effectiveness under the action on it of dust and slags. To determine the rules of interaction of a poured checkerwork of 20 mm diam. corundum spheres with a low-temperature flow of gases containing dust in the solid state investigations have been made [5]. In all of the tests [3-5] with flows containing dust and slag in different aggregate conditions the same Magnitogorsk Metallurgical Combine blast furnace dust was used. The characteristics of the dust are given in [3, 5]. Here let us give only the properties of fusibility of the dust according to GOST 2057-82: temperature of the start of deformation t I = 1220-1360~ of softening t 2 = 1290-1440~ of transition to the fluid state t 3 = 1380-1560~ average size of dust particles d = 23.10 -~ m. The dust concentration of the gases at entry into the checkerwork was 0.015-1.15% and the average rate of filtration of the gases 0.1-0,9 m/sec at 0~ [3-5]. In the tests [3, 4] the temperature of the top of the checkerwork t t varied within limits of I130-1570~ and the temperature of the combustion products in the combustion chamber located before the checkerwork tcc = 1350-1680~ In the tests [5] the average temperature of the air and the checkerwork was 25~ As the result of the tests the following relationships of the change in coefficients of hydraulic resistance of the checkerwork were obtained. With flow of the combustion products through the checkerwork from top to bottom and with t t < t I and simultaneously with tcc > tl a relatively dense layer of adhering and sintering particles (crust) gradually f o r m e d o n the checkerwork, as the result of which there is a significant increase in resistance of the checkerwork. The local coefficient of hydraulic resistance of the crust ~c may be determined using an equation obtained in processing of experimental data: c = a r / ( P a a),

All-Union Scnetific-Research Institute for Metallurgical Thermal Engineering. ted from Ogneupory, No. 12, pp. 24-29, December, 1991.

0034-3102/91/1112-0651512.50

9 1992 Plenum Publishing Corporation

(1)

Transla-

651

where F is the weight of dust (slag) delivered to a unit of area of the upper section of the checkerwork in kg/m 2, a is a coefficient and a = 21.3 [3, 4], and Pd is the bulk density in kg/m 3 . For the conditions tcc > t t > t I ("hot" tests [3, 4]) and t t = tcc and t t < t I ~ cold tests [5])slagging occurred in the hot tests and clogging in the cold tests of the upper portions of the checkerwork with a corresponding increase in their hydraulic resistance. The condition and hydraulic resistance of the lower portions of the checkerwork did not change in this case. The coefficient of hydraulic resistance ~s of the upper slagged or dustclogged portion increases approximately directly proportionally to F and may be calculated using the equation

ae ),

(z)

where ~s,0 is the initial coefficient of hydraulic resistance of the portion of the checkerwork considered, A is an experimental coefficient dependent upon the temperature conditions, s is the height of the slagged or dust-clogged portion of the checkerwork, H is the initial porosity of the checkerwork, and Pl is the apparent density of the slag (Ps in the hot tests or the bulk density of the dust Pd in the cold tests). For the conditions tcc ! t t > t I the value of s is determined by the location of the lower cross-section of the slagged portion of the checkerwork. This cross-section [3, 4] is located on the horizontal plane with a maximum temperature of the checkerwork in the cycle equal to t I. In accumulation in the checkerwork of solid dust not changing its aggregate condition in the checkerwork the value of ~s in m may be determined from the equation [5]:

~s = [Kl (F--x) + K~ ]/(p d R),

(3)

