DEPARTMENT OF ENGINEERING RESEARCH UNIVERPITY OF!MICHIGAN ANN A RF3OR report )on TII7 C[{ARACTIRiSTICS OIF A CAST FORD ALLOY AT IHIGHf TEMPERATUPRE Anrproved by Vr.tten by A. E. White J. W. Freeman Project Number M643 Report rNumber 1 for Ford Aotor Company of Canada, Ltd, Auxust 24, 1946

I~~~~~~~~~~~~~~~~~~

THE CHARACTERISTICS OF A CAST FORD ALLOY AT HIGH TEMPERATURES The Ford Motor Company of Canada had developed a cast alloy which was considered to have properties suitable for applications at high temperatures requiring exceptionally high strength. The University of Michigan was asked to determine the properties of the alloy at high temperatures, particularly as related to the requirements of gas turbines. The nominal analysis of the alloy was stated to be 20Cr-lONi-30Co-5M.o-5Cu-0.25Boron with low carbon. Rupture tests at 1500~ F were used for the preliminary evaluation of the alloy. The excellent properties at these temperatures led to a more complete evaluation by rupture tests at 13500,. 16000 and 18000 F. Check tests were made at 15000 F on two additional heats as a measure of the uniformity of the alloy. In addition a number of supplementary tests were made at the request of the client. These included creep tests at 15000 F to1ether with tensile, impact, thermal expansion, and oxidation tests at elevated temperatures. Miscellaneous characteristics?riefly investigated covered fatigue properties at room temperature, forgeability and structural stability after creep and rupture testing. The investigation was terminated, however, before all the properties were completely evaluated.

2. CONCLUSIONS The principal advantage of the Ford Alloy appears to be strengths at high temperatures comparable to those of the best available precision cast alloys combined with machinability, 7ood ductility and comparatively good stability characteristics. These properties are sufficient to warrent considerable interest in the alloy for gas-turbine blades. The strength characteristics were similar to those of such precision-cast alloys as Vitallium, S590, S816, 422-19 and X-40. The relative properties varied depending on the temperature and test. The tendency was for the precision cast alloys, p rticularly 422-19 and X-40, to develop higher strengths as the temperature increased to 15000 F and above. There are certain limitations to the data on the Ford Alloy: 1. The effect of section size of casting on properties has not been evaluated. Tests were confined to specimens cut from 12-inch long castings with a 8~-inch O.D. and a 1 1/8-inch wall. For this reason all comparisons have been based on the highest values reported for other alloys 2. No consideration was given to heat treatment, a factor of considerable importance in developing the best properties of some of other alloys. 3. Some erratic results, particularly in tensile and creep tests, indicated that the prooerties may be affected by factors not evaluated by this investigation. The rupture tests on three heats at 15000 F, however, showed good reproducibility between heats. 4. The oxidation resistance of the alloy at temperatures above 1500~ F was erratic and apparently can be influenced markedly by factors not understood.

3. MATERIAL INVESTIGATED Three castings were supplied in the form of tubes apwroximat ely 8~ inches in outside diameter with a 1 1/8-inch wall and 12-inches long. These had been cast in a permanent mold. Apnroxi. ately l/ inch had been machined off both the outside and. inside surface of the tubes. The calculated oercentage analysis was reported to be: Cr Ni Co Mo Cu B 20 10 50 5 5 0.25?An analysis for carbon at the University gave the followin_ value s. Casting Number Carbon ~ 358 0.056 359.078 360.040 Tie three castings were made from separate heats on dif'ferent days and were fromn Heatt Numbers 358, 359 and 360. They were melted in an induction furnace from stock orepared in an ilectric furnace. X-ray inspection showed sound castings. A few small cracks around the ends of one casting were revealed by zyglo inspection.

TEST PROCEDURE The three castinas were cut up into suitable strips lengthwise to the axis of the tubes. Most of the tests were made on Casting Num-ber 358 with some check tests on the other two castings. In eome cases enough material was not available from N.1umber 358 and the test material was taken from the other two castings. The strips were cut from the casting by means of a high speed cut-off wheel using plastic bonded abrasive discs. The actual specimens were machined from these strips with stEndarc1 hirh-sneed steel tools in the ordinary manner. The alloy wa.s not easy to machine but a mechanic experienced with heatresistin austenitic alloys had no difficulty shaping, turning, thnreading and drillirn the material. Standard 0.505-inch-diameter specimens were used for tersi e, rupture and creep specimens. These tests were conducted in a.ccordance with procedures recommended by the American Society for Testing Materials. A brief description of tests, where necessary, is included with the results. Data on thermal conductivity of the alloy as reported -vy Professor J. M. Cork of the Department of Physics from a sep-?rately sponsored investigation was included in the report at the request of representatives of the client.

