REPRINTED FROM VOL. XII, TRANSACTIONS OF AMERICAN CERAMIC SOCIETY. 
 (Read at Pittsburgh Meeting, February, 1910.) 
 
 THE BEHAVIOR OF FIRE BRICKS UNDER LOAD 
 CONDITIONS AT A TEMPERATURE OF 1300°C. 
 
 BY 
 
 A. V. BLEININGER anv G. H. BROWN. 
 sige OF ILLINOIS 
 
 AMS 
 
ee ae SF a 
 
 TO7 
 wie 
 a > 
 
 THE BEHAVIOR OF FIRE BRICKS UNDER LOAD 
 CONDITIONS AT A TEMPERATURE OF 1300°C.! 
 
 BY 
 
 A. V. BLEININGER AND G. H. Brown, Pittsburg, Pa. 
 
 PRELIMINARY STATEMENT. 
 
 In this paper the writers are dealing with the definite 
 problem of the load-carrying ability of fire bricks, and 
 they make no claim that the test described by them will 
 discriminate as to the general usefulness or value of a 
 refractory. A material which might make a poor showing 
 in this test might be useful for many purposes, where the 
 streneth at the temperature employed is a minor consid- 
 eration. 
 
 In the study of the effect of heat upon clays and clay 
 mixtures, it is necessary to keep in mind the fact that the 
 phenomena of fusion are by no means confined to tem- 
 peratures close to the so-called melting or softening point, 
 but manifest themselves already at heats hundreds of de- 
 grees below the stage of viscous fusion. In fact, we can 
 say that clays and and clay mixtures do not possess a 
 definite melting point. In testing a specimen of fire clay 
 or of fire brick, fusion is said to have taken place when 
 the test pyramid or specimen has softened sufficiently so 
 that its edges are rounded, and it has bent over in the well 
 known manner. Asa matter of fact, for all practical pur- 
 poses, a fire brick would have been completely distorted, 
 and would have shown all signs of fusion at considerably 
 lower temperatures owing to its inability to support its 
 own weight. It is simply due to the viscosity of the mass 
 and to the fact that no load is imposed upon it, that it does 
 
 1 By permission of the Director, U. S. Geological Survey. 
 
 ] 
 
2 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 not show the distortion coincident with softening. It is 
 evident, therefore, that the determination of the so-called 
 melting point of a clay or fire brick, though useful in 
 differentiating between low and high grade materials, 
 does not offer a reliable means of expressing the entire 
 refractory behavior. 
 
 Thus Purdy’ found the fusion points of evidently in- 
 ferior clays to be quite high. He reports, for instance, a 
 clay of the formula 5.1 SiO,: Al,O, ° 0.046 Fe,O,- 0.09 TiO, 
 to have a “fusion” point corresponding to cone 32. Simi- 
 lar results have been obtained in the Survey laboratory at 
 Pittsburg. 
 
 The function of viscosity in extending the softening 
 period is clearly recognized by all who have worked with 
 silicates. 
 
 Thus Day and Allen,? in their investigations, briny 
 out this point very clearly. Doelter, and his students in 
 fact, have endeavored to determine tlie viscosity of differ- 
 ent silicates at various temperatures. Vogt, likewise, has 
 sought to correlate the viscosity of fused magma with the 
 sequence and velocity of crystallization. 
 
 Greiner? determined, by means of a specially con- 
 structed apparatus, the viscosity of mixtures of Na,Si0, 
 with various other silicates at different temperatures. 
 Thus, in a mixture of Na,O, SiO. and Al.O, he found that 
 the viscosity was greatly increased by the addition of as 
 small an amount as 2.5 per cent of Al,O,. This is shown 
 eraphically in Fig. 1, where the abscissse represent the 
 temperatures and the ordinates the relative viscosity. 
 Curve I corresponds to the formula Al,O,°12 Na,O: 15 
 SiO, and Curve II to Al,O,:9 Na,O:12 SiO,. If alumina 
 thus increases the viscosity of Na,SiO,, which is one of 
 the most mobile silicates, it is evident that the internal 
 
 iT. State: Geol. Survey, Bull No: 4, p. los. 
 
 2 The Isomorphism and Thermal Properties of the Feldspar, Carnegie 
 Inst., 1905. 
 
 ’ Inang. Dissertation, Jena. 1907, p. 17. 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 3 
 
 TRANS. AM. CER. SOC. VOLXII. BLEININGER & BROWN. 
 
 78509 Boe ~ 150° 12002 
 Temperature m %. 
 iriction of highly aluminous combinations must be very 
 high indeed. 
 
 On the other hand, the criticism raised as regards 
 melting point determinations, does not apply to silicious 
 mixtures containing a base-like lime. In testing silica 
 brick it has been found that the melting point is well de- 
 fined, as we should expect from our knowledge of the 
 calcium silicates. Such a brick will stand up well in the 
 fire, even under heavy loads, without deformation, at tem- 
 peratures at a safe interval below its melting point. There 
 is practically no flow until the melting point has been 
 reached, when it will suddenly fuse to a liquid of low 
 viscosity. This is probably the reason why silica brick 
 are now being employed in many places where load con- 
 ditions are an important factor, as in gas benches, ete. 
 
 In order to overcome the viscosity effect of the fused 
 mass it evidently seems desirable to bring about conditions 
 which will neutralize it. For this reason some investiga- 
 tors have found it advisable to apply a slight load to the 
 
4 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 specimens to be examined. In some cases a platinum 
 weight is used which is placed on top of the small cylinder 
 or cone in the furnace. Howe’ has suggested a furnace 
 illustrated in Fig. 2, which arrangement makes it possible 
 to measure the rate of settling or flow as well. 
 
 TRANS, AM.CER.SOC.YOL XI. BLEININGER Kk BROWN, 
 U x 
 
 y 
 y 
 y 
 \ ) 
 
 = 
 
 For the solution of the problem at hand, viz., the re- 
 sistance of fire bricks to load conditions, it was decided 
 not to approach the softening temperature proper, at which 
 the whole mass becomes a viscous liquid, but to restrict 
 our work to average furnace temperatures at which only 
 the more fusible silicates are softening. In the average 
 fire brick, containing from 80 to 85% of flint clay and 
 
 1 Metallurgical Laboratory Notes, p. 50. 
 
ELHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 5 
 
 20 to 15% plastic bond clay, it is obviously the character 
 of the latter which would determine the behavior of the 
 brick from this standpoint, since the flint clay within this 
 region shows no fusing action whatever. In order to bring 
 out the point sought, it was evidently necessary to apply 
 loads somewhat greater than those used in practice, 
 though it was realized that it must not be excessive. 
 