where K I = 1.46, K 0 = 2.52, x = F with F ~ F 0 and x = F 0 with F > F 0, and F 0 = 50 kg/m 2. The value of A changes depending upon temperature conditions from 6 to 190 and in extrapolation fo the experimental data up to 400 [3-5]. Dust may be supplied to a regenerative heat exchanger not only with the combustion products but also with the heat carrier being heated such as with the blast furnace blast air. Let us consider only one geometric plan of a heat exchanger, vertical location of the checkerwork with supply of the heating heat carrier (such as combustion products) to the checkerwork from the top and exit of it from the bottom of the checkerwork and delivery of the heat carrier being heated (such as blast air) to the checkerwork at the bottom and exit of it from the top of the checkerwork. We did not make special tests simulating this case but it is possible to make quite reliable assumptions based on existing experimental data. The behavior of the checkerwork in accumulation of dust not having its own aggregate state will be similar to that described in [5] in delivery of a dust-air mixture to the top. With tcc ~ t t > t I the dust entering with the cold heat carrier will be held in the zone of the checkerwork with a temperature of about t I as the result of adhesion to the checkerwork of dust in which the liquid phase is starting to form. In this case processes similar to those investigated in delivery of the combustion products to the top of the checkerwork with tcc > tl > t t with formation of a slag crust in the zone of the checkerwork with a tempeature of about t I will occur. Let us use Eq. (i) in the following manner. We will assume that the dust particles and conglomerates deposited in the voids between the checkerwork elements form together with these elements of the checkerwork a slag crust with relationships of the hydraulic resistance exactly the same as for the crust formed at the top of the checkerwork. The thickness of the inner crust of slag is inversely proportional to the porosity of the checkerwork. The coefficient of resistance ~c,i of the inner crust will be equal to

~,i= a&/(~na),

(4)

where F c is the specific mass of dust supplied to the crust (area of formation of the crust). Earlier we did not consider the influence of the cyclic nature of the operation of a regenerative heat exchanger on the change in its hydraulic resistance. In this case let us also make the corresponding assumptions. In the presence in the heat exchanger of aggregate transformations of dust with solidification of slags as the result of the prevailing influence of forces of adhesion in comparison with all other factors it may be assumed that the cyclic na-

652

ture of operation of the regenerator with a periodic change in heat carriers and direction of their movement neither the values of ~s nor the quantity of slag being held by the checkerwork change. Apparently under these conditions the experimental relationships of [3, 4] and Eq. (2) may be used without any changes. In the case in which the dust remains in the solid phase in the whole checkerwork the periodic nature considered of the flows in the checkerwork clearly influences the change in hydraulic resistance. Some portion of the dust carried into the checkerwork with the preceding heat carrier will be blown out from the checkerwork by each heat carrier. The quantity of dust blown out depends upon a multitude of factors [7], including the flow rates. Calculations of the checkerwork resistance may also be made using Eqs. (2) and (3) but in this case decreased values of F are used:

Fe= KpF,

(5)

where F e is the effective specific mass of dust entering into the checkerwork from the side of the given heat carrier, F is the specific mass of dust entering into the checkerwork with the given heat carrier, and K F ! I is a coefficient dependent upon the conditions of flow and the characteristics of the dust and the checkerwork (for the conditions of the tests [5] and in operation of a regenerator with the same heat carrier velocities it may be assumed that K F = 0.6-0.9). The hydraulic resistance of the checkerwork may increase significantly [8] if under service conditions plastic deformation of it leading to a decrease in porosity occurs. The hot-blast stove must be constructed so that plastic deformation of the checkerwork does not occur. However, in practice plastic deformation does occur and therefore it is necessary to determine the increase in resistance as the result of creep of the checkerwork. The calculation of deformation of the checkerwork and of the change in its resistance must be made using the method of [8] with use of the actual temperature and pressure distribution. The values of the thrust pressures of the checkerwork over its height are determined using the method of [i0] and are used in calculations of the resistance using the method of [8] instead of weighted loads. As the result of a change in temperatures in the heat exchanger becasue of unsteadiness in its operating conditions displacements of the elements or the checkerwork, such as spheres, relative to one another and of the surface of the wall and the support surface occur under the simultaneous action of quite high stresses [ii]. In this case some abrasion of the walls and spheres and crushing of the spheres are observed [9]. The dust formed primarily remains in the checkerwork but a portion of it falls out and also is carried downward by the combustion products into the space under the checkerwork. The dust formed directly in the heat exchanger may be so much that it becomes one of the basic reasons for the increase in resistance of the checkerwork. This occurs in an existing block of blast-furnace hor-blast stoves with a poured checkerwork of Kosogorsk Metallurgical Plant. Making quantitative evaluations is quite difficult. The main portion of the dust formed has a high refractoriness and does not undergo a phase transformation in the checkerwork operating temperature range. At least some quantity of the dust formed by elements of the heat exchanger low-temperature zone structure with relativ-ly low refractoriness may change. In this case in entry of the dust into the high temperatuz zone conditions of transition of the dust into liquid phases are possible. The main portio~ ~f the spontaneously formed dust is concentrated in the lower portion of the checkerwork, wh~ze the stresses are a maximum and the local displacements of the structural elements are not rarely also a maximum. Judging from our observations in 1989-1991 over the block of hot-blast stoves with a poured checkerwork of Kosogorsk Metallurgical Plant an unceasing process of spontaneous formation of dust occurs in the checkerwork. As a first approximation the influence of this dust on the resistance of the checkerwork may be evaluated if it is assumed that the distribution of the dust over the height of the checkerwork is similar to the distribution during accumulation of dust with the blast air in the checkerwork. Then to evaluate the increase in the checkerwork resistance the results of [5] may be used. For this it is sufficient to know the quantity of dust formed or the height of the dust covered zone, the volume mass of the dust, the temperature distribution over the height of the checkerwork, and the t I temperature of thedust. As may be seen from this, prediction of the change in hydraulic resistance of a poured checkerwork presents some difficulty and has a number of indeterminacies. However, in development of the heat exchangers considered prediction of their operation is necessary. Therefore let us set up a method of prediction. 653