5 Several investigations were terminated at the request of the client before completion. Specimens had been heat treated and machined for rupture tests Cat 15000 F to show the effect of aging for 50 hours at 1350~ and 1500o F prior to testinv. The creep test data were not complete. W.;eldability tests werie started but terminated before any data were obtained. A <<ore thorou, h investigation of tensile and impact characteristics v'ould have been desirable to show the degree to which the values reported are typical. RESULTS The data obtained from the various tests conducted to evaluate the alloy are summarized as follows: Tens ile Tests Tensile tests were run at room temperature, 1350~, 15000, and 1800~ F with the results sunmmarized in Table I. The stressstrain curves are shown as Figure 1. Stress-strain curves were not obtained at 18000 F because there is practically no range of proportionality between stress and strain at this temperature. All threo heats were tested at room temperature and 1500~ F and only two heats at 1350 and 18000 F. The graphical summary of the data in Figure 2 indicates:

TABL I S''RT-T: _PT- T7 t.:ITi:i;,PB0P':RTi T FOR FqO FD aLLOTY AT ROFO' TE".PF'RATrPR:, I35J), 1500~, and 1l800F. Offset Yield Etrernth Pr,);Drtional Reduction C(Th tI n?.emp. Tensile Strensgth nsi Lir:it Etonration of Area No. F psi.2t 0 I20Kf psi n2 In. a 0 _' 358 Poom 91,100 31,000 49,500 54,000 15,000 18.0 18.7 359 Room 122,500 42,000 58,000 62,000 15,000 25.0 21.5 3&O Room 84, 500 38,000 51,500 57,500 20,000 9.5 9.2 3(- Roomn 88,000 42,500 55,500 59,500 17,500 12.0 11.8 358 1350 57,500 33,000 38,000 39,500 25,)00 15.0 23.4 359 1350 58,500 35,0'0 40,000 42,000 27,500 12.0 19.7 358 15003 50,000 32,000 36,500 38,000 22,500 11.0 14.5 3s9 1500 55,000 33,500 38,500 40,000 25,000 11.0 17.4 360 1500 47,75 0 0 33,500 35,000 17,500 16.0 27.0 358 1800 20,350 ------ -- 15.0 1.4 359 1800 19,500 - ------ — 12.0 13.6 O~

7. 1. Considerable variation in tensile properties between heats at room temperature. 2. Very little change in tensile strength and yield strength between 13500 and 1500~ F. 3. Very little change in ductility with temperature. 4. Higher proportional limits at 13500 and 15000 F than at room temperature. The surface of the specimens after testing showed that the gage section of the 359 specimens were largely composed of five grains. The Heat 358 and 360 specimens were largely coarse grained and had a dendritic fracture. The variation in grain size and probably an accompanying variation, in orientation are D'resumably responsible for the variations in tensile properties and modulus values. Rupture Tests In the rupture test, tensile specimens are held at temperature under a constant stress until actual fracture occurs due to creep. A series of such tests are usually run at each ter:~prature unrc'er decreasing stresses so as to cause fracture at several time periods up to about 1000 hours. When stress is plotted against time for rupture on logarithmic coordinates a straight line is obtained, or if the test material is unstable two intersecting straight lines result.

g. Rupture tests were conducted on specimens from Casting Number 358 at 1350~, 1500~, 16000 and 18000 F for time periods to establish the stresses to cause fracture in 100 and 1000 hours. The data are assumbled in Table II and the stress-rupture time curves are shown as Figure 3. The elongation of the specimens after fracture are also shown on this figure. In addition a limited number of tests were made at 1500o~ on specimens from Castings 359 and 360 to indicate the degree of reproducibility from heat to heat. In general this was as good or better than is normally found for such alloys. Thre ind@ications are that Heats 358 and 359 had nearly identical properties, except for one erratic test for 359. The limited number of tests indicated that Cast 360 would have a slightly flatter curve than the other two with the result that its rupture strength would be slightly lower at short time periods and slightly higher at longer time periods. The curves of Figure 4 were prepared to show the effect of temperature on the 100-hour and 1000-hour rupture strengths of the alloy. In one test at 1500~ and one at 1800~ F excessive oxidatlon occurred and influenced the results, particularly the ductility of the specimens. One test on Casting 358 under 17,500 psi was discontinued after 1230 hours because the establishment of the stress-rupture time curve by subsequent tests indicated that

9. TABLE II UPTURI.E T'vT CHARACeTi RISTIC$ OF FORD ALTLOY Test Runture Elongation Reduction Casti.ng Ter o. Stress Time in 2 In. of Area u-Tmber _ nsi hr % 5'A ~* 1350 45,0)0 39 9.5 17.4 37,) 0 3(0 17.0 24.7 34,00 7 49 1 (.0 29.8 35~ 1500 25,000 103 12.5 18.3 23,000 20A 13.0 23.9 21,000 508.5 12.5 18.3 17,500 Discontinued at 123I) hours 359 1500 25,000 130.5 13.0 20.6 21, -,),J 259 27.0 30.8 21,d000 507 12. 5 17.4 360 1500 25,003 42.5 23.0 28.9 25,000 43 17.5 33.0 21, 00 *421 - 6.5 23.4 358 1-00 17,500 108 16.0 21.3 15,000 440.5 12.0 22.7 I3, 03) 8Pf4 14.0 21.5 358 1800 10,000 11 15.5 22.0 6000 *214 * 5.0 18.0 5000 584 6.0 6.6'7-7xces sve oxidation. rU-Pl TVURE STRENGTITIS Tenoerature Stress for Rupture in Indicated Time Periods...._.,_.___ 100-hours 1000- hours 13.50 41,500 33,000 1500 25,000 19,500 1600 18,000 13,500 180 h6800 4500

the time for rupture would be excessive for the purpose of this investi ation. Creep T'ests The purpose of the creep test is to obtain data on the relationship between stress and rate of deformation which can be used to select design stresses for permissible total deformations in specified time periods. The data obtained from the test consists of a curve of total deformation versus time. These curves have a characteristic shape in which there is a certain amount of initial deformation upon application of the stress. The specimen continues to elongate (or creep) under constant stress. The rate of elongation usually decreases for some time and eventually reaches a constant rate. If the stress 1s higjh enough, or time long enough, the rate may again increase until failure occurs. The initial period of decreasing rate of elong,.Etion is usually referred to as "first-stage" creep; the constant rate period as "second-stage"; and the final period of incred.sing rate as "third-stage" creep. In the determination of the creep characteristics a sernes of tests under different stresses are made at any one termerature. The second-stage creep rates are measured from the time-elongation curves. A plot of such rates against stress usually gives a straight line on logarithmic coordinates, at least within the range of creep rates between 0.00001 and 0.0001

