25-Nav-B1 Applied Thermodynamics and Heat Transfer (25-Mec-A1) · December 2016
Question 5 of 6: Overhung Rotating Shaft — Fatigue Diameter and Deflection
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exam 98-Mar-B1, December 2016. Open book, 3 hours, 100 marks. Part I (Problems 1–2) is compulsory; three of the four Part II problems (3–6) are required. All six problems are solved below.
Given. Bearing A at $x=0$, bearing B at $a=7$ in; the transverse load $P=500$ lb acts at $b=9$ in — i.e. it overhangs bearing B by $c=b-a=2$ in; the shaft extends to $l=12$ in. Because the load direction is fixed while the shaft turns, the bending is fully reversed.
Given data
Quantity
Value
Bearing span $a$ (A–B)
7 in
Load position $b$ / overhang $c$
9 in / 2 in beyond B
Transverse load $P$
500 lb
Torque $T_{\min},\,T_{\max}$
$-100,\ 600$ lb·in
$S_{ut},\ S_y,\ E$
108 ksi, 62 ksi, $2.9\times10^4$ ksi
Surface / reliability / temp
machined / 99 % / room
Find. The shaft diameter for a fatigue safety factor $n=2$ (DE-Goodman), and the maximum deflection at that diameter.
The load overhangs bearing B by 2 in, so the peak bending moment is at B; rotation makes that moment fully reversed.
Approach. Statics gives the peak (fully-reversed) bending moment at bearing B; the torque splits into mean and alternating parts. Build the Marin endurance limit for the machined, 99 %-reliable shaft, then solve the DE-Goodman equation for the diameter at $n=2$, round up to a standard size, and finally compute the overhang deflection.
Bending moment (fully reversed). The overhung load bends the shaft most at bearing B:
$$M_a=P\,c=500\times2=1000\ \text{lb}\cdot\text{in},\qquad M_m=0.$$
Endurance limit (Marin). $S_e'=0.5S_{ut}=54$ ksi; machined surface $k_a=2.70\,S_{ut}^{-0.265}=0.781$; size $k_b=(d/0.3)^{-0.107}$; load $k_c=1$ (bending); temperature $k_d=1$; reliability 99 % $k_e=0.814$. Thus $S_e=k_a k_b k_e S_e'$ (a function of $d$).
DE-Goodman for the diameter. With $K_f=K_{fs}=1$ (no fillet/keyway data given at the bearing seat),
$$\frac{1}{n}=\frac{16}{\pi d^{3}}\!\left[\frac{\sqrt{4(K_fM_a)^2+3(K_{fs}T_a)^2}}{S_e}+\frac{\sqrt{4(K_fM_m)^2+3(K_{fs}T_m)^2}}{S_{ut}}\right].$$
Setting $n=2$ and iterating on the size factor gives
$$\boxed{\ d=0.90\ \text{in}\ \rightarrow\ \text{use }d=\tfrac{15}{16}=0.9375\ \text{in (next standard)}\ }$$
at which $S_e\approx30.4$ ksi and the realized factor exceeds 2.
Maximum deflection. The largest deflection is at the overhung load. For a load $c$ beyond the support, over span $a$,
$$\delta=\frac{P\,c^{2}(a+c)}{3EI}=\frac{500(2)^2(7+2)}{3(2.9\times10^{7})\,I},\quad I=\frac{\pi d^4}{64}=0.0379\ \text{in}^4,$$
$$\delta=5.5\times10^{-3}\ \text{in}\ (\approx0.14\ \text{mm}).$$
Check: $K_f=K_{fs}=1$ is assumed because no fillet radius or keyway is dimensioned at the critical section. A shoulder fillet or a keyway there would raise $K_f$ to roughly 1.6–2.5 and require a proportionally larger diameter; the design must be rechecked once the bearing-seat geometry is fixed.