First let us specify the parameters of the base condition of the heat exchanger: the consumptions of combustion products and blast to be heated; the periods of heating the checkerwork ~c and heating of the blast rb; the length of the total operating cycle of the heat exchanger ~cy; the concentrations and melting points of the dusts in the combustion products and the blast; the quantity of dust spontaneously formed in a unit of time in the checkerwork from erosion of the refractories and grinding of the checkerwork; the temperature tcc of the combustion products leaving the combustion chamber; the temperatures of the top of the checkerwork tt; the temperature t c distribution of the checkerwork over its height. To evaluate the influence of creep of the checkerwork on its resistance it is necessary to select from the possible checkerwork temperature distributions the distributions with the maximum checkerwork temperatures. In this case the thrust pressures of the checkerwork increase [!I] and the rate of deformation in creep increases as the result of the increase in stresses and temperatures in the checkerwork [12]. For the evaluations of the influence of the other "dust" factors the same temperature conditions may be used. Using the method of [i0] let us determine the distribution of pressures of the checkerwork on the wall over the height. Let us assume that the average stresses in the sections of the ~heckerwork are equal to the checkerwork pressures found on the wall in the same sections~ Using the method of [8] and taking [12] into consideration let us determine the change with time in the hydraulic resistance of the checkerwork ns(z) referred to the initial hydraulic resistance and caused by deformation of the checkerwork. The function ne(z) may be constructed from several values of n E and the time interval must be selected based on the maximum allowable increase in hydraulic resistance m of the checkerwork of the given heat exchange. For the conditions of blast furnace production m = 1.5 may be frequently assumed. Let us switch to evaluations of the change in hydraulic resistance of the checkerwork caused by the presence of dust and slags. Let us make refinements in the temperature tl. We should note that t I is determined from several tests for which, as a rule, the values of t l differ. Therefore the tl, a ! tl ! tl,b range with minimum tl, a and maximum tl, b t~ temperatures is always considered. For some evaluations it is possible to use the average temperature of the start of deformation of the dust tl,av: 11'av=

2

L e t us d i s t i n g u i s h n i n e ( i ) b a s i c c a s e s o f i n c r e a s e s c h e c k e r w o r k c a u s e d by t h e p r e s e n c e o f d u s t and s l a g s .