11. percent per hour. Such curves are used to select design stresses for a material. The tests require high sensitivity extensometers ind?.r% usIually run for at least 1000 hours. The stresses are considerably lower than those used for rupture tests and the total deformation during the 1000 hours is'usually less than one percent. Three creep tests were conducted at 15000 F on speci:.nens from Casting 35g. These gave the time-elongation curves of Figure 5. The minimum creep rates, at about 1000 hours, from these tests are plotted against stress in Figure 6. These three tests indicate a creep strength for 0.00001 percent per hour of about 10,000 nsi. This creep strength is commonly extranolated by engineers as the stress to cause 0.1 percent deformation in 10,000 hours or 1 percent in 100,000 hours. The first test was run under a stress of 10,000 psi anrd gave a creep rate of 0.000025 nercent per hour. On this basis it appeared that tvwo additional tests at 9000 and 12,000 psi would fairly well establish the stress-creep rate characteristics of the alloy. These two tests, however, gave lower creep r-ites than were predicted by th'e first test. Because no further tests were run the creep characteristics were therefore not fully rstablished at 15000 F. Creeo rates are very sensitive to structural variations. Erratic results are th'erefore not unusual in cast materials esrecially when the stress-creep rate curves are as flat as is indicated for this alloy at 1500~ F.

12. Impact Tests The following impact strengths were obtained from a limited number of Charpy tests on V-notch specimens: Test Charpy Impact Casting Temperature Strength Number (OF) Ift-lbs) 358 Room ^ 17 359 n 18 360 n. 8 359 1350* 38 360 " 30 358 1500* 35 _360 _ 32 *Specimens were held at temperature for one hour before testing. The impact strengths at elevated temperatures were good. Those at room temperature were somewhat low especially in the case of Casting Number 360. These single tests on each casting should not be considered as having established the impact characteristics. Ordinarily at least three tests on each material are required. Thermal Expansion Thermal expansion curves were determined using the apparatus and procedure described in American Society for Testing Materials "Standard Method of Test for Linear Expansion of Metals", A.S.T.M. Designation: B95-39, page 934 of the A.S.T.M. Standards, 1944, Part I, Metals. The curves are shown as Figure 7 and the

13. calculated average coefficients of thermal expansion were: Temperature Average Coefffcient of Thermal Expansion Fange (inch per inch per OF).(~F) -inch specimen 2-inch specimen 70-200 0.0000074 0.0000075 70-400.0000077.0000078 70-600.0000080.0000081 70-800.0000083.0000084 70-1000.0000086..ooooo86 70-1200.0000089.0000089 70-1400- --------.0000092 70-1600 --------.0000096 70-1800.0000100 The test on the four-inch specimen was in accordance with the specification and pave greater sensitivity. The two-inch specimen was used at some sacrifice in sensitivity in order to obtain nroner temoerature distribution at the higher temperatures when It was necessary to remove radiation shields used at the lower tTrheratxres. The expansion characteristics are intermediate between those for frrritic steels and for 18-8 steel. Oxidation Tests Simple oxidation tests in air were performed by heating small samnles in ordinary muffle furnaces for 1000 hours at 1350~, 1500~ and 18000 F. The samples were small cylinders 3/4-inch in diameter by 1/2-inch high. These were machined and polished througlh 2/o paner. Prior to testing the surfaces were carefully cleaned and the samples weighed. The samples were supported on

24. 20Cr-80Ni alloy wire in the furnace. Six specimens were tested at each temperature. Three of these were removed from the furnace and cooled to room temperature once a week during these tests. After exposure to air for 1000 hours attempts were made to remove the accumulated oxide from the surface to determine the metal oxidized. The following results were obtained: 13500 F -. not measurable. 1500 T ---— F not measurable. 1800~ F continuously heated, the loss in weight was 3.2, 5.8, and 9.1 grams per square inch of original surface on three of tihe specimens. 1800~ F ----- intermittently heated, the loss in weight was 0.22, 0.20, and 0.16 grams per square inch of original surface. The loss in weight at 13500 and 15000 F was reported as not?reasurable because a thin dark oxide film was Dresent which could not be removed by the method customarily used. This resulted in a slight gain in weight. The work was terminated >efore a suiteabe procedure was developed for removing this thin oxide. The degree of variation between individual samples at 1*O0~ F was more than is usually expected from such tests. Likewise tone smaller amount of oxidation of the samples cooled to room temperature once a week was unusual. Ordinarily samples intermittently cooled show a greater loss in weight due to cracking off of scale during the heating and cooling.