 DESCRIPTION OF TEST. 
 
 In preparing for this investigation the work of Lemon 
 Parker? was consulted, whose interesting results were 
 
 Ses 
 
 TRANS. AM. CER. SOC. VOL Xl.  BLEININGER & BROWN 
 
 See Shite, 
 rer re 
 
 2 Transactions American Ceram. Soe.. Vol. 7, Part II. 
 
6 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS, 
 
 found very suggestive, though in his work only small 
 specimens were used. 
 
 Since it was thought advisable to use full sized bricks, 
 a furnace was constructed as illustrated in Fig. 3, with an 
 interior space of 20’’x 20’x 21”, heated by means of eight 
 Fletcher burners, which were supplied with natural gas 
 and compressed air. The combustion gases escape through 
 holes along the back side of the furnace, which consists 
 of a movable dour made from a heavy wrought iron frame 
 lined with fire bricks. This frame is hung from two rollers 
 running on an overhead track, so that it can be easily 
 moved. The brick to be tested is firmly placed on end upon 
 a heavy fire clay block about 9 inches from the door, and 
 made to set plumb by means of fire clay mortar. The half 
 of a chrome brick is then placed upon the test specimen 
 and on top of it a long cylinder, made from high grade fire 
 clay which extends through a hole in the arch of the fur- 
 nace. The load is then applied upon a cast iron knife-edge 
 plate by means of a steel I-beam 9 ft. long, with a distance 
 of three feet between the pin taking the up-thrust and the 
 fulcrum, and six feet between the latter and the point of 
 load application. The weight of the beam is 155 pounds. 
 The weight applied consists of a wooden box containing 
 the necessary amount of iron shot. The beam was first 
 calibrated by means of a platform scale. The pin at the 
 one end of the beam passes through a cast-iron block slid- 
 ing in a channel iron provided with a sht and a heavy 
 fastening bolt. On top of the block two bolts are arranged 
 so that it can be lowered by loosening the fastening bolt 
 and tightening up the top nuts. In this manner the [-beam 
 may be kept level throughout the test, thus avoiding any 
 side strains upon the brick. 
 
 Through an opening in the door a thermo couple is 
 inserted, which is placed as close to the brick as possible. 
 The junctions of the element are placed in test-tubes im- 
 mersed in a beaker of water, the temperature of which is 
 read regularly. 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. T 
 
 Before beginning the regular tests it was, of course, 
 necessary to determine the best working temperatures and 
 pressures for the purpose in view, and in this the exper- 
 ience of Mr. Parker, as given in the article cited above, 
 was used as a guide. 
 
 The first point it was necessary to determine was the 
 rate at which a fire brick could be heated up, and the time 
 required until the interior of the test specimen had as- 
 sumed the temperature of the furnace. For this purpose 
 a hole was drilled to the center of a brick. A thermocouple 
 was then inserted and the hole well plugged up, leaving 
 the couple well covered for some distance away from the 
 brick. Another couple was placed close to the brick, but 
 not touching it. Fig. 4 shows the time-temperature curves 
 
 TRANS. AM. CER. SOC. VOL XII. BLEININGER & BROWN. 
 
 1300 
 
 oe Welle 
 Bee bette dak [er 
 roe Soh 
 
 RE LEE 
 Time /n Min. 
 resulting from this test, which illustrate that after 285 
 minutes, at the rate of heating followed, the inside tem- 
 perature was the same as that on the outside. Since it was 
 necessary to conduct the test with reasonable rapidity, ow- 
 ing to the limited hours during which power was available, 
 
8 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 the time-temperature curve shown in Fig. 5 was finally 
 adopted. 
 
 The temperature to which the brick was to be brought 
 and held for one hour in one series of tests was taken to 
 be 1800°C., based on the consideration that the more easily 
 fusible combinations of silicate mixtures vitrify and soften 
 within this heat region. In the second series the tempera- 
 ture was raised to 1350°C. The tests described in this 
 
 TRANS. AM. CER. SOC. VOL XII. BLEININGER & BROWN. 
 1400 
 00 
 [200-—> 
 
 ALU cares exec 
 
 | | 
 
 1000;—— 
 
 aY 
 4 
 
 Temperature (at 
 
 | | ! 
 + t . + 
 t T 
 } H 
 | 
 ) 
 | \ \ 
 
 +— 1 s Se SSR EO ee 
 | 
 
 ail Corio 
 zi 
 te 
 
 Eeesaes 
 
 tH 
 | 
 
 | dP asta Sn ENC, Nei Re 
 
 00 220 40 26) 280 300 320 34) 360 
 
 Time {na Minutes 
 
 article refer to the former temperature. As regards the 
 loads to be used, a considerable number of preliminary 
 tests were made with pressures ranging from 125 pounds 
 to 50 pounds per square inch, but in the first series a load 
 of 75 pounds per square inch was employed, and in the 
 second, 50 pounds. The present results refer to the 75 
 pound load condition maintained for one hour at 1300°C. 
 
 In addition to the load tests, the following determina- 
 tions were made: 
 
 we 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 9 
 
 en 
 
 Crushing strength of the bricks on end in the 
 cold condition. 
 
 Chemical analysis. 
 
 Softening temperature. 
 
 Porosity. 
 
 True specific gravity. 
 
 Decrease in water absorption on re-burning 
 to cone 12. 
 
 D> OUR go bo 
 
 1. The crushing strength of the bricks, placed in the 
 machine endwise, was carried on in the usual way, pains 
 having been taken to imbed the specimen in plaster. 
 
 2. For the purpose of the chemical analysis, a sample 
 sufficiently large was obtained by breaking up two half 
 bricks and crushing them to pass the 100 mesh sieve. The 
 disintegrated sample was allowed to flow in a small stream 
 over a powerful magnet until all of the particles of metallic 
 iron were removed. The analysis was carried on under the 
 supervision of Mr. P. H. Bates, chemist in charge of the 
 structural materials chemical laboratory, assisted by Mr. 
 A. J. Phillips. The methods advocated by Hillebrand were 
 used throughout this work. 
 
 3. The softening temperature was obtained by chip- 
 ping off specimens from the bricks and placing them in a 
 Fletcher gas furnace, using preheated compressed air and 
 natural gas. For the higher temperatures a small carbon 
 resistance furnace was employed, in which the muffle con- 
 sisted of a body containing 85% calcined alumina and 
 15% of kaolin, Fig. 6. This furnace was found very satis- 
 factory for the purpose. Seger-Orton cones were used for 
 the determination of the temperatures. The results were 
 checked in each case. 
 