in h y d r a u l i c

resistance

of the

In the first case (i = 1)the d u s t y h e a t c a r r i e r is the p r o d u c t s of combustion, t c c < t ~ , a, and t t < t z , a, t h a t i s , i n t h e w h o l e s y s t e m t h e d u s t r e m a i n s in t h e s o l i d s t a t e and does n o t u n d e r g o a g g r e g a t e t r a n s f o r m a t i o n s . I n t h e s e c o n d c a s e ( i = 2) t h e d u s t y ( s l a g - c o n t a i n i n g ) heat carrier is the products of c o m b u s t i o n , t c c ~ t l , a, and t t ! t l , a The d u s t in t h e c o m b u s t i o n p r o d u c t s i s f o u n d in a s t a t e o f a g g r e g a t e t r a n s f o r m a t i o n s and even may be in t h e l i q u i d s t a t e b u t in t h e c h e c k e r w o r k t h e d u s t may be o n l y i n t h e s o l i d s t a t e and s e t t l e s on i t s s u r f a c e , f o r m i n g a s l a g c r u s t . With i = 3 t h e d u s t y h e a t c a r r i e r i s t h e c o m b u s t i o n p r o d u c t s , t c c > t l , a v , and t t > tl,av. The d u s t i n t h e c o m b u s t i o n p r o d u c t s i s f o u n d i n a s t a t e o f a g g r e g a t e t r a n s f o r m a t i o n s o r i n l i q u i d form and i n t h e c h e c k e r w o r k i t s o l i d i f i e s in the area of the s e c t o n having the maximum t~,av temperature. With i = 4 a flow of blast air (blast) passes, t t ! tl,a, and the dust is found in the checkerwork in solid form. In the case of i = 5 a flow of blast air passes, t t > tl,av, F e < F c r (Fcr is the specific mass of dust entering into the checkerwork at which the dust reaches the section with the maximum temperature in the cycle equal to tl,av), and the dust is in solid form since it is in the checkerwork at temperatures lower than tl,av. In the case of i = 6 a flow of blast air passes, t t > tl,av, and F e ~ Fcr. A portion of the dust is transferred to the portion of the checkerwork with temperatures of t c ~ tl,av and as the result in this area formation of an internal slag crust occurs while the dust located in the lower portion of the checkerwork with t c < tl,av is in the original solid state. With i = 7 the checkerwork is clogged with dust formed in abrasion of the walls and checkerwork and grinding of the checkerwork, t t ~ tl, a, and in the whole height of the checkerwork the dust is in the original solid state. 654

In the eighth case (i = 8) it is the same as with i = 7 but t t > tz,av. In the lower portion of the checkerwork up to the section with t c = tz,av the dust is in the original solid state while above it is in a state of transition into liquid phases. With i = 9 the dusty heat carrier is combustion products, tcc > tz, a, tz,a < tt ! tz,av~ and the case is intermediate between the second and third. The dust both forms a crust on the checkerwork and spreads on its upper portion. The data of the investigations of ships for calculation of the coefficient to the initial coefficient of resistance for the above cases with i = i, 3, 4, 5,

[1-5] makes it possible to propose simple relationof hydraulic resistance of the checkerwork np referred in relation to the action of the dust. For example, and 7 we may write

rip.f= ~

Pe(T)+ l

ltplLlav

(6)

or

ACa~xlTs

np~=31,54 np,/rcy~v r+l.

(7)

where s is the height of the checkerwork in m, Tav and T s are the average by mass temperatures of the whole checkerwork and of its slag (dust) covered portion in K, 9 is time in years, ~j is the length of the period of heating of the checkerwork, the blast, or the whole cycle during which dust or slag enters the checkerwork in h, ~cy is the length of the heat exchanger operating cycle (heating + blast + changing of the valves) in h, v is the rate of flow (filtration) of the given heat carrier through the checkerwork in m/sec (at normal conditions), A is the coefficient of intensity of increase in resistance of the checkerwork, which is determined with transformation of Eq. (2) and use of experimental data, and Cad is the adjusted dust concentration in the heat carrier in mg/m 3 (under normal conditions):

cad = cK~

(8)

(C is the true average dust conoentration in the heat carrier under normal conditions in mg/ m3). Let us recall that for the case of i = 3 K y = i. To determine T s it is necessary to know the height s of the slag- (dust-) covered portion of the checkerwork. For i = i, 4, 5, and 7 s is determined from Eq. (3) for F = Fe(t). For i = 3 according to [4] Zs is equal to the height of the upper portion of the checkerwork to the section in which the maximum temperature in the cycle is equal to t~ 9 a v 9 The hydraulic resistance of the checkerwork under conditions of formation directly in it of dust was not investigated and not simulated. Therefore for the evaluations, as already noted, it is assumed that the process of accumulation in the checkerwork in these cases with regard to the hydraulic resistance is equivalent to the process of accumulation in the checkerwork in entry into it of dust with the blast air. For the calculations we determine F e or Cad from the quantity of dust spontaneously formed and remaining in the checkerwork. The evaluations of the change in hydraulic resistance for the cases of i = 7 and 8 are made on the basis of such an analogy. We take the coefficient A for the cases of i = i, 4, 5, and 7 from [5] and for i = 3 from [4]. In the latter case the coefficient A is a function of the temperature of the top of the checkerwork t t. Since in [4] the dust has definite properties, for use of the data of this work for dusts with a different t z for determination of A it is proposed to use the equivalent temperature of the top of the checkerwork te: te-=tt--t,.~+t,..