15. F.? t i.ue Tpsts Fatigue tests were conducted at room temperature on specimens from Casting Number 358. The machine used utilized the symmetrical loading on a rotating beam principle. Specimens were of the tph e -r shown in FP gure 8. The results obtained are given below and are shown as a S-N curve in Figure 9: Stress Stress Peversals (psi)..for Failure 60,000 184,000 45,000 756,000 42, 000 843,000 35,000 8,605,000 33,000 6,952,000 2g,000 59,705,000 without failure Thie S-N curve indicates the following fatigue strengths: Stress Cycles (psi) 106 41,000 107 34,000 10~ 28,000 (approximate) Comnparative d n ta at room temperature could not be found for other new heat-resisting alloys. FortTin Tests A very simple forging test was made on samples machined from Casting Number 360, using the following procedure:

16. 1. C-lrinders, 1-inch in diameter by 2-inches long were machined from strips cut from the tubing. The surface was finished by grinding and the ends were ground 2. _.The..ylinders r -': preheated for about, 1 minutes at 1500~ F and then transferred to a natural gas fired fLraee at the desired temperature and held 20 minutes after the furnace came back to temperature. Temperatures were measured by a thermocouple imbedded in the center oF a dunmy sample placod beside the test specimen. 3. The specimens were removed from the furnace and an attempt wcas rade to forge dowvn to one inch in height in one heat on the flat ends with a small mechanical hammer. The results obtained are summarized as follows and a ohotoraoph o. thle samples is included as Figure 10: Number of Decrease in Initial Forging LHammer Sample ITeight Temn. (oF) Blows (inch) Remarks 19.80 49 1-1/64 One crack - sample was probably too cold before forgin!r was stopped. 2050 21 5-3/64 A few cracks appeared after last blows. 2135 10 60/64 Cracked badly from start of forging. 2240 8 45/64 Sample burst under first few blows.

17..A a further test an attempt was made to forge similar saplles down to rod by hammering the round surface. From the nrevious tests it was estimated that the maximum forging temrnnerature was not over 20500 F. In the first attempt with an initial forgding temperature of 20600 F the sample immediately broke u. Ina second attempt using an initial temperature of 190~0 F and by reheating four times the sample was drawn down to a 1/2 x 1/2 inch square rod. This rod had cracks in one end as is shown by the photograph in Figure 10. Apparently the alloy can at best only be forged with difficulty and considerable trouble with cracking will be encountered. The upper limit of temperature in forging is about 20000 F and at these temperatures the process is extremely difficult because the metal is strong and rigid. This work did not take into account grain size and orientation or segregation effects which mirtht be rresent in an ingot. weldinr Tests An investigation of the weldability of the alloy was sttprted and then terminated before any results were obttained. Double-V end 2ingle-V welds were made with 18-8 + -'To steel rods on samples the full thickness of the original tube. No examination was made to determine the quality of the welds.

18. Thermal Conduct ivit Samples were submitted to Professor J. B. Cork of the Universityfs Department of Physics for a determination of the thermal conductivity of the alloy. Professor Cork made this determination as separate investigation directly for the Ford,,otor Company. At their request, however, his report is incThlded in this report as follows: "The thermal conductivity of a metal specimen submitted by the Ford Totor Company was found to be -- 0.035 Calories per Cm.2 per sec for a temperature gradient of one degree C. per Cm. This value is believed to be accurate to better than nlus or minus 8 percent." J. x!. Cork Feb. 25, 1946.,tabilitv Characteristic s The effect of exposure to stress at elevated temperatures were briefly studied by means of physical tests at room temperature and by metallovraphic examinations. hyvsical Properties at 7oom Temperature: Tensile and imoact tests were made at room temperature on specimens after creep testing to determine if the material was appreciably affected by exposeure at 15000 F for approximately 1000 hours under stress. The results obtained, Table III, show a considerable loss in room

LTAI;3LE III EFF-i'CT R C REEP TEE:'; IG AT 1500~ F ON Tl1TF, ROOM-TEMIPERATURE PHYSICAL PROPERTITS OF FORD ALLOY CASTITG NUM..ERP. 358 Creep Testinr Conditions Tensile Offset Yield Strength Proportiorn- Elongation Reduction Temp. Strecs Ti-e Strength (psi) al Llmit in 2 in. of Area (OF) (nsj! (hr) (psi).1..0.2 (p si) () {) original manterial 91,100 49,500 54,000 1.5,000 18.0 18.7 1500 9000 1078. 103,000 58,500 61,500 35,000 6.8 1500 10/,O)0 1173 89,000.47,000 61,500 35,000 6.0 7.3 Tiodified Izod Impact Strength*: (ft-lb.. 1500 12,000 1150 6.0, 6.0 *Specimers were 0.365-inch square rith t O.50-inch deep V-notch. H

20. temperature ductility and impact strength with some gain in proportional limit and yield strength after creep testing. The tensile strengths are erratic and indicate that the material was non-uniform. Duplicate impact tests on identical specimens were not made on original material before testing. The standard Charpy bars gave 17 foot-pounds (see page 12). On the basis of experience with tests modified Izod specimens would therefore have given considerably higher values than the 6 foot-pounds obtained from the creep specimens. The modified specimens used were the largest square that con b'e machined from the 0.505-inch diameter gage section of n creep specimen. Two specimens were obtained from the one creep specimen for duplicate tests. Metallographic Examination: A very limited metallo-.raphic examination was made of the original materials, a creep snecimen, and the specimens from the longest duration rupture tests. The principal object was to evaluate the effects of testinc on the structure. The etching characteristics of the alloy were considerably different from alloys for v.hich experience was available and the results were not entirely satisfactory. No attemnt was made to identify the various constituents. The structures of Heat 358 in the as-cast condition, after creep testing at 1500~ F, and after rupture testing at 15000,