 4. The porosity was determined by obtaining the 
 weight of a dry piece of the brick, the weight of the same 
 specimen saturated with water by boiling in vacuo and the 
 suspended weight. The results checked practically with 
 the porosity, calculated from the true and apparent specific 
 gravities. 
 
10 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 TRANS.AM.CER.SOC.VOLXI. BLEININGER X BROWN. 
 
 a Section A-A. 
 ar c& “yy Uy “f : 
 
 ne) ia 
 
 2" PRP 
 Re 
 
 VS 
 
 RS 
 S 
 
 SSS 
 
 Porous Material \w) clectrode 
 Fig ©. 
 
 5. The true specific gravity was obtained from 50 
 gram samples of the pulverized bricks, under the custo- 
 mary precautions as to the removal of air by boiling in 
 vacuo and making corrections for the final temperature 
 of the water. : 
 
 6. The original weighed bricks in this case were 
 tested for water absorption by placing them flatwise in a 
 covered tank, containing 114 inch of water, for 48 hours. 
 After weighing they were dried and burnt in a down-draft 
 test kiln to cone 12. They were then again immersed as 
 before, and the water absorption determined. 
 
 In this work 26 brands of fire bricks were tested, and 
 each manufacturer was requested to furnish 20 bricks. 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 11 
 
 The writers desire to take this opportunity for thanking 
 the various companies for their co-operation, and the in- 
 terest taken in this work. 
 
 RESULTS OF TESTS. 
 
 In Table I the chemical analyses, as well as the calcu- 
 lated empirical formula, are compiled. 
 
 The results of the physical tests are arranged in Table 
 II, and the data referring to the load test, as has been said 
 above, apply to the load of 75 pounds per square inch and 
 the temperature of 1300°C. 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 12 
 
 091+08NX0v0+0%N+0°H = OU: 
 
 | | 
 660°0| 200°0 | S$00°0| TT0°0 | 910°0 1 090°0| 820°0}| O'T | £622 ar°0| 00/240 8-0) 06" 9F°S| SV OF 
 00T°O | 800°0 | $00°0 | §00°0 | 610°0 | €90°0 1 SS0°0| O'T | SSP? Py ee 9°01-S6'T169°T 1 60° 68 
 Gol 0 | 600°0|9T0°0! 800°9 | T20°0 | 890°0| TILT’O| OT | $96°9 | 0LT°'0/| 82°0| 90°01 22°0| T° T | 9° T | 00°6T 
 ZST'0| SE0°0 | L100 | 020° fy L60°0|690°0) 820°0) O°T | T28°8 | 210°01| 08°0| €2°0| 6S°0 ;ao°T GeO h 6G 
 10Z°0 | 910°0| 200°0| 2€0°0 | 2g0°0 POT'01|090°0)| O'T | O86°E 83°0| 610 | 3h 0 8G°0 | SE.¢ | 9E TL | 99°82 
 69L° 0 | 900°0 | €T0° oat a) oe 04:09070 1 0° E0226 04°01) 86504-6504 SRO. Sh CSE 0S"6¢ 
 9010 | 910°0| 200°0 | 920°0 | F00°0 | £00°0 | 290°0! O°L | 220°% 98°0 | 1°0/ 8L"0 QE 0.) 06::0-) CLE GLP SPF 
 880°0) F60°0 Cae That 00: hoo 690°0) 0°T | O070°T | 92°0)08°T166°016¢6'°0) 07 TL) 02° | 86°21 
 G60 | 600°0 | 900°0| 280°0 | gF0°0 00T°0| T90°0| O'T | PSE°E. | 8T°0)} 8E°0/ 95°00) G2°0)] SF°S| €S'°T | 08 TE 
 80%°0| 200°0 Ae RL ale ESO:.0) OFF 2766) Se) S102) SS" 01.092 Os E 2 0-S9" Ser Vl Gece 
 TOT’ 01 €T0°0| ST0°01 2T0°01 710°0!| 270°0| 690°0| O'T | S6°T | 7E°0| 09°0)| 62°01 88°00) SFL] 80°C 1 72° FP 
 80¢6°0| FI0°0 | €T0°0 qa 180, cee 990°0| O'T | PE9°S 6&0 | &b°0 CG OPS LE PSS) 106° E179 > 9S 
 Ge 0| €10°0| STOO | 020°0 | €90°0) €TT°01920°0!] OT | 89T°S |92°0) 970) 92'0| CLT | 68°2| 76°T | 69 SE 
 89Z°0| 200° 0| 020° 0 i es GIT’0| 190°0| O°T | Lg9°4 cz'0 | 29'T 6P 0.) 98. 0:1 96° SSS) ESL Se 
 9T6°0 | 800°0| TS0°01 €70°0 | 20°00 | 060°0| S80°0| OT | T6S°% | 8T°0] €L°T | 19°0] 6F'0) 6S9°2] OF ' Z| 2S 98 
 L9T°O | 600°O | 660°0 ee aes 790701 2120°0| O°T | 96T°S | 27°00; 98°0/| 09°0| 97°01 STS] 86° TL | S9° SE 
 66°00} 600°0 | T20°0 660° 0 | 060° 0 0S0°0|S20°0!] OT | LH6°S | 06°0/29°0| T8°0) 6E°0) LET | $0°? | E8 rE 
 91T¢°0 | 6600 | €90°0 |} SZ0°0 | LT0°0 | 680°0| SOT’0| OT | OTL°€ | 07°00] TL°T | 62°0| 22°01] S0°S | OF S| 82° 62 
 SPL 01 TL0°0 | SZ0°0 | 820°01 Sh0°0| 680°01 S20°0! OT | @rT'S 86°01 6°0| S7'0| GOT | 86°T | 97'S | 6S TP 
 8610 | ¢80°0 G&0°0| TS0°0 | 610°0 090°01 790°0| 0°T |e9e°z |82°0192°T| 82°0| Th'0 68 °T}] 66°} 06° 8€ 
 8LT°0} 80°00! F700! FE0°0 | 9T0°0 | 9F70°0 | F90°0) OL | 988°% 16°0| LEP | 6S0 1-SEs OEP Ly 66 b 98 28s 
 69T°0 | $80°0 | €70°0 | €80°0 | €10°0 | 270°0| 290°0} OT | &7E°S | 79°0: LO°T| 29°00 | 62°0 | GET | 98° TL] SL °6E 
 1&Z°0| 6£0°01 080°0| 980°0 | 920°0 OFL O1-FSO05 0: OF Di S8P°S | Cr 0.-60>-E 1 SS:.01 2S: 0 OFF LET Ses US 
 STE" 0 | 090° 0 | 880°0} 880°0 | 260°0| G60°0 | 980°0}] OL | OFT’ L | 29°00) 6F'T | 82°00] 8F'01 GET] 29° 0)| SST 
 LG6¢°0 | SE0° 0 | 480°0 1 990° 0|1.660°01 280°01980°0!] O° | 282°2 | LE°0| 16° +! 88°00} 16°00) 6T £1 6h 0) ZF LT 
 GVE 0} VL0°0 | 620°0 | 9F70°0 | SF0° nee: aa a a OEt ate a a LE°0|ST'°T167°0| SS’ FT 
 \ | 
 