(9)

where tz, z and tl, z are the average temperatures t z of the dust investigated in [4] and of the given dust. Equations (6) and (7) are identical but it is more convenient to make calculations for the cases of i = i, 3, 4, and 5 with use of Eq. (7) and for i = 7 of Eq. (6). On the basis of Eq. (I) in analogy with Eq. (7) we obtain the equation for calculation of the increase in resistance for the case of i = 2:

np~

=

aColttc r-l.- ], 3I ,54 ~ld~,~T~rey

(1o)

655

where T t is the temperature of the top of the checkerwork in K and ~0 is the original coefficient of hydraulic resistance of the checkerwork for the heat exchanger conditions considered. The cases of i = 6 and $ present some difficulty for evaluations. The following model is proposed. Let us follow the spread with time of the dust at the top in the checkerwork accordin to Eq. (3) in relation to F = Fe(~). Let us record the moment of time rcr at which the dust reaches the section with a maximum temperature in the cycle equal to tl,av. Let us assume that all of the dust entering and remaining after this moment goes into formation of an internal slag layer while the resistance of the lower portion of the checkerwork remains constant and equal to that reached at ~cr" Using Eqs. (4), (6), and (7) it is possible to approximately write __(Pe-&,cr)Lav]

/,_

. ~

Ii)

or

~p;----[@

Fe,cr~ 3J.S4Q

Cad~T"av ~ (T---~c~]/ (17PdTav)'r'~

d~o~cy

or

npi=3~,54[~ %r+a~ov ( T _ ~ c, ) C'j a~dWb-~ cyt ~

13)

where i = 6 for 8, Fe > Fe,cr , ~ > ~cr, Tl,av is the average temperature of the start of deformation of the dust in K, and Fe,cr is the specific mass of dust which has entered inta the checkerwork at the moment of its spread to the section with t~,av in kg/m =. Equations (12) and (13) are more convenient to use for the case of i = 6 and (Ii) for i=8. The coefficient np, 9 is most conveniently determined by the method of quadratic ulterpoiation with respect to t t between the value of np, 2 to t t = t ~ a and the values of np,~ for t t = tl,av + A~ and for t t = tl,av + Az. It is convenient to use a~ = 50 and A 2 = 100~Co Actually more than one of the cases considered of the change in hydraulic resistance of the checkerwork may occur simultaneously. For example, in the hot-blast stoves of Kose gorsk Metallurgical Plant we noted plastic deformation of the spheres, accumulation of dust in the checkerwork from the top (i = 3), accumulation of dust with the blast (i = 5 and 6), and formation of dust in the checkerwork (i = 7). Determination of the local resistance of the checkerwork (on the basis of which it would be possible to determine the total resistance of the checkerwork and the increase in it) under the simultaneous action of more than one factor is a completely unsolvable ~roblem. A simple approach is proposed. Let us assume that all of the considered processes occur independently of one another in the checkerwork. Then the total increase in checkerwork resistance as a function of time n(~) is determined from the product of the time functions for the individual processes: 9

and for the cases k = i not realized in the given exchagner np, i = I. Having determined the time function n(~) for the area considered at n(~) of the less allowable value of n = m (more accurately 0 < n < m) for the given equipment let us determine the service life of the checkerwork until repair. For conditions of the absence of deformation in creep of the checkerwork and occurrence of only one of the first eight considered cases of increase in hydraulic resistance by transformation of Eqs. (7), (i0), and (13) we obtain the final equations for calculation of the service life of the checkerwork ~sv: for i = I, 3, 4, 5, and 7: fi f i q 1 7 1 ~

656

|~

HPJTcYTav

(!5)

for i = 2:

(16)

z sv 2= 0.0317 mi-- 1~pdd~~Tav*-eY ,

9

(

, aCvTtx c

'

and for i = 6 and 8:

T s v ,,=0;0317(m,--1)//aP~C ~~ where m i is sistance of with values with i = 4,

vTb ' + ( 1 - -

ATscl~~

) ~cr ,~,

(17)

the maximum allowable for the given i increase in the coefficient of hydraulic rethe checkerwork of the given exchanger and m i = np,i(~sv). Equation (17) is valid of m i > mi,cr for which isv, i(mi,c r) = ~cr,i and ~j = ~c with i = i or 3, zj = ~b 5, 6, or 8, and ~j = Tcy with i = 7.