21. 16000 and l800~ F are shown in Figures 11 through 15. The samples were etched electrolytically with-Aqua Regia in glycerine. The structures of the rupture specimens at the fracture and at the surface about 3/8-inch from the fracture are shown at 100 diameters. Structures at 1000 magnification were not shown for the rupture specimen because of the similarIty to the creep specimen. The information obtained from this examination is summarized below: 1. No evidence of pronounced precipitation normally found in such heat-resisting alloys was observed. Almost all etching reagents tried resulted in staining in a manner suggestive of the presence of very fine or submicroscopic precipitation. This was most pronounced in the 16000 F rupture specimen and least severe in the 1800~ F specimen, a characteristic suggestive of fine precipitation. 2. The fractures were largely transcrystallihe although considerable intergranular cracking was present'on the surface of the rupture specimens adjacent to the fracture. 3. The degree of surface cracking increased with testing temperature. The cracks were oxidized. It is impossible to determine if the cracking or oxidation came first. The probability is that both acted to initiate fracture. The cracking and oxidation are not unusual. In fact most alloys show a greater tendency for intergranular cracking in rupture tzesting than was shown by these specimens. This pre'sumably accounts for the good hot-ductility in the rupture tests.

22. DISCUSSION OF RESULTS Ingeneral the properties of the alloy were comparatively 1ood especially at 13500 and 1500~ F. The following tabulation compares the ruoture strength with those of several cast allovs considered to have outstanding properties: Temp. Stress for Rupture in Indicated Time Periods Allor (F) 100 hours 1000 hours Ford 1350 41,500 33,000 Vitallium. 1350 36,500 28,000 422-19 1350 47,000 36,000 X-40 1350 45,000 33,000 Ford 1500 25,000 19,500 S590 1500 21,000 15,500 S816 1500 29,500 23,000 Vitallium 1500 24,000 15, 000 422-19 1500 30,000 23,500 X-40 1500 30,000 26,000 S:.0 1 -00 12,5,0' 0''.~r; 18,00, "8 o:,' o S. 811":i l1..... S'590 1800 8,000 5,800: 5816 1800 10,500 7,800 Vitallium 1800 9,400 7,000 422-19 1800 10,000 7,100 X-40 1800 11,300 9,800 The values at 135G0 and 18000 F were taken from a paper Pnttled F;JIi -h Temperatuure Al.oys Deve!onerd fnr Aircraft Turbosllnerchargers andr rqas Turbines" by J. W. TFreeman, E. F.. Peynrolds -nc A.. F J. tVhite. Those at 1500~ and 16000 F wiere t. ken fro., a'~::~-l-, entt1i*r ""~at'-odistin: let,:Ils co$r Cao s Tu. ine Oarts" **h T-,i y;;, C t`j- S T T it

23. The paper by Cross and Simmons presents creep data at 15000 F for the alloys which indicates the following approxi-.-te comparative creep strengths: Stress for Creep Rate of Allo 0.00001, per.ir. Ford 10,000 Vitallium less than 6000 422-19 9,500 X-AO 14,000 At 13500 anr 1500~ F the Ford Alloy has equally as good or better rupture and creep strength than precision-cast Vitallium, S590 and 422-19 and is only consistently outranked by X-O and c ast S816. On the basis of rupture strength it is not quite as good at 16000 and 1800~ F. In appraising these comparative values it should be recognized that the other alloys Twere tested in the form of precision cast 0.250 or O.505-inch dianmeter specimens. The difference in casting procedure and size between them and the Ford Alloy may have had a pronounced effect on the relative properties. At the oresent time data are not available on which to base any esti-ation of this effect. In addition the values for the precision cast alloys are the highest reported for these alloys. Tests on several heats have.given a range in pronerties for all of tile alloys. Furthermore the high properties in some cases may have been due to special heat treatments, a factor not investiga&ted for the Ford Alloy.

24. The F crd Alloy has two exceptionally good features. One is its uniform ductility with time in the rupture test. The other is the fact that it is fairly machinable. In so far as the data obtairned in this investigation are concerned the following comments may be of interest: 1. Tlh.- tenSile and creep tests show considerable nonuniformity between castirn s and between specimens from the same castinp. This finding may be indicative of considerable variation in properties with section size and casting conditions. This non-uniformity apparently did not have much affect on tensile properties at elevated temperatures or on ru:oture strength. The agreement in rupture test results between three heats was very good. The creep test, however, is much more sensitive to non-uniformitieS, which accounts for the erratic creep rates and total deformation oP the creep curves. 2. In general the stability characteristics under the conditions of creep'and rupture testing were better than those showrn by -many heat resisting alloys. Greater changes in physicr.l properties at room temperature, especially impact strength, are not uncommon. The visible microstructural changes were practically nil and the intergranular cracking was less than is found in many good heatresisting alloys after rupture testing. 3. The tensile characteristics suggest an aing pnhenomena taking place under the influence of stress at 13500 and 15000 V. If such changes do not occur then a considerable decrease rather than constant tensile and yield strengths should have resulted. More tests, however, would have been desirable in order to be sure that this apparent effect was not Just a happanstance from specimens with variable properties. 4. The alloy was very resistant to oxidation by air up to 1500~ F. The abnormally rapid oxidation of one rupture test specimen at 1500~ F and one at 18000 F suggests that the oxidation resistance may be quite erratic due to extraneous factors. This is further substantiated by the difference between the periodically air cooled and constantly heated oxidation specimens at 1800~ F.

25.,. Tile forging tests were not conclusive. They do indicate, however, that if the alloy is to be forged it will be difficult because of the low temperature required and the temperature control necessary.

