73“: \; ll n... :n l~4 C) "C 3) II) C) T] (’1 a“ (3 O U TD H' CH 3} (Tr—VP- . l...) ” Alloy Steel and Articles Made Therefrom ” Thesis for Degree of Metallurgical Engineer RICHARD ELWOOD BISSELL ’13. 1927 THESIS ‘7? /<,. -_. A.) .. . . . 44 -_ .. .. w L" . . R Iii-DEA II‘ V .. n x .\ hr... z x y u 5,: w n p x ., fur!“ "ALLOY STEEL AND ARTICLES MADE THEREFRDM” Thesis Presented to the Michigan State College In Partial Fulfillment of Requirements For the Degree of Metallurgical Engineer " I '\ .'~. \ . _ q. a (- \ ‘\k:-\ t: I By Richard Elwood gull 19270 THE I" '3 "lLLOI STEEL.AND ARTICLES MADE THEREFROM” The leed of no: material of Different Than Normal Characteristic fer Aircooled Aircraft Engine Poppet Valves. 1 Search of the Automotive Peppet Valve Material Field. Test of Materials or Compositions Intermediate in Composition to Previous Known Compositions but with the Addition of Silicone Discovery of Something Valuable in the New Chrome Nickel Silicon Steel. Characteristics of the new'laterial. "ALLOY STEELS AID ARTICLES MADE THEREFROM" Demands for greater efficiencies, tOgether with less weight, in automotive engines and in particular aircooled aircraft engines, have caused engineers and metallurgists to carefully check over the require- ments of each individual part. The materials available and the treat- ments which make them the more suitable to fill the requirements as listed, are carefully studied. In the case of the exhaust valve for the aircooled aircraft engine, the requirements are many and the materials heretofore available for other valve service in any condition of treatment, do not thoroughly satisfy the designing engineer or metallurgist. Dr. 1. 2. 3. 4. 5. 6. 7. 8. The 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.- 14. Aitchison‘ lists the types of valve failures as follows: Elongation of the stem of the valve. Distortion of the head of the valve. The presence of cracks in the valve face. Excessive wear of the valve stem. Excessive wear of the valve foot. Burning out of the head. Scaling of the valve. Breaking of the head or neck due to self hardening. writer has added, modified and extended this list as follows: Tensile stretching of the valve. Warping or curling of the valve. The presence of cracks due to defects in the original steel - due to expansion or contraction stress fatigue or - due to direct service load stress fatigue. Excessive wear of valve stem. Excessive wear or riveting of the valve foot. Oxidation, scaling or burning due to a combination of temperature and mixture, including added or deposited chemical compounds, from doped fuels. Scaling of the valve. Breaking of the head or neck due to self hardening. Grooving, battering or excess wear of the valve seat. Peaning, riveting, or burring of spring retainer supporting surface edge. Pitting, corrosion or erosion of valve head. Pitting or corrosion of valve stem. Breakage due to brittleness. Foreign matter adhesions to original valve surface, commonly called "pick up". ‘ Paper read before the Institute of Automobile Engineers, Nov.5, 1909. Dr. Aitchison has described the preperties a valve steel should possess: l. The greatest possible strength at high temperatures. 2. The highest possible notched bar value. 3. The capacity of being forged easily. 4. The capacity of being manufactured free from cracks, whether these arise in the manufacture of the steel bar or are produced during the forging of the steel. 5. The capacity of being heat-treated easily, regularly, and reliably. 