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).
Calculations and analysis of crank stiffness and torsional resonance speeds, comparing calculated results with test data.
| Identifier | ExFiles\Box 63\4\ scan0070 | |
| Date | 9th March 1929 guessed | |
| 2. (Handwritten: 16c 151) Dimensions: d₁ = 2 d₂ = 1.5 r = 2.26 (Handwritten: 2.25) S₁ = 1.25 S₂ = 0.875 ω = 2.25 (Handwritten: 2.20) a = 1.45 h = 0.75 (Handwritten: .73) b = 0.525 Equivalent length 2b + 0.8h = 1.65 inches (Handwritten: 1.634) 3 (d₁⁴ - S₁⁴) a/4 (d₂⁴ - S₂⁴) = 3.29 inches (Handwritten: 3.30) 3 (d₁⁴ - S₁⁴) r/2 h w³ = 5.35 inches (Handwritten: 5.89) Total l = 10.29 inches (Handwritten: 10.82) Stiffness per crank = 11.8 x 10⁶ x π x (d₁⁴ - S₁⁴)/32x10.29 = 1.53 x 10⁶ lb.inches per radian. (Handwritten: expl. stiffn comp.) Working from drawing M.3751, I make the calculated stiffness per crank for Shaft 15.c. to be 1.9 million pounds inches per radian and the test stiffness 1.718 millions. The calculated result is thus 10% higher than the test value, as compared with 18.5% given in your letter. I note that you have made tests on many other shafts and that the calculated results are within plus or minus 6% of the test results. I should be very glad to have particulars of these checks on the formula if you can manage to send them. (Handwritten in margin: 1.646) As regards agreement between calculated and observed resonance speeds, I cannot think that the simpler method of calculation you mention is more satisfactory than the equivalent flywheel method. I suggest that the test bed conditions have affected the issue and that with a sufficiently heavy flywheel combined with a flexible drive beyond to the brake, the agreement would be better. The flywheel system method throws more light on the whole matter of torsional resonance for it enables minor (Handwritten in margin: final bed appears most real.) criticals / | ||