L' a 1;-T — 0 — r- -- r` ~1 TS' —f'^- -T _| _ |'................ T ilr r i I I - t 1 t - 1X'- -1 i 1< - -;. I: 0 I i -i 04: f X IX X I l. _: t It I 1 1 t w-o4- -l — - — f- t - z t 1 t _ SS 1r:1 K —I I;] iT t I I ( t 1-, 1 — 74 iii I - l I 1- 11 0 -t 1 Q! 1 1- 1 1 |-i{0; F- 1 t-o-loof X'l'~~'~ ~X ~ X ~~' *-*1' s w1: s'10 X;-1 1-+4'-tir Nt - -- 4 >0-0 e-i tt - - + 1 ft —4- -t ~~" t; -- -- W -r —-1'-L L O I i__-.. 1% $-t: —-p -c —*rn-tlt —' —- 1 —-t-` 1-1 —- t~ _-0 _ _ 1-'t-~ 11 7 — <t- — 1-:rI 4- 2-S w P - - -- - Itoi__:~~~~~~~~~~~Tt 11: I t - - - - ---— s —-;;t bn1k l -l 77~~~~~12 ~41 -I1 8, 1'- -''' A -'' --' t- 1v1 ~;:' v -—' - - N - - - i -' -'- H fV0 F et: A~~ __ [~~ ~ S _0. n5 2Y| 711 0 tt:t1 _ 14W~ 1 ~T45 tu t 1 t1i 2 _ I; j 1 1- X |' t X on atiI 1 L.~~~~~~~~~~~~~~~. I~~~~~~~-. 1 $:t: t —f-T- 1 tt t t 1 - I I- 4.; t W 11 tj~:- r -i-i1 —--- -t~ -— I - 1j —-t4' -1 —c —-!i —-----— t —-- --- -: tiT ——'- -- -= — h: -: -~ 1 0',0- ~ - - --: - t 1; & i XXtf1,0|.j,!0| k --- -T —- l -l -- t r 1 - -- -— t - - -l — k- l r i: 1lt-+ i- 11 Fl -1 -t

7 —-- J:J I- II____ Ti f - __ —-- -- 3~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ i~1 1- 4 536 i- -*1 3 4 I~ ~ ~ ~ ~ ~ ~~~~i —t ______~ —t ---- _________ ______ —-j —-r-; -r — t — i- 4 4 + --— ~~ —- ______f I 0 c-stin Ni4 - 4. ~~ 477i — i -f - } —- -4 _c ____ _____-I- ---- ~1i -- f — - -i i 4_ ____ ____ -1Zi -- 4 t I'po __ ____ __ I~~~~ 74. f-I —- - f ~~~~~ —-~"t~z;- C~z_~ ___ iuf'.P'B" 0 ">P~ISCO ~TINJUED _i I _ I- ____ ___ ___ _____-. I~ ~~ ~ ~ ~ ~ ~ ~ ij 0) Ii~ L ____ __ ___:.fj.t:1. 8R (07~~'t — - L.- t-r__ Lii_ - - - ___ - 4I E15i__ __ __ I ~stin~ ~H1mb r. 5. __ _ j'~'I t;.a~~sr~~ln U rFIGU STR FVO t+ —TO Ari~ 1 789 3 4 - 6-7 -92 3 4 5 6 78 91

t — v-t t t 1 1 1 ---— i!-.!V -. i ~_ ~..; t t —t-e- --— t —-& —— S-t — w --— i-i 1 te;-,, PI 141 1 I- I 77 111c IL b t -, b I"tI i-t -! —-~-4.....__ _ i 1...ii.......'...L...1 —r' 4 TI To -, | -f- t~~~~~~~'i I t,::,:: l:-t,! L:_! t J_ * - - _-_-_' - --- |. -+i i- } 1: 14 T t I L~. + X ~,~ ~~~ -t~~ -j~4 I 1t,; - r- - r -i' -. t4~-' }.' t' 1:. X 1 1' -: -l:,- I t: l/:,0-/, i4:t 1 - r |: __ T-' 1 - V:I,,1..,'ri.-.1.. e.,::.r1 ij' S,.I!- l:-'t; l:t;-:: l fl; l;- t;l N;t:t t l; X; --;-;! 1t t j t 01 l-;i;-ti-l,,'l —— l-F-i t.-1N:1 I''' 1 I ll f1 1 —- — 0tr — l- -7-T-0TE

-Jk< +_ _X______- r - 1 —'0il1- __ -- 71 i- ____:- __ 8! l~_ _ _ _ __ _ _ _ H _ _ _ _2 ~z I _ _ _ _ _ 11 - -ILt2,.,t___<1z — 4 +1zzi2I]IKt -___ } 4- +- ____ ___ ___l 1 + - 8s i_ __Pt6 |__ S —$0-Cp. -T DR _'_'t_5Tpl|tFt 5;|:|-| ir-~~~ - _ _ _ _ _ _ H -; | - - _ _+t+=-i; -;1-' 4 — ~ 1 1 W — t 0 0~~~~~~~T[ W S1 F.- 1-.o "-: ~~ ~ _ I - ____ s|SS 11 tP-___17777. I -— ~ —:tI 11 1I'I1'1'1' —t1-H- -: - I;:-~ 0 114-I -.-1 _____ tE<~____ ____ 0F4*' 1 1 1 -;- -tfr~ ____ -~-t-4 -t t t- -* --— vt~ I < — j —-l - -— l —---— l —--— t —----— l —---------— ________ —-~ —'4i —0 —-— C-rf- - r I1_ -- _ -. - _i _|| I ____'_ ____ -t -- 1 - v;|!I1 1 -,0c q __ _ __ 1-: - |+ -— i —. -7j77 r.t41 0 ___-ft —.* -v+0 L11 2 1 j l 1 | S 1 [: | l l:1 l l! 1 1 2 1 I! 6 I odo.;i 1- — 3-4 — 7 (7 7I jI8 i.cOl 67830-~~ 3 4 567.o~