6. The least possible tendency to scale, and if scaling does occur, the scale should be as adherent as possible. 7. The ability to retain its original physical prOperties after frequent heating to high temperatures, followed by cooling to normal temperature, also after being heated to an elevated temperature for a considerable length of time. 8. No liability to harden when cooled in air from the temperature which it will attain when used normally as a valve in an engine. 9. The capacity of being heat-treated after forging so that it is free from strains liable to produce dis- tortions. 10. Sufficient hardness to withstand excessive wear in the stem. 11. The capacity of being hardened at the foot of the stem with considerable ease if necessary. 12. The capacity of being machined easily and satisfactor- ily by ordinary methods. Although this article was prepared about eight years ago, it can be seen that at that time the steel was supposed to possess much. The eight years have not decreased the list, and we are now quite content to eliminate "easy" “normal” Operations in the attempt to create a satisfactory valve. The writer has added: 13. The capacity of Operating at high temperatures without adhering to foreign matter. On the first examination of the list it will be noted that the steel with the greatest possible strength at high temperatures (1) will not have the capacity of being forged easily (3) in strict sense of the word, and also it will be noted that with all the preperties from #1 to #11 perhaps even grinding must be resorted to in place of #12. 3. Such materials as are available for solid one piece exhaust valves up to date might be listed as follows: Straight Chrome Steels Type A. Type B. Chrome Cobalt Steels Type A. Type B. Chrome Silicon Steels Type A. Type B. Type C. Type D. Type E. Chrome Nickel Steels Type I. C .45 .55 .20 .40 .35 .50 .45 .60 .20 .30 .35 .45 C .40 .50 Cr .50 .80 Co (0 .60 .80 1.15 1.85 Ni .50 flax .45 .75 Mo .60 .90 Judging from the writer‘s experience with these different grades of material, together with careful interpretation of exten- sive laboratory tests and scientific data, a chart is compiled indicating the relative merit of each material for each of the thirteen prOperties. The steel possessing the desired property to the greatest degree is marked 1, and the steel possessing it to the least degree is marked 5’. 1 2 3 4 5 6 7 8 9 10 11 12 13 Chrome Type A 4 1 4 5 s s 4 4 2 1 1 2 2 Chrome Type B 5 1 3 4 s 2 s s 2 2 2 2 3 Cobalt Chrome Type A 2 5 4 4 4 2 1 1 1 1 1 4 4 Cobalt Chrome Type B 3 4 4 3 4 2 1 1 1 1 1 4 4 Chrome Silicon Type A 4 2 l 1 1 l 1 l l l 1 4 5 Chrome Silicon Type B 3 3 2 2 2 2 1 1 l 1 1 4 3 Chrome Silicon Type C 2 2 2 3 1 3 l 1 1 1 1 2 2 Chrome Silicon Type D 3 2 2 2 3 3 2 2 l 1 2 3 4 Chrome Silicon Type E 4 4 2 3 3 3 3 3 2 1 2 3 4 Chrome Nickel Type A 1 2 2 l 4 1 2 1 l 5 5 4 2 Returning to the writer's list of fourteen types of valve failure and bearing in mind the conditions of Operation Of an air- cooled aircraft engine exhaust valve, it is seen thqt failures of type 3, 4, 6, 8, ll, 13 and 14 should be particularly guarded against. This seems to require a merit 1 of steel properties 1, 5, 6 and 10 in particular, and if there be any merits 5 they should be in prOperties pertaining to ease of manufacture. Valuable as they are - - the steels listed do not in any instance fulfill exactly the requirement and the indications are that a new material must therefore be sought. Chrome Nickel Steel