For calculations using Eq. (17) it is necessary to know ~cr and T s. F e through Cad , v and T:

First let us express

(18)

Fe=31,54CadV~JX.

Toy

Then l e t us introduce the understanding of the height of the portion of the checkerwork ~g from the bottom to the section with the maximum temperature in the cycle equal to Tl,av. Lgt us r e c a l l that with s ! s s is calculated according to Eq. (3) for F = Fe: I s = [K,(Fe - x ) +

Kox]/(gdII).

( 19 )

In calculation o f Ls a c c o r d i n g t o Eq. ( 1 9 ) t a k i n g i n t o c o n s i d e r a t i o n Eq. ( 1 8 ) we r e v e a l t h a t ~s i s a f u n c t i o n o f t i m e ~ s ( X ) and Z c r i s t h a t moment o f t i m e when ~ s ( Z c r ) = ~g. To d e t e r m i n e t h e t e m p e r a t u r e T s we a s s u m e a p p r o x i m a t e l y t h a t ~s = ~g" The s e r v i c e t i m e o f t h e c h e c k e r w o r k X s v , s may be d e t e r m i n e d by t h e m e t h o d o f q u a d r a t i c interpolation between the value of ~sv,2 for t t = tl, a and the values of Zsv,~ for t t = tl,sv + A I and t t = tl,sv + A 2. The remaining parameters in the calculations are the same. With i = 2, 3, 6, and 8 some quantity of slag in the liquid state will be found in the checkerwork and chemical interaction will occur between the slag, the checkerwork, and the lining of the exchanger. It is necessary to evaluate the influence of this interaction on the service life of the exchanger. For this by transformation of Eq. (6) we obtain the equation for the ratio of the weight of slag M s in the checkerwork in the liquid and solid states to the weight of the checkerwork M c in which the slag is located: M s __(m -- I)HpcTav t

Mc w h e r e 0c i s t h e

bulk density

of the

Ap e ~ l s

(20)

'

checkerwork.

The c o e f f i c i e n t A changes from approximately 400 a t t t = 1290~ t o a p p r o x i m a t e l y 10 a t t t = 1600~ For m = 1.5 and the corresponding v a l u e s i n Eq. ( 2 0 ) we f i n d t h a t Ms/M c c h a n g e s from approximately 0 . 0 0 2 a t t t = 1290~ t o a p p r o x i m a t e l y 0.06 at t t = 1600~ Consequently for a significant increase in the resistance of the checkerwork entry into i t o f an i n s i g n i f i c a n t quantity of dust is sufficient. Simultaneously even in a rough evaluation of possible chemical reactions between probable d u s t s a n d c h e c k e r w o r k s o f c o r u n d u m and mullite-corundum refractories and also taking into consideration the fact that the main portion of the slag in the checkerwork will be in the solid state, i t may b e c o n c l u d e d t h a t t h e c h e c k erwork is almost completely free of corrosion and most probably preserves its design strength and effectiveness. Let us determine how much the products of this small amount of corrosion may influence. the hydraulic resistance. The dust used in the tests was very unfavorable for the checkerwork since it contained more than 50% iron oxides. In interaction if it with the checkerwork there is formed some quantity of low-melting phases, which increases the total weight of slag flowing on the checkerwork. In interaction with the corundum refractory the increase in slag weight is not more than 6% of the weight of slag entering into the checkerwork while in interaction with mullite-corundum refractory not more than 17-30% (the increase from 17 to 30% is related to the increase in the temperature t t from 1300 to 1550~ Therefore a small additional increase in checkerwork hydraulic resistance may be expected. However, as the result of its small value even taking into consideration the quite large error in the proposed evalua-