 Z a < a ra oa (et n ls 4 Ss 
 
 Ce ee ee 
 
 SLNATVAINOT | is 4 
 
 ee 
 Gy ee: 
 QZ" 99 |° Coed 
 SG Bees 
 Be aie 
 ee 
 eras 
 es 
 ee 
 92085 ws 
 gG° 9G)". GE 
 Lb 091 Wk 
 SLE SOAs i 
 +S a “GL 
 Gerona. SEI 
 LEO a Ot 
 GS cok eae 
 ee ae 
 So Va dhs 
 Le ese 
 Lens Cages 
 SoS ey 
 Sg eee 
 ee 
 Ne cy 
 P8 
 
 ‘SOSATBUY [BOTULBYO 
 T WTaViL 
 
13 
 
 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 8L°6 | Gt 0t 06°6L peysnsy LVGT peysnip 
 
 eonne ae eee 66, \@) 9 YT LOZL 86) 
 SF's Cro 06° IT % T6Z1 Ug 
 98°¢ 9),°8 29° ST % 0 SZ | 
 L6°T oP OT 68°31 abe POST 91g 8 | 
 ee°T 02°8 $9°6 és PESZl 148 
 £6°0 Zs OT GT IT ors TEST ors g 
 16°0 89°) 6G°8 7 SIZ £8 
 PI'S 78° OT 86ST % Gel 8 
 G0"e TS°Z1 9G° GT 91-81 PLZT 9T- 8 
 ta ed Jet aly ite eas sag? 
 6o'°T es") Z8°8 T GOZT 8 
 88°% P88 ZL OL ii Pest %), 
 6P'T 68° 88°8 9T-6 LILI 9T-G 8 
 00°S @G29 GG’ peysnig OfIL | peysnap 
 6h 'G 007g | 6F'OL | peusnap GhIL | peusnig 
 00°S Elf Cl Gre | a Sort AL 
 18°Z 9e°8 hess b 6LIT a 8 
 SPs | 09°6 | 80° ST | peysnig T6TT | peusnD 
 ee jos (ie | ow [ae] & 
 eo°T 22° GL°8 | % | a HS | 
 gig EG") T2°6r | % | T6TL 8 | 
 Crs |. SGTL 16°6 | peysnig O8TT | peysnig | 
 76'S Sa*g LP IT | poeWosiq 
 
 | Alpe OZ | peioisig | 
 
 : | 
 
 | 
 
 . . . . ° . ° . 
 
 — —. $$ 
 
 q q | 
 
 ee $$ eee 
 
 I MAOM- Primo tNW MAMA MON OME ON 
 NANATANANNANNNNNNNNMNNNNANN AN 
 
 WODMmAaAWMI AWWW HDD OO 00 N OD 10 9 
 
 —— Ee ————— en) 
 
 NI ra <H 
 oon 
 oo OD oD 
 
 68° 0 00°SE 68° ST peysniy ra a poysniy 
 . | . r 
 dpe gees [oases | ee | ge] fF 
 602 fog aw 558 ey = of 
 150 $45.5 732.3 5 a ae 
 5.25 rae < a's oS a. 7 oe 
 Pee | ooh | gh 53 a Be 
 ch es . @ ad 
 age? Sho4 ego 2 bs Cen 
 ang ~w i. Oo g S 
 gs esas ae 5 SS wy 
 4 Saas 36 ° aay ee 
 a 8 A 35 = of ar 
 o av a] o a mae 
 “ = et= 4 a. = 2 we 
 Bae ° 5 ° o 
 es 3 =a re) a | 
 
 sayouy 
 ON 
 
 +82] POT J10j0q 
 
 42H $0 14319 
 $2U0d UT 
 
 ‘dma surusjjos 
 surqsnig 
 
 Aso10g % 
 *pua 00 YoTNIg 
 
 ‘ut ‘bs sad ‘sqy ‘proo 
 
 *q3uo7s 
 
 AYABID DyIIEdg on, 
 
 ‘S1S80,,, [BVOISAUgG 
 Tl HIAVL 
 
14 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 'ANS.AM.GER.S0C VOL XIloge gages 
 
 Fig. 7 represents the fracture resulting from the 
 crushing test in the cold. The break is normal, and ayrees 
 with the theoretical considerations. In the following plates 
 the photographs of the fire bricks. after having been sub- 
 jected to the load test, are reproduced. 
 
 % 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 -——— a = — SS 
 
  TRANS.-AM. CER. SOC. VOLXI. ———BLEININGER & -BROWN 
 
 = Figs Ht, 12,13. 
 
16 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 VOLKL. 
 
 Figs 23,2 4, De, 
 
18 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 19 
 
 “TRANS. AM. CER. S0C. VoL Xl. BLEININGER & BROWN.” 
 
 Fig 3h 
 
 Fig oo. 
 
 ES 
 
 Irom these it appears that in the cases where the 
 bricks were badly distorted and crushed, in each case a 
 certain degree of softening took place, as is clearly indi- 
 cated by the curved surfaces. It is evident that these 
 bricks attained a viscous condition in which they were not 
 able to carry the load imposed upon them, though ‘the lat- 
 ter 1s small compared with the crusting strength at the 
 atmospheric temperature. In no case is the fracture as 
 Sharp as is shown in Fig. 7, but invariably clear evidence 
 of flow is produced. 
 