A leads the column in property 1 with merit 1. Although this material has great strength at heat, it is always relatively soft cold being austentic. It therefore does (‘Similar to, but not a copy of, Aitchison's Table XVII.) 5. not resist wear On stem or foot and cyaniding Or carbonizing can not be resorted to, to improve the 91 nation. It appears to the writer that nickel and chromium in com- bination are necessary for the high strength at heat required and that when sufficient nickel is added for suitable effect here - it is necessary to improve the Oxidation resistance and the ability to forge by the addition of some such element as silicon. In order to harden the material, it will perhaps also be necessary to have the elements in such proportions as will allow the production Of martensite for resistance to stem and tip wear at least. For this purpose the writer prOposes to study then the alloys between the straight chrome Type B and the Chromium Nickel Type A, but with added silicon which would appear necessary for the pur- poses above mentioned. The test alloys are as follows: C Mn P 8 31 Cr Ni TEST ALLOY l .30 .21 .01 .025 2.00 13.5 3.9 ” " 2 .35 .25 .02 .02 2.05 13.4 5.6 " ” 3 .33 .22 .02 .015 1.90 11.9 8.2 " ” 4 .26 .21 .02 .025 2.60 12.0 10.9 TEST ALLOY #1 Characteristics: Br Sc Hardness as received ———==' — 340 --- " after forging and air cool - 320 55 " 16500 F 2 hr cool 50 °/Hr to 1350° F Packed in Boxes with Mica 436 64-65 " ” 1550 Q.O. Following above 400 60-61 " " 1600 0.0. ” ” -—- --- " " 1650 Q.0. " ” 436 65 " " 1700 Q.0. ” ” 436 65 " ” 1750 0.0. ” " 476 70 ” " 1800 Q.O. " " 476 70 " " 1850 0.0. " " 456 65-70 " " 1900 0.0. " " 400 65 " " 1950 Q.0. " " --- --- " " 2000 0.0. " " -- ~---- " " 2200 Q.W. " " 228 46-47 Spec. Cr. 7.78 TEST ALLOY # 2 Characteristics: Hardness as received ---- — H W Hardness as received — after forgéng and air F4Hre 50 after forging and air cool 1 s a s 3 3 3 1e50° F 2 Hrs cool 50 DF/Hr to 1550°F in Boxes 4 Mical Packed 1550 1600 1650 1700 1750 1800 1850 1900 1950 2000 2200 Q. Q. 000090008 0. 0. 0. 0. O. O. O. 0. O. 0. W. Following above H N H N 7' N N 9' 333333 3 3 Spec. Gr. TEST ALLOY #3 Characteristics. 33333333333 1650 1550 1600 1650 1700 1750 1800 1850 1900 1950 2000 2200 Q.0. 0.0. Q00. 0.0. 0.0. Q.0. Q00. Q.0. Q.0. Q.W. ° F/Hr to 1350° cool = 7 .81. Boxes 4 Mica Q.0.Following above Q N N 9' H H '9 fl " H O . 33333333 Spec. Gr. TEST ALLOY #4 Characteristics: 7.84 Hardness as received ----——- ——- after forgingz and air cooled 1-1 16500 F2 1550 Q. 0. Following’above 1600 0.0. Following above 1650 0.0. N '9 H n H fl 9' N N N N N N 1700 Q00. 1750 Q.0. 1800 Q.0. 1850 0.0. Hr 5o °F/Hr to 1350°Boxes & Mica Q 7' fl " " I. 9' N Br. 241 200 400 400 400 400 360 380 436 400 315 280 205 186 210 180 370 370 370 400 355 296 243 233 201 190 160 132 163 148 190 158 158 158 154 166 143 138 Sc. 44 43 55 55 55 55 55 55 58 58 42 37 36 40 37 58 61 58 58 45 40 34 34 28 33 32 39 38 37 36 37 39 35 35 7. TEST ALLOY #4 Characteristics: (Continued) Br. Sc. Hardness after 1900 Q.O. Following above 132 33 " " 1950 Q.0. " 128 32 ” " 2000 Q.0. " " 126 31 " " 2200 Q.W. " ” 114 31 Spec. Cr. 7.84 It is noted that Test Alloy #3 is apparently an a stentic steel when forged and the special 1650O - 2 hr. cool 50 /Hr. to 13500 (Packed in boxes with mica) treatment allows it to at least partly revert to martensite which is again changed to and retained as austenite starting on retreatment at 1700 to 17500 F followed by a quench. It is noted that after