From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Analytical study on the dynamic fatigue life of rubber under various temperature, shear, and strain conditions.
| Identifier | ExFiles\Box 178\2\ img018 | |
| Date | 15th January 1940 | |
| JANUARY 15, 1940 ANALYTICAL EDITION 23 FIGURE 9. EFFECT OF TEMPERATURE ON DYNAMIC FATIGUE LIFE GRAPH Y-AXIS: RATIO OF FATIGUE LIFE AT A GIVEN TEMPERATURE TO THAT AT 100°F (LOG SCALE) GRAPH X-AXIS: RUBBER TEMPERATURE °F GRAPH LEGEND: A: DROP IN FATIGUE LIVES IN A DEPENDS ON FREEZING CHARACTERISTICS OF STOCK B: DEPENDS ON AGING CHARACTERISTICS OF STOCK FIGURE 10. RUBBER VIBRATED IN SHEAR % minimum shear = (d_min. / T) (100) % maximum shear = (d_max. / T) (100) a in Figure 10 is a sketch of a double shear unit which consists of two identical rectangular bodies of rubber of thickness T bonded between two outer metal plates and a central metal plate. If the central plate is displaced a distance d relative to the two outer plates the rubber is put in shear; and we express the magnitude of the shear deflection either as the shear ratio d/T or as the shear percentage (d/T) 100. b represents the two oscillation extremes of such a unit vibrated in shear. The minimum deflection in the shear cycle is d_min.; the maximum, d_max.. The shear cycle is then specified as one of vibration between the shear percentages (d_min./T) 100 and (d_max./T) 100. Figure 11 illustrates results obtained on shear samples of the 50 durometer stock discussed above. The two outer metal plates were held fixed, and the central plate was vibrated back and forth along its length between two extremes as shown by the dotted contours. The oscillation length was a 50 per cent shear oscillation. In shear mountings the rubber may be placed in lateral strains which are normal to the center plate. Three conditions of lateral strain are shown in the figure: 0 per cent, 12.5 per cent compression, and 25 per cent extension. The first row corresponds to a -25 to +25 per cent shear cycle; the second row, to a 0 to 50 per cent shear cycle; and the third row, to a 75 to 125 per cent shear cycle. The striking observation about these test data is that the results in shear fatigue are entirely in accord with what would have been predicted from the linear fatigue data. Consider samples B and C. B, vibrated between 0 and 50 per cent shear, had a fatigue life of 1,000,000 cycles. C, vibrated between 75 and 125 per cent shear, had a fatigue life of 15,000,000 cycles. The shear oscillation cycle for both samples was one of 50 per cent. C had a much greater fatigue life than did B. This is due to the fact that the rubber elements in C were in strain at all times during the shear cycle, whereas those in B went back to a condition of zero strain once each cycle. Essentially the elements in C were working up on the linear extension fatigue hump, whereas those in B were being vibrated at the linear fatigue minimum. In sample A the shear cycle was one from -25 to +25 per cent, a total shear cycle of 50 per cent. The fatigue life of A was seven times that of B. Actually A was subjected to an alternating 25 per cent shear cycle; once from 0 per cent shear to -25 per cent shear, then once from 0 per cent shear to +25 per cent shear. The double 25 per cent shear cycles are essentially two repeated 0 to 25 per cent shear cycles. A shear unit of this same 50 durometer stock vibrating through a single shear cycle of 0 to 25 per cent shear has a fatigue life of about 14,000,000 cycles. It is therefore consistent that A—each of whose cycles represents two vibrations from 0 to 25 per cent—should last only 7,000,000 cycles. Sample D had a greater life than A; E had a greater life than B. The reason is that the rubber elements were being vibrated essentially with a minimum strain which falls in the compression region, and on the basis of linear fatigue data D and E should have the higher fatigue lives. In F, however, the dynamic fatigue life was considerably less than in C. In F the rubber elements which under no shear were under lateral compression were essentially relieved from the compression by the extension resulting from the higher shear; and the rubber elements were being worked in a low dynamic fatigue region near the zero of strain. Sample G had a higher fatigue life than A; G should have a higher life because the rubber elements were always under extension strain during the vibration. Finally, I had a higher dynamic fatigue life than did C. The rubber elements in I were in higher extension than in case C; in other words, the shear strain and the lateral extension strain essentially added to place the rubber unit in a more favorable fatigue region. FIGURE 11. DYNAMIC FATIGUE RESULTS ON SHEAR SAMPLES OF 50 DUROMETER STOCK TABLE DATA: SHEAR CYCLE / LATERAL STRAIN NONE | 12 1/2 % COMPRESSION | 25% EXTENSION -25% TO +25%: A 7-MILLION | D 20-MILLION | G 12-MILLION 0% TO 50%: B 1-MILLION | E 2-MILLION | H 2-MILLION 75% TO 125%: C 15-MILLION | F 2-MILLION | I 40-MILLION Many theories can be devised to explain the way the dynamic fatigue properties of rubber vary with the strain and the strain oscillation conditions, but such theories are still conjectures. Literature Cited (1) Busse, W. F.{Mr Friese}, IND. ENG. CHEM., 26, 1194 (1934). (2) Cassie, Jones, and Naunton, Trans. Inst. Rubber Ind., 12, 49 (1936). (3) Cooper, L. V.{VIENNA}, IND. ENG. CHEM., Anal. Ed.{J. L. Edwards}, 2, 392 (1930). (4) Jenkin, C. F.{Mr Friese}, "Report on Materials of Construction Used in Aircraft and Aircraft Engines", London, H.{Arthur M. Hanbury - Head Complaints} M.{Mr Moon / Mr Moore} Stationery Office, 1920. (5) Rainier, E.{Mr Elliott - Chief Engineer} T., and Gerke, R.{Sir Henry Royce} H.{Arthur M. Hanbury - Head Complaints}, IND. ENG. CHEM., Anal. Ed.{J. L. Edwards}, 7, 368 (1935). PRESENTED before the Division of Rubber Chemistry at the 98th Meeting of the American Chemical Society, Boston, Mass. PRINTED IN U. S. A.{Mr Adams} | ||