657

120, a

11o -

- -

-z'~'Jl I I I

~80 ~ro

I

I~ I i I

I '

~,, ~

~ i I

3:

I

]j

t.f.E"ltT,a tlav

t01

i',, 4,r162 ', i / 1 /

[, o! 1000

!

il, i I , j

30 20 10 0

i

I

!t

I

i

I

I

1100

1200

7300

I~00

I

i

1500 7600

Fig. i~ The chagne in service life of the Tsv of the poured checkerwork of a Kosogorsk Metallurgical Plant hot-blast stove in relation to the temperature of the top of the checkerwork t t and the temperature t~ of the dust entering the checkerwork with the combustion products: a) complete figure; b) a fragment; I) tl, a = 1230~ tl, b = 1370~ t~ av = 1300~ 2) 1280, 1420, and 1350~ 3) 1330, 14~0, and 1400~ I, II, III, and IX) cases with i = i, 2, 3, and 9 for variation of the dust 1 (for variation of the dust 1 Fig~ la shows the method of separation of the area of t t into the characteristic cases); the productivity of the hot-blast stove VTVN-I is 60% of the nominal; dust concentration in the burner air 0.3 mg/m 3. tions this factor may be completely disregarded or taken into consideration by introducing in{;~ the corresponding equations for n coefficients somewhat greater than unity and in the equations for ~sv less than unity. On the basis of the above it is possible to evaluate the change in and the service time of an actual exchanger. As this exchanger we will take a of Kosogorsk Metallurgical Plant with a poured checkerwork with a height spheres in service in 1989-1990 for 189 years until shut down for repair. coefficient of hydraulic resistance increased by about 100%.

hydraulic resistance hot-blast stove VTVN-I of 5.5 m of corundum During this time the

The parameters of the basic operating conditions were temperature of the dome and the top of the checkerwork 1500~ and the temperature distribution of the checkerwork over its height may be seen in [15, Fig. 5, curve 2]. The productivity of the hot-blast stove was about 60% of the nominal. The blast air and the air going to the burner contained about 0.3 mg of dust per 1 m ~. The average size of the dust particles was less than i0 ~m. One of the samples of trapped dust was analyzed in the Ukrainian Scientific-Research Institute for Refractories. Its chemical analysis in wt.% was 16.9 SiO 2, 3.6 AI203, 5.34 Fe203, 34.6 CaO, 3.2 MgO, 0.55 Na20, 0.8 K20, Amcalc 22.8%, and traces of Mn. The t I temperature was 1320~ A single analysis was insufficient for determination of the average properties of the dust in service. Nevertheless the most probable range of the average t~ temperature of the dust is 1300~ ~ tl,av ~ 1400~ Figure i* shows the relationships of the hot-blast stove service life to t t and t I with m = 1.5 calculated using the method presented for cases of entry of dust with the combustion air. It may be seen that Tsv depends substantially upon t t and t I. The maximum service life *N. L. Brun'ko participated in the calculations for Fig. i. 658