a 2200°F heat followed by water quench, the hardness is particularly low. In this condition it is readily turned and milled but drilled or sawn with difficulty. Since air cobled aircraft motor valves rarely exceed a 1520°F temperature, and since this alloy seems to possess apparent desirable features in that it requires 1700 to 17500 heat following a hardening by slow cooling before it is materially changed, it is desired to study its characteristics further. With this idea in mind, these test results are presented. (1) Forge 92) Reheat to 16500 F - 2 hrs. cool 50°/Hr. to 1350°F. (Packed in boxes with mica.) (3) Reheat to 2200° F, quench in water. (4) Reheat to temperatures shown below, followed by test. BRINELL SCLEROSCOPE BEND‘ FRACTURE 400° 2 Cool in Air 130 so a bends Fine 7 Reversals Jagged 800° F Quench in Oil 130 so 5 bends Silky o 4 Reversals 1000 r Quench in 011 124 so 4 Bends Silky 1200°F » w n 134 so 4 v ." 1250°r n w w 132 so 4 " 5" 1300°F Cool in.Air 136 so 3 " " 1350°r n w w 140 31 1 " " 1450°r « w n 128 so 5 " " 1550°r w n n 135 31 1 " " 1650°F n n n 154 34 '1 " " 1750°F n n n 152 34 l ” " 1850OF " " " 160 34 l ” Coarsening up 1950°F " ” " 152 so ? Fine ’ (Piece held in vise and bent back and forth with hammer blows.) 8. It is assumed by the writer that the number of reversals of an austenetic steel before fracture are a measure of the proximity Of stable martensite. At 1850°F some tendency appeared to be hardening the material and changing the fracture slightly. It now seems desirable to put the material through a variety of treatments to determine how it changes from one condition to another, as, from the two tables presented, it is clear that after the same temperature and type of quench, it can be in different conditions, depending upon the preceding treatments it has received. GROUP TESTS A TEST ALLOY #3 ANALYSIS OF STEEL USED 0 M5 Si N1 Cr. .38 .33 2.16 7.33 11.50 BRINELL SCLEROSCOPE ROCKWELL Hardness of stock as received 241 1. After 2200°F Quench in Water 169 29 11 2. As above, followed by 1650- 4 hr 50 /Hr. 1350 421 61 47 3. As above 1 & 2, followed by 1250°F 8 min. in MOlten Salt Bath, 0.0. 364 56 42 4. As abov , 1, 2 & 3, followed by 1450 F 8 min. in Molten . Salt Bath, 0.0. 395 60 46 5. As above, 1,2,3 4 4, followed by 1200 F, 8 Min. in Molten Salt Bath, Q.O. 358 55 40 GROUP TESTS B TEST ALLOY #3 ANALYSIS OF STEEL USED (Same as A) Preliminary Treatment: (1) Heat to 2ZOOOF Quench in Water, 0 o (2) Reheat to 1650 r - 4 Hrs, Cool 50 /Hr to 1350 r. (3) Reheat to temperature shown below, quench and test. BRINELL SCLEROSCOPE ROCKWELL. Reheat in Molten Salt 8 min.at 900 F QC 427 62 47 " ” ” ' 8 ” " 950 F ” 410 61 47 ” " ” " 8 " ” 1000 F " 381 59 44 ” " " ” 8 " " 1050 F " 369 56 4s 9. BRIhELL SCLEROSCOPE ROCKWELL Reheat in hiolten Salt 8 min at 1100 F 00 364 55 43 " ” " " " ” ”1150 " 366 55 44 " ” " ” ” ' " 1200 " 354 55 42 " ” " ” " " ” 1250 " 364 55 42 " ” " " ” " ” 1300 " 372 57 43 " ” " ” " ” ' 1350 " 398 61 45 H N N N 9' N N 1400 '9 391 61 46 ” ” " " ” " " 1450 " 396 60 46 Reheat in Molten Lead 8 min at 1500 ” 402 60 47 ” ” " " " " " 1550 " 410 61 46 " " " ” " ” ” 1600 " 375 57 43 fl " N N N N N 1650 H 343 54 38 " " ” " " " " 1700 " 262 42 29 GROUP TESTS C TEST ALLOY #3 ANALYSIS OF STEEL USED (Same as A) Preliminary Treatment: (1) Heat to 22000 P, Q. 3. 155 28 11 (2) Reheat in Molten Lead for 8 Min at 1650 F Q.0. 