REPORT ON RUPTURE PROPERTIES AT 1100~F OF SIMULATED TURBINE WHEEL FORGINGS FROM "17-22-A"V, "17-22-A"S, AND REX 448 STEELS By R. Jackowski A. I. Rush J. W. Freeman PROJECT 842 May 7 1954 REPORT 200 May 7, 1954 THE TIMKEN ROLLER BEARING COMPANY STEEL AND TUBE DIVISION CANTON, OHIO

INTR ODUC TION The investigation covered by this report was undertaken to determine the rupture properties of "17-22-A"V steel in the form approximating the forged discs of the type used as rotors in the gas turbines in the jet aircraft industry. In addition to the "17-22-Af"V discs, comparison discs were supplied from "17-22-A"S and Rex 448 (12 Cr Mo V Cb) steels. Rex 448 barstock was also included for comparative purposes. Previous work on barstock from small induction heats of 1t722-A."V and Rex 448 analysis indicated that the rupture strength and ductility of the " 17-22-A"V steel at 1100~F was substantially higher than that of the "17-22-A"S steel. However, both the rupture strength and ductility values for "17-22-A"V steel were lower than those of the Rex 448 analysis. This report presents the properties established for each of the forgings and in addition are compared with the properties previously established for "17-22-A"V and "17-22-A"S barstock.

2 SUMMARY AND CONCLUSIONS Appraisal of the data from this investigation led to the following conclusions: 1. The rupture strength and ductility of the disc material were equal to those obtained previously for barstock from "17-22-A"V steel. 2. The rupture strength and ductility at 1100~F obtained for the disc made from "17-22-A"V steel were substantially higher than those for the disc from "1.7-22-A'S. The " 17-22.-A"V material had 1000-hour rupture strengths of 30,000 to 36,500 at 1100~F as compared to the 22,000 to 24,000 psi for " 17-22-A."S. 3. The "17-22-A"V steel disc data indicated equal and possibly higher rupture strength than for the Rex 448 steel forging for time periods of up to about 300 hours. The steeper stress - rupture time curves for the "17-22-A"V material resulted in lower 1000-hour strengths. 4. The Rex 448 barstock oil quenched from 2115~F was approximately equal instrength and ductility at 1100~F to the Rex 448 disc, whereas that of the 1850~F normalized barstock was lower than either. 5. The ductility of the specimens from barstock and disc of "1722-A."V steel was in all cases substantially lower than that of the Rex 448 steel. 6. The 1000-hour rupture strength for the "17-22-A"V disc material varied from 30,000 psi to 32,500 psi for the tangential specimens and from 30, 000 psi to 36, 500 psi for the radial specimens. These variations in rupture strength obtained for the specimens from the "17-22-A"V disc are difficult to explain because it is uncertain whether the variations were

3 caused by differences in tempering conditions, location of the original specimens in the disc, or a combination of both. 7. The rupture strength of the disc from "17-22-A"S steel was higher than that of barstock from a commercial heat of similar analysis. 8. The discs from both steels -- "17-22-A."V and "17-22-A"S -- apparently had greater resistance to tempering than had been expected. Previous work done on these materials did not show the same resistance. TEST MATERIALS Machined 0. 505-inch diameter tensile specimens were supplied from "17-22-A"S, "17-22-A"V, and Rex 448 steels reported to have the following analysis: Heat Chemical Composition (percent) No, C Mn Si Cr Ni Mo V Cb Al N2 " 17-22-A"V Steel 02359 0.25 0.56 0.66 1.24 0.27 0.55 0.73 —. " 17-22-A"S Steel 02358 0.29 0.50 0.68 1.29 025 0.50 0.25 Rex 448 Steel 02336 0. 084 0.90 0. 10 10.64 0.28 0.80 0. 18 0.32 - (1) 0. 102 1.08 0.49 10.73 0.84 0.70 0.23 0.50 0.013 0. 045 (1) English made disc having in addition the following: Zr - 0. 16, A1203 - 0. 006, Cu - 0. 16, SiOz - 0. 006. The V/C ratio of Heat 02359 was somewhat lower than the approximate values of four which was originally considered desirable for "17-22-A"V steel. The composition of Heat 02358 was typical of "17-22-A"S steel.