is in the area of low and increased temperatures and the minimum with t t close to tl, a. For the operating temperatures of the hot-blast stove with t t = 1500~ the service life for all of the variations of t I considered is more than i0 years and in I~ years of service the coefficient of resistance of the checkerwork even in the poorest case increases by less than 15%. Calculations of the influence of dust entering into the checkerwork with the blast showed that in this case during the same i~ years the coefficient of resistance increases by only 1.5%. An evaluation was made of the increase in resistance of the checkerwork asthe result of dust formed in the checkerwork. Based on the results of unloading the checkerwork during repair s = 0.8 m. It as found that in this case the coefficient of hydraulic resistance of the checkerwork may increase by 40%. An approximate evaluation was made of the influence of creep deformation of the checkerwork under the~ action of thrust forces in it on hydraulic resistance. According to the data of the operating conditions of the hot-blast stove considered [ii] the thrust forces in the area of deformation of the checkerwork (with temperatures above 1300~ are approximately eight times more than the weight loads. The deformation rate of the checkerwork increases proportionally to the square, of stresses in it [8, 12]. Taking this and the data of [8] into consideration for t t = 1500~ we find that as the result of deformation in creep of the checkerwork the coefficient of resistance of the checkerwork may increase by about 60% in 189 years. Therefore the basic reasons for the increase in resistance of the Kosogorsk Metallurgical Plant hot-blast stove are deformation in creep of the checkerwork and spontaneous formation of dust in it. This is confirmed by the condition of the checkerwork during unloading of it from the hot-blast stove during repair. The most significant changes were expressed in the presence of a quite large quantity of conglomerates of spheres with crushed contacts as the result of creep and the presence of a large quantity of dust primarily consisting of materials of the checkerwork and the walls and located in the lower portion of the checkerwork up to a height of approximately i m from the bottom. Slag clogging of the checkerwork is expressed very weakly at individual points. There is also little dust introduced with the blast air. The amount of plastic deformation of the checkerwork was about 10% and the porosity decreased by 10-20%, which agree with each other. With this decrease in porosity a 20-60% increase in the hydraulic resistance of the checkerwork may be expected [8]. The increase in the coefficient of hydraulic resistance from the combined action of these two factors (deformation of the checkerwork and spontaneous formation of dust in it) must be 60-100%, which is sufficiently close to the increase in the coefficient of resistance occurring in the hot-blast stove (about 100%). An evaluation made simultaneously makes it possible to reveal the secondary factors influencing the service life of the checkerwork, entry of dust into the checkerwork with the blast and the combustion products. At present the question of thorough cleaning of dust from the blast and combustion air at Kosogorsk Metallurgical Plant is being considered. The use of cleaning is expensive and judging from the data of the evaluation does not provide the necessary effect. CONCLUSIONS On the basis of laboratory investigations and an analysis of operation of production regenerative heat exchangers with a poured checkerwork a method has been developed for calculation of the change in hydraulic resistance of the checkerwork and prediction of its service life until repair. The primary reasons for the observed increase in hydraulic resistance of hot-blast stoves with a poured checkerwork at Kosogorsk Metallurgical Plant were revealed. The method is recommmended for use in development of regenerative heat exchangers with a poured checkerwork and also for analysis of operation of existing exchangers. LITERATURE CITED i.

2. 3.

L. N. Toritsyn and L. V. Uzberg, Teplofiz. Vys. Temp., No. i, 145-151 (1989). L. N. Toritsyn and L. V. Uzberg, Teplofiz. Vys. Temp., No. 2, 374-378 (1989). L. N. Toritsyn, N. L. Brun'ko, E. D. Lekomtseva, and L. V. Uzberg, The Change in Hydraulic Resistance of the Poured Checkerwork of a Regenerative Heat Exchanger in Heating of it with Slag-Containing Combustion Products [in Russian], Sverdlovsk (1988), Deposited in the All-Union Institute for Scientific and Technical Information 11/14/88, No. 8065-1388.

659

.

5. 6. 7. 8.

9. i0.

ii. 12. 13.

660

L. N. Toritsyn, N. L. Brun~ko, E. D. Lekomtseva, and L. V. Uzberg, Teplofiz. Vys. Tempo No. 4, 797-806 (1989). L. N. Toritsyn, N. L. Brun'ko, E. B. Shamgunova, and M. A. Babina, Ogneupory, No. 8, 25-28 (1991). I. F. Usatikov, N. M. Chudnova, T. M. Shlyakhova, et al., Ogneupory, No. 2, 39-43 (1990}. L. N. Toritsyn, V. L. Sovetkin, and E. I. Bulatnikova, Izv. Vyssh. Uchebn. Zaved., Tsvet. Met., No. 4, 67-71 (1985). L. N. Toritsyn, Teplofiz. Vys. Temp., No. 5, 1032-1034 (1989). L. N. Toritsyn, Ogneupory, No. 6, 23-29 (1991). L. N. Toritsyn, Calculation of the Pressure of a Poured Checkerwork on the Walls and Bot ~o tom of a Regenerative Heat Exchanger [in Russian], Sverdlovsk (1989), Deposited in the All-Union Institute for Scientific and Technical Information 4/27/90, No. 2274-B90. L. N. Toritsyn, Ogneupory, No. 4, 25-29 (1991)o L. N. Toritsyn, Ogneupory, No. 12, 9-13 (1988). L. N. Toritsyn, Ogneupory, No. Ii, 34-41 (1990).