 In carrying on the load test as the temperature rises, 
 the beam is first raised, due to the expansion of the furnace 
 bottom and the brick, a quiescent stage is then reached 
 after which, from 1130 to 1290°, a weil defined deflection 
 begins, caused by the contraction of the brick. In some 
 bricks this deflection continues at a very slow rate, or 
 reaches a condition of equilibrium some time after the 
 temperature has been raised to 1300°. This kind does not 
 fail under the conditions of the test, and the later deflec- 
 tion starts the more apt is the brick to stand up. The 
 materials failing under this test show a more or less early 
 
20 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 g@, and the rate at which this takes place 
 increases with the temperature, till finally it becomes so 
 rapid that it is impossible to keep the beam level. [Failure 
 then is merely a matter of minutes, and takes place very 
 suddenly. In every case, therefore, softening precedes 
 failure. With a third kind of material (of silicious com- 
 position, low in fluxes) the rise in height remains station- 
 ary for quite a while, followed by a long quiescent period. 
 After the maximum temperature is reached a very slow 
 and gradual settle takes place, which is quite small and 
 decreases in rate as the temperature is kept constant. Such 
 materials remain perfectly straight, and shrink but a small 
 amount. 
 
 Undoubtedly some of the bricks tested would have 
 been deformed badly on long time tests, which it was im- 
 possible to carry out under the conditions of the plant, but 
 in several cases where the time of testing was extended 
 considerably a kind of equilibrium seemed to be reached, 
 after which no observable changes took place. It is very 
 likely that in these cases solution of the more refractory 
 portion of the brick took place, which stiffened up the 
 bonding material. | 
 
 Inspection of the failures shows plainly that the more 
 refractory flint clay has not softened to the slightest ex- 
 tent. The grains have lost none of their original identity. 
 They seemed to have slid upon each other, the bond clay 
 acting analogously to a lubricant. From this it follows 
 that, no matter bow excellent the major constituent of the 
 brick may be as to refractoriness, if the bond clay is defi- 
 cient the load carrving power of the product is impaired. 
 In other words, as far as this property is concerned, a fire 
 brick is not any better than its weakest constituent. 
 
 start in settling 
 
 It is observed that there is some rough relation be- 
 tween the softening temperature and the ability of the 
 bricks to stand up under this test, but it is also seen that 
 this is not true unqualifiedly. Test No. 16 is one of the 
 decided exceptions. The failures not only show a softening 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 21 
 
 point below cone 30 but also a certain high content of 
 fluxes, i. e., K,O0, Na,O, MgO, CaO and FeO. On the other 
 hand, while a brick may soften below cone 30, if its content 
 of RO constituents is low it will resist the conditions of 
 the load test without difficulty. In No. 24. we have an 
 illustration of this kind, and many silica brick show similar 
 results. Nevertheless, it seems quite clear that the load 
 test is in a measure a determination of the refractoriness, 
 in spite of the low temperature at which it is conducted. 
 
 TRANS. AM. CER. SOC. VOL XIl. BLEININGER & BROWN 
 
 pele fe tae eee 
 Spee oe er er 
 ae ee Se a PEP 
 
 pee PE eer ie 
 
 sice GAS an ie 
 
 | | nk ica 
 8 eRe sae 
 ie eee eae 
 
 B 
 peer) | t ® ia od oe che oe ce ae pee 
 ©) en 
 Spe «| 5 | 17'® 22 \. | | 
 5 pe ge ag el Ae Be | | ] Page eit | 
 6 } 
 181 i : ae 
 10} Nae | 
 | ao ae cRsaeeees 
 : Se Sie eae | 
 | | 
 
 Motecvlar L£ovty 
 
 see ee 5.0 i a .{ of 10,0 
 ofecules OTG , Alz% =/ 
 
 No. 19 is not properly placed in this chart, as will be seen by reference 
 to Table I.—Editor. 
 
 In ig. 34 the failures due to the crushing of the 
 bricks are plotted, and it is observed at once that the most 
 
22 BEHAVIOR OF FIRE BRICKS UNDER LOAD. CONDITIONS. 
 
 significant factor seems to be the RO content. It is not to 
 be expected that a sharp division line exists between the 
 area of failure and that of satisfactory load carrving abil- 
 ity, Since in some cases the FeO, which constitutes part of 
 the RO, may be present in the form of larger grains, in 
 which condition it is not detrimental. The fact is again 
 shown that neither the softening temperature nor the 
 chemical analysis alone produces the information brought 
 out by the load test. Given all three of these data, how- 
 ever, a very good presentation of the refractory resistance 
 of a fire brick is obtained. 
 
 Examining the question from the chemical standpoint, 
 and assuming that the fire brick body consists of a refrac- 
 tory constituent corresponding in composition to the kao- 
 lin formula, and of a more fusible cementing component, 
 the following method might be pursued, taking for example 
 one of the failures, say sample No. 4. The entire brick, on 
 calculation from the average analysis, possesses the em- 
 pirical formula: 0.019 Na,O, 0.030 K,O, 0.036 MgO, 0.026 
 CaO, 0.14 FeO, 0.054 TiO., 1.00 Al,O., 2.485 SiO... Assum- 
 ing that the alkalies are derived from feldspar, the alumina 
 corresponding to 0.019 -- 0.030 alkali is 0.049 equivalent. 
 Deducting this from 1 we obtain 0.951 equivalent of clay 
 substance, which contains 1.902 equivalent of silica. Sub- 
 tracting this from 2.485 leaves 0.583 silica. Beside 9.951 
 equivalent of pure clay base, this would leave the rest of 
 the constituents to form the more fusible mixture: 0.019 
 Na,O, 0.030 K,0O, 0.036 MgO, 0.026 CaO, 0.14 FeO, 0.049 
 Al,O;, 0.583 SiO... Throwing this into another formula 
 with RO equal to unity, we obtain: RO- 0.195 Al,O,° 
 2.322 SiO. It is the amount and character (fusibility 
 and viscosity) of this mixture which should govern the 
 refractory behavior of the product under load conditions. 
 I'rom a rough estimate, it would seein that this composi- 
 tion is analogous to a quite fusible clay or acid slag, with 
 comparatively low viscosity. The amount of this mixture 
 might be calculated as follows: 
 
BEHAVIOR OL FIRE BRICKS UNDER LOAD CONDITIONS. 23 
 
 0.049°37.35—= 1.83% AIO, 
 0.583°54.58 = 12.83% SioO, 
 .69% FeO 
 .54% CaO 
 .538% MgO 
 .02% K,O 
 .42% Na.O 
 
 2.485 
 
 oro C &8 
 
 20.86% 
 
 In this case, then, the fusible silicate mixture of the 
 formulua RO-: 0.195 Al,O,° 2.322 SiO, would constitute 
 20.86% of the total weight of the brick body, sufficient to 
 cause it to behave as it did in the load test. In this 
 theoretical speculation we must keep in mind that the 
 effect is more marked, owing to the heterogeneous mixture 
 of the assumed pure flint and impure bond clay. ‘The more 
 thoroughly the two materials are ground and blended to- 
 gether the better would the body behave in the load test, 
 until finally the effect would be that corresponding to a 
 single clay of the average composition, though even in this 
 case it would remain a somewhat inferior material, owing 
 to the high content of fluxes. The more impure the bond 
 clay, therefore, the more complete and intimate should the 
 mixture be. 
 