4 The tensile specimens were prepared from discs and/or barstock heat treated as indicated in Table I. It will be noted that retempering was required in the case of the "17-22-Al" type steels to obtain the desired hardness level of about 300 BHN. Metallographic specimens were cut from the end of tensile specimens without knowledge as to the position of the specimen in the disc itself. Where the stress for rupture was 40, 000 psi or more, the diameter of the specimen was reduced to 0.400 inches in order to maintain load requirements within the capacity of the stress-rupture units. RESULTS AND DISCUSSION Tests were carried out to establish the rupture characteristics of both the tangential and the radial specimens from the forgings made of "17-22-A"V, "17-22-A"S, and Rex 448 steels. The results of the stressrupture tests at 1100~F are presented in Table II and Figures 1 through 4. In each case, comparative stress-rupture time curves from barstock were included. The following sections present and discuss the results of the data obtained. " 17-22-A"V Forgings As will be discussed later, there were complications in the data due to tempering difficulties. The data appear to indicate that the radial specimens had higher strength and elongation at 1100~F than the tangential specimens for time periods longer than about 350 hours. (See Figure 1.) The radial specimens were also somewhat higher than the original experimental heat

5 of barstock (Heat A80), Comparative rupture strengths for 200 and 1000 hours were as follows: Rupture Strength Elongation at Type Heat BHN (psi) Fracture (%) Specimen Treatment 200-hr 1000-hr 200-hr 1000-hr Heat 02359 Tang- N. 1800~F + 6 hrs 302/ -- (32,500) -- 6 ential 1225~F + 6 hrs 1200~F 323 + 6 hrs 1225~F Tang- N. 1800OF + 6 hrs 311/ 53,000 30,000 ^5 <3 ential 1225~F + 6 hrs 1200~F 321 + 6 hrs 1225~F Radial N. 1800~F + 6 hrs 323/ 45,500 36,500 8,5 44.5 1225~F + 6 hrs 1225~F 337 Radial N. 1800~F + 6 hrs 311/ -- (30,000) - 1225~F + 6 hrs 1200~F 321 + 6 hrs 1225~F Heat A80 Barstock N. 1800~F + 6 hrs 302/ (50,000) 34,500 -,7 <5 1200~F 311 Values for rupture in 200 hours were used, rather than the more common 100-hour values because it is doubtful that the data could be reliably extrapolated to 100 hours. It will also be noted that the number of tests was so limited as to constitute a survey, rather than to exactly establish values. As originally heat treated, the disc had a higher hardness than was desired, 352 - 375 BHN. Retempering was carried out under the several conditions listed in Table I to bring the hardness down closer to a 330 maximum. Examination of Figure 1 suggests the possibility that the retempering of finished specimens at Michigan led to higher strengths than these materials retempered by Timken,. It appears very possible that this was more

6 important thanthe difference between radial and tangential specimens. It appears that the forging structures were more resistant to tempering than barstock normalized from the same temperature. The microstructures of the "'17-22-A"V disc shown by Plates 1 through 3 indicate considerable variation in structure between specimens. The structures of the radial specimens varied from a banded type structure of ferrite and a finer grained tempered bainite, Plate lA,to a mainly coarser accicular, tempered bainite structure, Plates 2A and 3A. The tangential specimens, on the other hand, showed a more uniform structure of fine grained ferrite and bainite, Plates 1B and 2B. Plate 3B shows a number of larger ferrite patches. No significant difference in structure was observed between the specimens with the various retempering procedures. Variations between specimens, probably due to variation in location in the forging, obscured any conclusion regarding tempering conditions. Since the photomicrographs were made from metallographic samples taken from ends of tensile specimens, the photomicrographs therefore may not be representative of the structure in the gage section, particularly in view of the fact that the exact position of the samples in the disc is unknown. It is uncertain whether the differences observed between the radial and tangential specimens from the "17-22-A"V steel were entirely due to the location of the specimens, to the differences in microstructure, to the tempering conditions, or to a combination of any of these factors in the original disc.

7 "17-22-A."S Steel The rupture test data obtained at 1100~F are given in Table II and plotted in Figure 2 as stress - rupture time curves. The limited number of tests indicated a higher rupture strength and lower ductility than is charactereistic of barstock. Comparative rupture strengths and elongation values were as follows: Rupture Strength Elongation at Type Heat BHN (psi) Fracture (%) Specimen Treatment 200-hr 1000-hr 200-hr 1000-hr Heat 02358 Tang- N. 1725~F+ 6 hrs 321/ 39,500 22,000 <1 Il ential 1200~F + 6 hrs 1200~F 331 Radial N. 1725~F + 6 hrs 306/ 40,000 24,000 <1 I1 1200~F + 6 hrs 1225~F 331 Heat 16030 Barstock N. 1725~F + 6 hrs 311/ 35,000 20500,-3 5 1200~F 331 There was no significant difference between tangential and radial specimens, although the indications again were that tangential specimens were somewhat lower in strength at the longer time periods. Again there was a difference in tempering conditions to complicate the comparison between types of specimens. There was also some indication of increased resistance to tempering for the disc forgings. Representative microstructures of the'17-22-A."S disc are presented in Plate 4. The structure shown, typical of both the radial and tangential "17-22-A"S specimens obtained from discs, consisted of a relatively