r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ TAPER PER FT. INCLUDED ANGLE MUST FIT GAUGE 7. -Y~~~~~~~~~~~~~~~ 010"_ o ~~~~~~~~~~~~~~~~~~_ _ __.__~__ _l/ l 1 23 / ~~~~~~~~~TAP, 2 U.S.S.I _____ _ 13 _,, 7 — >,~~~~~~~~~~~~~~~~~~ DRILL =g(I6" i-6, ~~3 ~~~~~2,/ ~DEPTH -2 HOLES 4 ~~~~~~~~~~~~~~~~~~~~~C',SINK, 60'X1 23 TAP -24. S$ < I 23 3 2 DEPATH TEST SPEC ___'I ESSI18 FIGURE 8.'FATIGUE TEST SPECIMEN

k N__ L X a- IDS,;I-iI --- SS,20 X 0 l aft~~ II t TI.Xi I i 10 AmA0X Xo YC X I *TX1T m an I.o, P-t F),9Fk~i E$T S-N URV R__'TT. P y ___

C4,0 43 Fo. rwipp,, Temr. OF.- Ori inal 1980 0 9:0 2240 No, of'Blows: Sample 49 21 1 8 (A)- Sanples from tests atteraDtinp to forge 2-inch cylinders down to 1-in6h in one hp-at. Original J. F, i n g T,!n —,.i n2 1 5 0F Forring Temp, 19500 F (4 rehearings) (B) O-RM-01'4- test attemptIng to forve bars from 0'4,-inch cylinders FIGURE 109- FORGING TEIc..OT SAMPLES OF FORD ALLOY*