 An analysis of the relation between the initial, cold- 
 crushing strength and the load behavior, shows no appar- 
 ent connection, but the fact is brought out that low inital 
 streneth isa handicap. Bricks Nos. 4 and 11 are examples 
 of this observation. While No. 4 would have failed irre- 
 spective of its cold-crushing strength, the failure was more 
 complete on account of its weakness, and No. 11 in all 
 probability would have shown a very much smaller con- 
 densation, in fact, would have stood the test. 
 
 The hardness of burning in a general way is a factor 
 worthy of consideration. Although burning to a high tem- 
 perature cannot, in the nature of the case, effect any fun- 
 damental change, and cannot convert a low grade material 
 into a good one, our work has shown that well burnt bricks 
 stand up hetter than soft burnt products. This is due, not 
 only to the greater compactness of the hody, but also to 
 
24 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 the change in the composition of the bonding material 
 where such is used. In other words, hard burning will 
 cause the plastic clay, usually decidedly less refractory 
 than the flint, to dissolve some of the fine part of the better 
 
 material, thus increasing its own refractoriness and hence 
 
 its resistance to load conditions. For instance, No. 26 
 would have shown up better if it had been burnt harder. 
 
 To what extent other changes, such as the formation 
 of sillimanite at higher temperatures, would have a bene- 
 ficial effect, it was not attempted to determine. 
 
 For the purpose of selecting materials for Government 
 
 use, the tentative requirement was made that under the 
 conditions of this test a 9” brick should not shorten more 
 than one inch. 
 
 The results of the second series of tests in which the 
 load is 50 pounds per square inch and the temperature 
 1350°C., are very much like those obtained in the first 
 series, but appear to bring out more the question of flow 
 due to softening being somewhat less dependent upon the 
 initial strength of the brick. It is intended to use the lat- 
 ter conditions in all further work. 
 
 REFRACTORIES MADE FROM ONE CLAY. 
 
 As to fire bricks made from a clay material which is 
 sufficiently plastic so as to be used alone, it is clear that its 
 ability to carry loads depends simply upon its composition. 
 Some of the very best tests have been obtained from ma- 
 terials of this kind. This case is the simplest and needs 
 no particular attention, since a clay will stand up or fail 
 by virtue of its own quality. A pure clay of moderate plas- 
 ticity would be the ideal material for the manufacture of 
 fire brick. Since, however, such clays are usually not avail- 
 able, the question of judging them as to their composition 
 deserves some attention from the standpoint of their load 
 carrying capacity. From this point of view the most im- 
 portant consideration is that of the amount of fluxes. 
 
 At the high temperatures involved, the presence of 
 
 Ap 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 25 
 
 even a sinall amount of fluxes becomes a potent factor, and 
 hence it is far better to select a clay higher in silica and 
 low in fluxes than a clay possessing the silica-alumina ratio 
 of koalin but higher in fluxes. The effect of silica in lower- 
 ing the refractoriness of clays is commonly exaggerated 
 as far as practical results are concerned, and it is usually 
 not necessary to reject a clay on this account. However, 
 high silica and high fluxes make a dangerous combination. 
 From this it follows that a clay fairly high in fluxes, corres- 
 ponding for instance to 0.22 RO: Al,O, 2 SiO., would be 
 improved by the dilution with a silicious material low in 
 fluxes, or even by the addition of a clean sandstone, the 
 principal requirement being intimate blending and erind- 
 ing. 
 
 In regard to the fluxes present in a clay, the state 
 and size of grain of the iron is of importance, for it is 
 evident that coarser grains of iron’ minerals will do no 
 harm, though in the analysis they contribute their share 
 towards raising the amount of fluxes. An effort should be 
 made to determine the amount of such grains and to make 
 the proper correction in the analysis. 
 
 The fluxes are an important factor in the ability 
 of refractories to carry loads, inasmuch as they become 
 active at low temperatures, as has been shown in the table 
 of results, by forming easily fusible silicates. As the tem- 
 perature rises increasing amounts of silica and alumina 
 are dissolved, and it becomes, therefore, simplv a matter of 
 the amount of this fused matter and its viscosity, whether 
 or not a given refractory will stand up at the temperature 
 in question. As the temperature rises or the load is in- 
 creased, a point is reached where the entire body becomes 
 too viscous to retain its shape. 
 
 By grinding the clay as coarse as possible, conditions 
 are improved, as in doing this the formation temperature 
 of the fusible silicates is raised, since the process of solu- 
 tion is hindered, and by these means it may be possible to 
 bring the product out of the danger zone. In time, however, 
 
26 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 equilibrium conditious are approached closer and closer, 
 so that finally the brick may fail under the same condi- 
 tions under which they stood up at first. 
 
 BRICKS MADE FROM FLINT AND BOND CLAY. 
 
 In this case we are dealing with comparatively coarse 
 grains of flint and fine grains of bond clay. ‘These two 
 materials are mixed and blended as far as the processes 
 customary at the present time permit it. The bond clay 
 breaks up more readily, and the plastic mass produced by 
 it cements together the grains of flint clay and caleine. 
 Assuming that the flint clay is of good quality, the load 
 ‘carrying capacity of the product depends upon the bond, 
 aS has been shown above. It is evident that if the flint is 
 inferior the product suffers accordingly. These two ma- 
 terials may be considered from the standpoints of chemical 
 composition, vitrification range and fineness of grain. 
 
 Flint Clay. 
 
