NivaarExam PrepOfficial exam papers ↗

22-Mec-B1 Advanced Machine Design · May 2015

Question 2 of 6: Bolted-Joint Design — Size, Preload and Safety Factors

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

Paper format. 22-Mec-B1 Advanced Machine Design — National Exams, May 2015. Open book, 3 hours, 100 marks. Part I (Problems 1–2) compulsory; answer any three of the four Part II problems (3–6). All six problems are solved here as a complete study resource.

Reference texts. R. G. Budynas & J. K. Nisbett, Shigley’s Mechanical Engineering Design (10th ed.) — deflection §4, fatigue §6, shafts §7, bolted joints §8, journal bearings §12, brakes & clutches §16; R. C. Juvinall & K. M. Marshek, Fundamentals of Machine Component Design; R. L. Norton, Machine Design: An Integrated Approach; R. C. Hibbeler, Mechanics of Materials.

Question 2: Bolted-Joint Design — Size, Preload and Safety Factors (30 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Check (assumed data). The figure is dimensionless—it labels only the bolt diameter $d$, the member grip $l=l_1+l_2$ and the load $P$—so, as the exam permits (“assume any missing data and state it”), the design is carried out for a representative separating load $P=20\text{ kN}$ carried by a single bolt through two steel members of total grip $l=40\text{ mm}$ ($l_1=l_2=20\text{ mm}$), $E=207\text{ GPa}$. The method is general; only the numbers depend on these stated assumptions.

Given. A single bolt clamps two steel plates that an external load $P$ pulls apart; grip $l=40$ mm, members steel ($E=207$ GPa), assumed load $P=20$ kN.

Given / assumed data
QuantitySymbolValue
External separating load$P$20 kN
Grip (both members)$l=l_1+l_2$40 mm
Member / bolt modulus$E$207 GPa
Trial bolt—M12×55, Class 8.8

Find. A bolt size and preload $F_i$; the factors of safety against yielding and against joint separation; and the preload (as a percentage of proof strength) that maximizes them.

P/2P/2P/2P/2d (bolt)l = 40 mml1l2clamped members
Bolted joint: one bolt of diameter $d$ through two steel members of combined grip $l=l_1+l_2=40$ mm; the external load $P$ (shown split as $P/2+P/2$) pulls the members apart.

Approach. Compute the bolt and member stiffnesses to get the joint stiffness constant $C$; the bolt then carries $F_b=F_i+CP$ while the members shed $(1-C)P$. Write the yielding and separation load factors as functions of preload, then equate them to find the preload that maximizes the smaller of the two.

  1. Select a trial bolt and list its areas. Take an M12 coarse-thread bolt, Class 8.8 ($S_p=600$ MPa proof, $S_y=660$, $S_{ut}=830$ MPa). Its shank and tensile-stress areas are $$A_d=\frac{\pi}{4}d^{2}=\frac{\pi}{4}(12)^2=113.1\text{ mm}^2,\qquad A_t=84.3\text{ mm}^2.$$
  2. Bolt stiffness $k_b$. For an M12×55 bolt the threaded length is $L_T=2d+6=30$ mm, so the unthreaded shank is $l_d=55-30=25$ mm (all within the 40 mm grip) and the threaded portion in the grip is $l_t=40-25=15$ mm: $$k_b=\frac{A_d A_t E}{A_d\,l_t+A_t\,l_d}=\frac{(113.1)(84.3)(207000)}{(113.1)(15)+(84.3)(25)}=5.19\times10^{5}\text{ N/mm}.$$
  3. Member stiffness $k_m$. Using Shigley’s frustum fit for a two-plate steel joint (constants $A=0.78715$, $B=0.62873$): $$k_m=E\,d\,A\,e^{B d/l}=(207000)(12)(0.78715)\,e^{0.62873(12/40)}=2.361\times10^{6}\text{ N/mm}.$$
  4. Joint stiffness constant $C$. The fraction of external load felt by the bolt is $$\boxed{C=\frac{k_b}{k_b+k_m}=\frac{5.19\times10^{5}}{5.19\times10^{5}+2.361\times10^{6}}=0.180.}$$ So the bolt sees $CP=0.180(20)=3.60$ kN of the external load and the clamped members lose $(1-C)P=16.4$ kN of clamping force.
  5. Proof load and preload window. The proof load is $F_p=S_p A_t=(600)(84.3)=50.6$ kN. Good practice keeps the preload in $F_i=0.75$–$0.90\,F_p$; the exact optimum is found next.
  6. Yielding and separation load factors. The bolt reaches proof when $F_i+n_y\,CP=F_p$, and the joint separates when $F_i=n_0(1-C)P$: $$n_y=\frac{F_p-F_i}{CP},\qquad n_0=\frac{F_i}{(1-C)P}.$$ $n_y$ falls and $n_0$ rises as the preload increases, so the two curves cross at the preload that maximizes the smaller factor.
  7. Optimum preload. Setting $n_y=n_0$ and solving, $$\frac{F_p-F_i}{CP}=\frac{F_i}{(1-C)P}\;\Rightarrow\;F_i=(1-C)F_p.$$ $$\boxed{F_i=(1-C)F_p=(0.820)(50.6)=41.5\text{ kN}\;=\;82.0\%\ \text{of proof}.}$$ This sits neatly inside the recommended $0.75$–$0.90\,F_p$ band, so the M12 Class 8.8 bolt tightened to $\approx82\%$ of proof is the design.
  8. Resulting safety factors. At the optimum both factors take the common value $$\boxed{n_y=n_0=\frac{F_p}{P}=\frac{50.6}{20}=2.53.}$$ The joint is safe against both bolt yielding and separation with a comfortable factor of about $2.5$.
Problem 2 — results
QuantityValue
Bolt selectedM12×55, Class 8.8 ($A_t=84.3$ mm²)
Joint stiffness constant $C$0.180
Proof load $F_p$50.6 kN
Optimum preload $F_i$41.5 kN (82.0 % of proof)
Factor vs. yielding $n_y$2.53
Factor vs. separation $n_0$2.53