~ $sd OO0~I jo ssea~9 ~ agpufl sanoH O{It ~[oj,4 oO0~I ~ ~ut~so$ do~aO,[o~,JV 8~~.Ieqmnl~ ~ul~s~O jo san~on;t~so.~oI~-'UI oan~ta OOOIX.... Lho~,, OOIX ~,~o~/, q. ~ "' ~' "" "'"" ~'' ~. ~' st..'.... /,-',. ~..-,.' ~-J''*" l' r ~.. -,.,,'"' ""'''' " ~''' -:''~''r"'~'~ ~,' / ~'.~..,,,~-..,.....,...,..,,.....:..... i,.- ~..i'..'..~ ~...''.'.:.~'~... ~..,. ~ ~.'. -.,.~, -...... i':':.... (,....... -:..:.' _ r',..:. -{,.~,y /. ~,.~:.,{...-...,. ~ -.,'.-.. -..,. ~ ~(..;..,[ ~...., ~..:...,. -.. ~ i >,, ~ ~..-. ~.., %"-: ~'"~5, ~,':..'......'"',';'.', ~ ~?, "' ~ i~v"' 0/,..:......,......,..,'. ~-,:''~.;:~ i..~:,:~::;.':'-"' ~:.':~, 0:,.i:.:%''~'''':?" ~j..,..- -..-... -..,.:.[ "' e,.,'.'-....~.-,~, -...-.'-...,:, —...:.,....;J ~. ~':~:''. t..,.-'." ~'.'..'"' ~ ~ -';' ~'' ~'...!?: ~-i.:: b.:' —'-.: ~.,'.':'~'.".', > _ > c: "~r'";''. -'; 5'::'"':'"..~ 5',:'- ~: ":'":':!:i::i:' ~.~ ~''''/ ~''.' ).,v>' - % ~ ~:!?i ".:':-',: ~-.:i":-, " -'':~,, %,......'' " ~......, ~'"'~ -":'";' "':" "'.'.....'' " " 5',;,~:-......,'::,.... -, >',.' - 0.-;..,:f..,.~,:.:,, ~,,,...:.j,;,:. <.,. ~..,. i' ~.. ~.,,',5.'{(.i ~i'~:'7~ 5': -'.':..'..<:,.: )':;?"::;;;.,:"'!'5:'.,'. —',,.,~ ~.,.,:,~."> ".C___'~.~,.,.L..':-.,,,'x5'~'.''.."..,.:' ~ ~: -' ":'"'"":' ":"'-.~_ ~.'"',~;t~.:',,......'" ~' ~'''~ \'' "'~'':"i' ".'/'..'"'.7-<..''-', —b" ".-...,.....: ~,,',! ~: "~7~.~~.....,......:...,........ ~;:_";.;/..., -~" M.:.:'% "."..'. ~......'' ~.::.-..?',,"//~, %".: ~'gg".''' "'~" "' J t i',-...'e' i' ~,'..,.'"'': "'"':' -—.,...... -' ~,....~ ~ - ~'-.",\./..~..-.'(,. -..,..,.. )>,~,:.'~..-:.:,.:..,.',[,i, ":,,T.:-:.' x\ ~. -,,,, ~ ~ ~.-': 3';''~,"~.":~'~'Lt,~ ".,.' -.::.'?-'.:,'..", * "' ~'...~.v';...'-:....., r,... ~ ~'-""'5.' -.: -'i":'' i i,::~ )' ~ "t..,,.,:..toqa $o oan ona soaot teul IaO -'It oan Ix 000IX OOIX ~ohl c~o~t:';C',*.-,' r.,./'A:,,,'~,~':';;'~ ~ ka~.:-~,.~.. -.'", /: -~ t~ ~.. -~,".'iiv-~ j, bL.~'%,?'-"' ~: ii~..,~"g)"~.:~.'-~'.'~.~-;-.?/'.~3": "::6~'"~;'~.~,..~tSrl'Jl~,d~'"mk./f -'~., J.-~"'";'"'~'"-"'Y'' "';'.... 5:." ~'.~ i,,-,'.. ~ ~.'~.',:~.-~' ".-,."'~' ~ t:.~,,..''.'?:.;.' I..':....-,.,,,.,e.~i~.....,,~.,...1,''~l e.~ ~ -.:...'~-%v,%~r.::,,...,- ~..~,) ~... ~. ~. ~:,,,,.~., ~:......',., - t.:' ~,,:~," ~,::,~".':. j,~:,',,.i..'q~',~:.':;.:.*:..~,,~.*',.~: ~( ~.'''' ~' ~ ~ ""- ~...'. "~...,.-~_%5"'"TM~ ~'......"'~""' — "''' ~', "'':.!'~1 "-'..:'_."//:-".::;',( \?'~'~'%. ~- - ~:' "i':'.~"::-'~;~",'~"' "~'::.:''-~''~"'.e:.",:'?..'::;,,.'...... ~:'5 "~:~'.,' ~:~.',-.':' )::,. ~...',,..'-_"' (.,..,~,~,~:..:....:. a:..::.:,..(~;~ ~......: ~ ~.:.' q'' —....." —.,/....',...,-__.,,~_~.~..;,., x~ ~:-:~,......;~......,.. ~,, ~. ~,.....,-: %:.,.,........., ~,.-.,. ~-.,,.'.:.-~,.....' ~.~.'., ~.,..;,,.',.~.... _.,,:.. _...:.: ~. ~,= ~,.~'.~,,.~-'~''~':"' "i' " ~'" ":',"'.%~ —~.;:(~,~,':73,:.,.:.-'..~:'' --,....,~".5 I~"'.2 "~.:;.~?.'.'',':'.~,,,,:,'"'~ ~'.:.'- It~-lr~,:. >'-..... ~ -" ~' -(': ~',~,~-?,. - ~.' ~'./,, II~r.,~'..;:;':;.'."'''..- (,'"' lr.'~ "'~'-~' -,'-', — JJ~'x7.: -~"...-. ~' ~. -'* "~ "-. AV..'~';:~:".'""' ~,..... ~.,;.,: /n....:,-'.el?'f::-';:;,:.'.:". ~ ~;.':.:'~ / ~'-..,.,.~.... "V ~. ~... ~....'L.~ -,.;';."%."'...'.~:tJ:':..~-''.':"L;.-'."'-::'''.-, —, ~': -. ~,;~,.?'.~'~'_,'', ~...~.. ~ ~...-:. ~.:.:. _N.;~.~,,~ ~. e -...:.....~ ~,'-~ o'' "' ~ ~,-5~,'~.%,~:~'".......... ~;: ~ ~.... -.~,,.... ~:~,...............-,..,~...~.,~a~ ~:. ~.:..,, v..... e' ~," ~ " / ~' ~ -.'....i~. ~ ~'.t15~,.~.' ~'.:~.. > "'.,'-;; —'.:.'~,". —;~:,~.-...''i, ~'~'...., ~ % " ~.~'~,"/,', 3,,:~': h~..'."..'. ~ ~'. ~ ~.,'_ c ~' ~' / ~ * ~:'.,,~'~ ~.."~ a"'"'..~ ~...',....'', ~' ~'..',......,~"' ~.,.. >-~...-,:-T~"-'.'.~,7...........,,:,~$ ~ ~ ~,.~_

I'R ~' 1,' v:, K~ ~~~~~~~~~~~~~, ~.''1 A~~ /1 ~.. Fracture- XlO0 PI~~/ Rupture -JQ X10 0SC, Figure 13.- Microstructure of Casting Number 358 after Rupture in 508.5 Hours at 15000 F Under a Stress of 21,000 psi.,~~-' o' -i t,~.&.',' -- r.ar -~., v>.;~ j. K',d J~,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ *.~~ q A'' " $ u 4~~6 or.at- 160 F Une teso 350pi''r~~~~ ~~~ c B ~1 ~-,~~~~~~~~~~~~~~~~~~~~e~~~~~~~~~~~. d~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~b ~~~~~~~~~~~~'.,... ~~ "''' p;)~ 9 b~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...: ~.',, h. ~ ~,,.,. i ~ ~Fatr,' ~s-, r - XlO0r Fract~~~~~~urfae - ]C0 Y ~ utr,Xl0 /"S Figure 14.- Utcrostructure of Casting Number 358 After Rupture in 5085 Hur at1500 Unera Stress of 213,00 psi. 86& Hours a~ ~ ~~~~t 1600~ F Unde

1'~~~~~~~~~~~~~~~ri. /.,.9.~~~~~~~~~~~~~~~~~~~~~~'1 ~i...... ~~-,,.~..-/,.:.: ~' ~"- ~ — ~d"',, -,. - a 0r i.'"..~, i ~ ~. c. Fr'acture X0- Srface X i00' Fi~re 5.-Micostuctre ffCasting Number 358 after Rupture in5.-4 Hoursstact18000FUndere 5000opsi. 9 ~ ~ i 58 Hour at100 Udr 50 p

UNIVERSITY OF MICHIGAN 3 9015 02826 5281