 This material, the geological origin of which is still 
 in dispute, is deficient in plasticity, although by very fine 
 erinding it may become sufficiently plastic to be molded. 
 It is of conchoidal fracture, often quite hard, and at its 
 best approaches pure kaolin in composition, being often 
 surprisingly low in fluxes. Its drying shrinkage is prac- 
 tically nil, but in burning it suffers as a rule a decided 
 contraction Jn exterior volume. Its refractoriness, when 
 pure, is very high. A brick made from flint clay (by fine 
 and long continued grinding) is able to carry high loads. 
 The vitrification range of flint clay has been well shown 
 in the paper by Knote, in this volume. As far as refractor- 
 iness is concerned, it is good practice to grind it quite 
 coarsely for reasons already indicated. The fact that oe- 
 casionally it is in part calcined does not affect its refrac- 
 tory behavior, but simply corrects the shrinkage suffered 
 in burning. : 
 
 The physical characteristics of flint clay are not al- 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. ay 
 
 ways a synonym for high refractory quality, as the writers 
 have seen elays of this type which are decidedly inferior. 
 
 Plastic Bond Clays. 
 
 In these materials a good grade of plasticity is desir- 
 able, though this is less important where their refractori- 
 ness is high. In clays of greater fusibility high plasticity 
 enables the manufacturer to cut down its amount to a 
 minimum. In the selection of a bond clay it is evident 
 that its vitrification range should be as long as possible, 
 i. e., the temperature at which it becomes dense and non- 
 absorbent should be as high as possible. The chemical 
 composition and inherent mineral structure are thus most 
 readily expressed by means of porosity-temperature curves, 
 such as have been frequently described in these transac- 
 tions, and which have been employed by Purdy in the 
 study of Illinois fire clays. In making such tests, the time 
 of burning should approach as much as possible the dura- 
 tion of the burns in actual practice. What has been said 
 of the chemical composition of clays under the head of 
 bricks made from one clay applies also here. The main 
 consideration is that the amount of fluxes be as low as 
 possible, irrespective of the silica or alumina content. 
 
 As.to fineness of grain, it might be said that the grind- 
 ing should not be carried any farther than is necessary to 
 develop good plasticity. 
 
 Improvement of Bond Clay. 
 
 If a flint-bond-clay mixture should prove unsatisfac- 
 tory as regards its ability to carry loads, the first step 
 would be to cut down the amount of plastic clay if possible. 
 If this should not be possible or beneficial, the plastic ma- 
 terial could be improved by grinding it intimately with a 
 portion of the flint clay to such a degree of fineness as 
 would insure thorough incorporation. This could be done 
 by dry or, preferably, wet grinding. In the latter case 
 it might be possible to pug the additional, coarser flint clay 
 into the slip, which naturally would be maintained as thick 
 
28 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 as possible. The finely ground flint would not remain 
 inert but would be sure to possess some plasticity. 
 Another alternative would, of course, be the sub- 
 
 stitution of another clay for the unsatisfactory plastic 
 
 bond. Since not infrequently silicious clays are at hand, 
 this might prove feasible, even though the ainount of bond- 
 ing material would have to be increased. Where the flint 
 clay is becoming scarce, it might be possible to find some 
 soft sand rock which could be ground into the plastic bond 
 clay, and which would thus improve its behavior under load 
 conditions. 
 
 By some such means as have been indicated here, it 
 would not seem a difficult matter in most cases to manu- 
 facture products fulfilling all reasonable conditions as to 
 
 crushing strength at the usual furnace temperatures. In 
 
 addition, by paying attention to the sizing of the ground 
 clays, So as to produce as dense a structure as possible, by 
 the use of less water or by adopting dry pressing, the load 
 carrying ability of the brick is increased. Likewise, burn- 
 ing at a higher temperature assists in bringing about the 
 same result. Experiments carried on by the writers have 
 shown that harder burning has increased the streneth of 
 fire bricks appreciably. 
 
 CONCLUSIONS. 
 
 1. The softening temperature of fire brick specimens 
 is not a safe criterion of the resistance to load conditions 
 at the average furnace temperatures. The same is true of 
 the chemical analysis. By means of both the so-called 
 melting point and the chemical analysts, the load behavior 
 may be approximated. This cannot take the place, how- 
 ever, of a direct test. The RO constituents (including the 
 iron as FeO) should be as low as possible, and for best 
 results should not exceed 0.22 equivalent. The content of 
 fluxes is more important than the silica content. 
 
 2. The test described appears to fulfill the require- 
 ments of a load test for fire brick, since the results are 
 
 t= 
 
 te; 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 29 
 
 consistent and admit of close checking. It is recommended 
 that the load be fifty pounds per square inch and the tem- 
 perature 1350°C. A No. 1 brick should show no marked 
 deformation and should not be shortened more than one 
 inch, for the standard nine inch leneth. The test estimates 
 in a measure the refractoriness. 
 
 3. From the results of this test it appears that the 
 amount and fusibility of the least refractory clay consti- 
 tuent govern the behavior of the brick under load. Im- 
 provement will result by grinding the less refractory clay 
 intimately with flint clay or even silicious materials, low 
 in fiux, improving the density, and by harder burning. 
 
 DISCUSSION. 
 
 Mr. Parker: I certainly take a great interest in the 
 paper of Professor Bleininger, as I at one time undertook 
 to make investigations of this very kind, but all were at 
 higher temperatures.and with smaller loads. We were 
 having trouble which was attributed to shrinkage, and I 
 was soon satisfied that it was from some other cause. I 
 would like to see these experiments carried a little further 
 and at higher temperatures, which would result in data of 
 very great value. 
 
 Mr. Hipp: I would like to mention that recently we 
 have done some work along this line, although our appli- 
 ances were nothing like Professor Bleininger’s. I would 
 like to ask whether Professor Bleininger thinks that the 
 viscosity temperature, the temperature at which the bricks 
 become viscous, has any direct bearing on the strength that 
 the bricks might withstand? 
 
 Mr. Bleininger: It appears that the viscosity of the 
 brick at lower temperatures is the principal cause of 
 failure. 
 
 Mr. Hipp: Wo you think burning the brick harder in 
 manufacture would increase the strength? 
 
 Mr. Bletninger: J think so. It would tend to hold it 
 up at higher temperatures. 
 
30 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 Mr. Ramsay: As 1 understand Mr. Bleininger’s re- 
 marks, the higher the crushing strength the better it will 
 stand the heat. Take for illustration a fire clay, grind it 
 very fine, and burn it ina kiln at a very high temperature. 
 Will that brick stand more hot crushing strain than the 
 brick that is not ground fine? 
 
 Mr. Bleininger: Given two bricks of equal refractori- 
 ness, the one with the greater initial compressive strength 
 will have somewhat of an advantage. I*ine grinding does 
 
 not necessarily mean a decrease in porosity, but rather the — 
 
 reverse. The pores are smaller but the porosity is greater. 
 