8 coarse-grained tempered, accicular bainite. This structure is somewhat more accicular and coarser in nature than those observed in the past in the "17-22-A'S barstock with the same heat treatment. Again, no difference due to tempering conditions was observed. Rex 448 Steel Barstock and disc material from a Timken heat and a disc from a British heat were tested. The heat treatments and rupture data are presented in Tables I and II respectively. There appeared to be very little variation in rupture strength or ductility between specimens from barstock or disc or from Timken or British made heats when air cooled or oil quenched from 2115~F, as may be seen from the stress-rupture time curves of Figure 3 and the following tabulation: Rupture Strength Elongation at Type Heat BHN (psi) Fracture (%) Specimen Treatment 200-hr 1000-hr 200-hr 1000-hr Timken -- Heat 02336 Tang- A.C. 2115~F + 6 hrs 294 49,500 41,000 11 15 ential 1200~F Radial A.C. 2115~F + 6 hrs 277 48,000 41,000 14 24 1200~F Barstock O. Q. 2115~F + Te:mp- 285 48,000 (40,000) 10 ered at 1200~F Barstock N. 1850~F + Tempered 277 40,000 (31,500) 14 at 1200~F British Heat Tang- O.Q. 2115~F + Temp- 285 -- (41,000) -- 11 ential ered at 1200~F Radial O.Q. 2115~F + Temp- 285 46,000 40,000 N15 12.5 ered at 1200~F

9 Normalizing the barstock from 1850~F appeared to cause a decrease in rupture strength of approximately 8,000 psi, though the ductility remained the same. Microstructures typical of Rex 448 steels are shown inPlates 5 and 6, In all cases, the main structure was tempered martensite. Variable amounts of delta ferrite were present in all samples. It is impossible to say whether the ferrite variation was due to heat treatment, source of material, or to normal variation between specimens. Normalizing from 1850~F gave a finer structure than heat treating at 2115~F, The specimens from the british heat were definitely coarser grained. There appeared to be no significant difference in structure between oil quenching and normalizing or between radial and tangential specimens from the discs. Comparison of Properties of'17-22-A"V, " 17-22-A"S, and Rex 448 Steel Forgings The rupture properties of the'17-22-A"V disc forgings are compared to the forgings made from the other type materials in the stress - rupture time curves of Figure 4 and the following tabulation:

10 Stress for Elongation Type Heat Rupture (psi) (% in 2 in ) Specimen Treatment 200-hr 1000-hr 200-hr 1000-hr "17-22-A"V Steel - Heat 02359 Tang- N. 1800~F and -- 30,000/ — /5 3 / 6 ential Temper-ed 53,000 32,500 Radial N. 1800~F and 45,000/ 30,000/ 8.5/ — 4.5/ — Tempered -- 36,500 "17-22-A.'S Steel - Heat 02358 Tang.- N. 1725~F and 39,500 22,000 <1 1 ential Tempered Radial N. 1725~F and 40,000 24,000 1 1 Tempered Rex 448 Steel - Heat 02336 Tang- A.C. 2115~F and 49,500 41,000 11 15 ential Tempered adial A. C. 2115~F and 48,000 41,000 14 24 Tempered Rex 448 Steel - British.Heat Tang- O.Q. 2115~F and -- 41,000 - 11 ential Tempered Radial O.Q. 2115~F and 46,000 40,000 15 12.5 Tempered The data at 1100~F show that the "17-22-A"V disc has substantially higher rupture strength and ductility than the "17-22-A"S disc. The steep slope of the stress - rupture time curves for the "1722-A"V disc material results in lower rupture strength beyond about 300 hours than that of the Rex 448 disc,.

11 The ductility of the "17-22-A"V disc is considerably lower than that of the Rex 448 steel. The "17-22-A"V disc material showed more variation in structure between specimens than did.the M"17-22-A"S or Rex 448 disc material.