185 30 16 Followed by: 1650° F, 4 hrs Cool 50° F.Hr to 1350°F .387 59 43 Then 1650° F, 8/m in Lead, Quench in 011 302 47 35 " 1280 F, 8/m in Salt, " " 348 54 42 " 1650 r, 8/m in Lead, " v " 331 49 39 " 1450 F, Ofm in Salt, " " " 384 59 46 GROUP TESTS D TEST ALLOY #3 Pieces treated consecutively (Last piece receiving all previous heats) BRINELL SCLEROSCOPE ROCKWELL c (1) 2200°r, quench in water 140 31 10.25 (2% 1700 F, v 011 - 172 32 15.0 (3 1450 F, " v 011 172 32 16.0 (4) 1450 r, " " " 182 36 ---- (5) 1450 F, " " " 189 38 --- (6) 1450 F, " " " 210 42 --. (7) 1450 r, v " v 205 42 -- (8) 1450 F, " " N 205 43 --- (9) 1450 F, " " " 238 44 ---- (10) 1450 F, " " v 260 47 --- (11) 1450 F, " " " 279 56 -- (12) 1450 F, " v " 316 57 -.-- (13) 1450 F, v " " 328 57 --- 14) 1450 F, " " " 356 58 -- 15) 1450 F, " " " 372 58 -- 10. From the characteristics as determined above, the writer concludes the method of manufacture of a valve Of this material should be as follows: (a) Forge (b) Heats to 21000 F, Quench in Water (6) Machine (excgpt drilling, sawing or threading) (d) Heat to 1600 F, 6 hours, Cool 50°F/Hr to 1350, then cool in air. (e) Heat to 1250° F, Cool in Air (f) Drill, saw, or thread. (g) Heat to 14500 F, Quench in Oil. The valve will then be 55 minimum SclerOsc0pe or martensitic head, stem and tip. Another method perhaps equally as good, would be as above for a, b, c, d, e, f and g, followed by heating of head only to tap of guide section to 2100 F, and cool in air. This would give an austenitic head on a martensitic stem. TEKSILE TESTS HEAT TREA MEET (1) Heat to 2200°P Quench in Water. (2) Reheat to 165065, 8 Hrs. Cool 50°F/Hr to 1350°F, (Packed in boxes With M1030 Pulled at room temperature: Elastic Limit 203,750 to 207,500 #/sq." Tensile Strength 212,500 to 217,500 i/Sq." Elongation in 2" 1 . Reduction ofNArea 31% Pulled at 1500° F: Elastic Limit 29,300 to 30,000 #/Sq." Tensile Strength 34,525 #/sq.' Elongation in 2" 12.0 to 14.5% Reduction of Area 12.0 to 14.8% OXIDATI 0N RES IS TANCE TESTS : Pieces heated at 16500 F for 8 hours in a relatively tight furnace with pure oxygen inserted continuously from an oxygen cylinp der, and with oxygen coming from all minute cracks and crevices, indicate almost entire freedom from scaling. This is considered a very good test. COEFFICIENTS 0F EXPATSION 32°F - 302° F .00001048" ° F 302 - 392 .00001013 392 - 482 .00001019 482 - 572 .00001036 572 - 662 .00001048 662 - 752 .00001053 752 - 842 .00001070 842 - 932 .00001079 932 - 1022 .00001084 1022 - 1112 .00001088 TRANSFORMATION POINTS Pieces submitted to The Leeds & Northrup Company, 4901 Stanton Avenue, Philadelphia, in the austenitic condition. Results as follows (Letter Of October 20, 1925):‘ "These curves were made with the regular transformation point indicator apparatus but we could not discover any critical point. In addition to this, the laboratory made tests taking time temperature readings with a set up much more sensitive than the transformation.point apparatus. These time temperature readings have also been taken with" - - - - - etc. "One of these runs have been plotted so far; blue prints of which are enclosed. These prints have to have the corners matched so that the curve extends across both sheets. No signs of a critical point appear." "You will note that the thermocouple readings are plotted in.mocr0v01ts. The thermocouple used is one of our standard Rhodium thermocouples which matches up with Table No. 12-A."* "The variation in heating curves were caused by changing the current through the furnace.” ‘* Curves and tables are found in leather pocket inside Of back cover. 11. 