 Mr. Ramsay: With coarse ground material the dis- 
 tance between the particles is greater than with fine 
 ground. Under the heat conditions it takes longer to come 
 in contact as they are further apart. 
 
 Mr. Bleininger: We must remember that where flint 
 clay is used in making fire bricks we are dealing with two 
 materials, the flint clay and the bond clay. The latter is, 
 as a rule, very much finer than the former. By a suitable 
 combination of coarse and medium fine flint and fine bond 
 clay, the greatest initial streneth is produced. It is evident 
 that the more porous a brick is the less solid material it 
 has to sustain the load. The coarser the clay is the longer 
 it will resist the heat effect, but we must remember that 
 in furnaces operated continuously we have a long time 
 effect. It is obvious, however, that the bond clay cannot 
 be coarse since this would defeat the purpose for which it 
 is used. The weak point, therefore, is with the bond. If 
 the latter is of high grade the brick will stand under the 
 load, if not it will fail. But I have shown clearly how the 
 product may be improved if the bond should not be of 
 first quality. 
 
 Mr. Ramsay: I wish to bring out, if possible, what 
 relation the cold crushing strain will have to the hot crush- 
 ing strain? 
 
 Mr. Blewinger: There is very little relation. Ifa 
 brick is physically weak under ordinary burning condl- 
 
 eae 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. dl 
 
 tions and is defective in composition, you can improve that 
 material somewhat by making it. stronger, but it will not 
 be a fundamental remedy. 
 
 Mr. Ramsay: In other words, it one brick is soft 
 burned and one is hard burned, the hard brick will stand 
 the greater heat? 
 
 Mr. Bleininger: WHard burnt fire bricks seem to stand 
 up better than soft burnt. 
 
 Mr. Hamilton: In Professor Bleininger’s investiga- 
 tion he has found that the less refractory brick stood the 
 greater test when cold? 
 
 Mr. Bleininger: The cold strength, as I have said 
 before, has direct relation to the strength in the furnace. 
 We have tested highly refractory materials that showed 
 a good strength when cold. 
 
 Mr. Hamilton: And while the photographs do not 
 show up as badly in the hot tests, that is the less refrac- 
 tory brick. Now, we make a brick that we think is very 
 highly refractory. That would be a rough grained brick 
 containing a binder of 20% of plastic clay. That brick 
 would be very porous and I take it that the hot strain 
 would be very low, but at the same time that brick would 
 stand a greater heat than 1300 degrees where it did not 
 come in contact with a load. A brick containing flint and 
 bonded with 30% plastic clay would not show up as well, 
 but would bea satisfactory brick for the roof of a furnace. 
 The thing that I would like to find out is, would it be 
 better to make a fine ground brick than one coarse ground ? 
 Your experiment has not shown that. 
 
 Mr. Bleininger: I want to say that I think I have 
 been misunderstood. By grinding the whole brick mixture 
 finer you of course improve the standing up quality of an 
 inferior bond clay. But it is not necessary to do this. If 
 for instance you were to grind the bond clay with some of 
 the flint very fine, and then mixed this combination with 
 your coarser ground flint, as usual, you would obtain the 
 best result. I say distinctly that I am speaking of brick, 
 
32 BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 
 
 suitable for carrying loads at furnace temperatures, and 
 realize fully that a brick may stand up at a high tempera- 
 ture if no load is imposed, while it would fail in this test. 
 There is a very rough connection between the ability of a 
 brick to stand up under this test and its so-called refrac- 
 toriness. <A brick softening below cone 30 would not stand 
 much of a chance in the load test. 
 
 Mr. Parker: I think if Professor Bleininger goes a 
 little farther in his experiments he will find that the re- 
 sults do not agree at higher temperatures. Beyond 1300 
 degrees C. they would not agree at all. At least that is 
 my experience. 
 
 Mr. Bleininger: Weselected the temperature because 
 it is a very common furnace temperature. 
 
 Mr. Parker: JI was compelled to go higher, and I 
 found that one kind of clay standing unusually high tem- 
 peratures was not as good a load carrier at a minimum 
 temperature of 2400° to 2500° T°. as a less refractory clay. 
 
 Mr. Hamilton: The point is, | do not think we can 
 get a highly refractory brick sample and at the same time 
 a brick that will stand up well under load conditions, be- 
 cause you cannot introduce the bond and hold up with the 
 temperature. 
 
 - Mr. Yates: At higher temperatures in the making of 
 fire brick you take all the shrinkage out of your clay. 
 
 Mr. Hipp: I would like to say concerning the per- 
 centage of plastic clay used, that we made some tests 
 recently, trying different amounts, and found that two 
 bricks of practically the same refractoriness, one contain- 
 ing 30% of bonding clay and the other 50%, showed con- 
 siderable difference as to their ability for withstanding 
 the strain imposed. I would like to ask Professor Blein- 
 inger whether in testing two fire bricks, each containing 
 a different amount of plastic clay but the porosity about 
 the same in each brick, the crushing strength would vary 
 to anv great degree? 
 
BEHAVIOR OF FIRE BRICKS UNDER LOAD CONDITIONS. 33 
 
 Mr. Bleininger: IJ cannot answer this question satis- 
 factorily. 
 
 Mr. Ramsay: I think you have stopped too soon. We 
 know your position in this line of work, and our customers 
 would be apt to take your opinion rather than ours, and IL 
 think sometimes your conclusions are erroneous. I think 
 that sometimes the clay used in bond has some effect upon 
 the brick. 
 
 Mr. Blemminger: Of course, we stand back of our re- 
 sults, as they have been made with care, and I cannot re- 
 tract one word of what I have said. When Mr. Ramsay 
 speaks of the effect of the bond clay, [ wish to call atten- 
 tion to the fact that this is the very thing which we have 
 emphasized. 
 
 Mr. Parker: I would like to say that at higher tem- 
 peratures Mr. Bleininger will find different results. At his 
 temperature they agree with my experiments. You will 
 find at higher tempertures that an ordinary fire clay prom- 
 ising good results as a load carrier, though still far from 
 its fluxing point, may lose its efficiency in this respect, 
 while the more refractory clay will prove more efficient 
 with a moderate load to a temperature much nearer its 
 fusion point. 
 
“a 
 
 =a nS 
 rae a 
 Aah? ae 
 — eer a 
 “ot eh 3” 
 
 y 
 ~~ 
 7 
 
 - 
 
 iS ag ©. 
 
) 
 
 | 
 
 106076166 
 
 a 
 
 30112 
 
 INNA 
 
 Leip yy eter rt tht