TABLE I Section Size, Heat Treatment, and Hardness Data for "17-22-A"V, "17-22-A."S, and Rex 448 Steels Type Stock Heat Treatment BHN "17-22-A "V Steel Heat 02359 (1) N. 1800~F + 6 hrs at 1225~F 352/375 (2) N. 1800~F + Tempered at 1200~ + 1225~F 311/321 (3) N. 1800~F + Tempered at 1200~ + 1225~F 302/323 (4) N. 1800~F + 12 hrs at 1225~F 323/337''17-22-A"'S Steel Heat 02358 (1) N. 1725~F + 6 hrs at 1200~F 352/363 (5) N. 1725~F + 12 hrs at 1200~F 321/331 (6) N. 1725~F + 6 hrs at 1200~F + 6 hrs at 1225~F 306/331 Rex 448 Steel Heat 02336 Barstock N. 1850~F + Tempered at 1200~F 277 Barstock O.Q. 2115~F + Tempered at 1200~F 285 Forging A.C. 2115~F + 6 hrs at 1200~F 277/294 British Made Heat Forging O. Q. 2115~F + Tempered at 1200~F (presumably) 285 (1) 2" x 22" forging. (2) 2" x 22" forging tempered for 6 hours at 1225~F. Half the forging tempered an additional 6 hours at 1200~F. l"-square by 7"-sections were cut and retempered 6 hours at 1225~F. (3) 2" x 22" forging tempered for 6 hours at 1225~F. 0. 5051' diameter specimens retempered for 6 hours at 1200~F + 6 hours at 1225~F. (4) 2" x 22" forging tempered for 6 hours at 1225~F. The 0.505" diameter specimens retempered for 6 hours at 1225~F. (5) 2" x 22" forging tempered for 6 hours at 1200~F. Half of disc retempered 6 hours at 1200~F. (6) 2" x 22" forging tempered for 6 hours at 1200~F. 0.505" diameter specimens retempered 6 hours at 1225 ~F.

TABLE II Rupture Data for "17-22-A"V, "17-22-A"S, and Rex 448 Steels at 1100~F Code Brinell Stress Rupture Elongation Reduction Heat Treatment No. Hardness Time of Area...___.._____..____._ ~Number (psi) (hours) (% in 2 in.) (%) "17-22-A"V Disc Heat 02359 N. 1800~F + Temper (2) Tang 352/375 44,000 353 10.0 17.8 N. 1800~F + Temper (3) Tang 311/321 50,000 235 4.5 5.9 45,000 323 4.6 6.6 35,000 660 3.0 (1) 11.0 N. 1800~F + Temper (4) Tang 302/323 30,000 1421 6.1 12. 1 N. 1800~F + Temper (3) Rad 311/321 38,000 533 12. 5 15.6 N. 1800~F + Temper (5) Rad 323/337 45,000 219 8.5 15.3 38,000 876 5.5 10.0 35,000 1199 4.3 12.0 "17-22-A"S Disc Heat 02358 N. 1725~F + Temper (6) Tang 321/331 40,000 195 1.5 2.0 25,000 699 1.5 1.5 N. 1725~F + Temper (7) Rad 306/331 40,000 198 1.0 0. 5 25,000 869 2.0 0.5 Rex 448 Disc and.Barstock "Timken" Heat 02336 (Disc) A.C. from 2115~F + 6 hrs at 1200~F Tang 294 50,000 171 11.0 49.2 40,000 1111 16.5 67.6 Rad 277 50,000 137 9.2 52.3 40,000 1149 24.5 70.5 "Timkern' Heat 02336 (Barstock) N. 1850~F + Tempered at 1200~F N 277 45,000 65 18.0 73.0 40,000 188 14.1 72.4 35,000 483 16.0 71.6 O.Q. 2115~F + Tempered at 1200~F Q 285 50,000 134 14.0 62.0 45,000 348 12.0 65.0 British Made Disc O.Q. 2115~F + Temperd at 1200"F Tang 285 40,000 1212 11.0 37.1 Rad 285 50,000 65 15. 1 62.8 40,000 1073 12.5 44.8 (1) Broke in fillet. (2) 2" x 22" forging tempered for 6 hours at 1225 F. (3)2" x 22" forging tempered for 6 hours at 12250F. Half the forging retempered 6 hours at 12000F. 1"-square by 7"-sections were cut and retempered 6 hours at 12250F. (4)2" x 22" forging tempered for 6 hours at 1225'F. 0. 505" diameter specimens were retempered for 6 hours at 1200~F and 6 hours at 1225~F. (5) 2" x 22" forging tempered for 6 hours at 1225~F. The 0. 505" diameter specimens were then given an additional 6 hours at 1225~F (6) 2" x 22" forging tempered for 6 hours at 1200~F. Half of disc then retempered for 6 hours at 1200~F. (7) 2" x 22" forging tempered for 6 hours at 1200~F. The 0. 505" diameter specimens were then retempered an additional 6 hours at 1225'F.