12. MAGNETIC TL STS: Below is quotation from letter of Dr. John A. Mathews, Vice President of the Crucible Steel Company Of America, under date Of August 2nd, 1926: "Some tests that we made of material taken from one Of your shipments Showed that after the high quench, the material was fully austenitic and quite soft. The mater— ial did not harden until a drawing temperature of 1400 was reached, when it showed a Brinell of 321 and was becoming magnetic. After a 1500 tempering, the Brinell was 375 3nd the permeability was considerably increased. At a 1600 draw the Brinell was 444 and still higher in induction, indicating the conversion of gamma to alpha iron. The drawing times previously mentioned were one-half hour. On holding for six hours at 16000 the hardness remained the same, but the induction changed from 5500 to 9200. On again drawing atoa lower temperature, the material softened to 364 at a 1200 draw and with an induction of 10,500. 0n bringing up to 15000 again for one-half hour, it once more showed 444 but with an induction Of 8000. For all temper- ings and retemperings from 10000 up, the material showed a high coercive force of from 55 to 67, indicating magnetic hardness, with the Single exception Of the redraw at 1000 after fully hardening, when the coercive drOpped to 28." IMPACT TESTS: Treatment: (1) After 21000 F Q. OW. 116 Foot pounds (2) Followed by 16500 F 5 hrs. (Packed in.Boxes with Mica) Cooled 50°E/Hr. to 1350°r 3 Foot pounds (3) Followed by 1280°F Cool in Air 10.5 Foot pounds (4) " 1450°F Quench in Oil 7 5 Foot pounds. CONCLUSION. Let us now see where the new material should be classified on the chart indicating the relative 2merit Of each material for each of the thirteen prOperties (Page 2): PROPERTY 3 4 5 6 7 8 9 10 11 12 13 TEST ALLOY #5 2 l l l 2 l l 1 2 5__,2 P‘P‘ nan: This fulfills our original idea that fer aircooled aircraft engine valves the steel should.have merit l of steel prOperties 1, 5, 6 and 10 in particular, and if there be any merits 5 they should be in properties pertaining to ease of manufacture. Recent reports from three of the largest builders of aircooled aircraft engines - two in America and one in France - confirm the findings as presented. l. 2. 3. 4. B I B L I 0 G R A P H Y "Valve Failures and Valve Steels in Internal Combustion Engines", By Leslie Aitchison, Institute of.Automobile Engineers, Nov. 5, 1919. "Materials for the Exhaust Valves of Internal Combustion Engines", By J. E. Hurst and Harold Moore Engineering Vol. 108, pp. 672-74. "Characteristics of Material for Valves Operating at High Temperatures", By J. B. Johnson and S. A. Christiansen, American Society for Testing Materials, June 1924. "Valve Steels", By P. B. Henshaw, The Royal Aeronautical Society, 7 Albemerle Stre. W. 1., Dec. 2, 1926. r” 3") . 5380521 {3353. Cit-EN a l J a); ‘ I/Xy 71/5 99732?" / M. V. pen-0 Deg. F. Table No. lZ-A THE LEEDS & NORTHRUP COMPANY PLAT. VS. PLAT-l- 10% RHODIUM THERMOCOUPLE Degrees Fahrenheit Cold Junction 00F. 800 ° 100° 200° 300° 500° 600° 900° 1000° 1100° Millivolts 2.55 .0032 .0037 .0042 .0046 .0048 .005 0 .0052 .0053 .0054 .0055 .0056 .0057 1200° 1300° 1400° 1500° 1600o 1700° Millivolts 1800 ° 1900 ° 2000° 2100 ° 2200 ° 7.81 ' 8.43” 9.05 _.S:16 7.97 "8.58?” ,. 11.85 _> ___11.88 12.52‘_H_‘____13.18 .0061 .0082 .0066 .0066 .0007 .0006 2800‘7 29000 Millivolts .0060 2300" . July 10, 1923 Deg. F. M. V. per" Deg‘. F. M. V. per° Deg. F. M. V. pa" Mllllllllllllsllllrllll llllllllllllvlllllllllllllm 3 1293 03058 0595