22-Mec-A2 Kinematics and Dynamics of Machines · December 2017
Question 3 of 7: Radial cam with flat-faced follower
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper: National Exams — December 2017 · 16-Mec-A2 Kinematics and Dynamics of Machines · 3 hours, open book · Answer five questions, at least one from Part B (Part A mechanisms & machine dynamics Q1–5, Part B mechanical vibration Q6–7). Marks: 20 each. All seven questions are solved here.
Reference texts (subject). R. L. Norton, Design of Machinery, 6th ed. (instant centres & Kennedy’s rule §6.3–6.4, cams Ch. 8, epicyclic trains §9.7–9.9, balancing Ch. 13); Uicker, Pennock & Shigley, Theory of Machines and Mechanisms, 5th ed.; C. E. Wilson & J. P. Sadler, Kinematics and Dynamics of Machinery, 3rd ed.; S. S. Rao, Mechanical Vibrations, 6th ed. (Part B: Ch. 3 base excitation, Ch. 6 & Ch. 9 multi-DOF / torsional).
Check (measured geometry). Questions 1 and 2 carry no printed dimensions — the exam instructs the candidate to scale every length off the drawings (1:6 for Q1, 1:10 for Q2). The joint coordinates used below were read from the printed figure of the figures; the resulting linear quantities are therefore accurate to roughly ±3–5 %. Scale-free results (time ratio, mobility, angular-velocity ratios, mode shapes) are unaffected.
Question 3: Radial cam with flat-faced follower (20 marks)
Given. Flat-faced translating follower; $\omega=1500$ rpm $=157.08$ rad/s (const); rise $h=2$ in over $\beta=90^\circ=\pi/2$; dwell; fall $h=2$ in over $\beta=90^\circ$. Objective: minimum peak acceleration with finite jerk.
Find. Motion program & its $s,v,a,j$ equations; $a_{\mathrm{max}}$, $v_{\mathrm{max}}$, $j_{\mathrm{max}}$ (rise and fall); base circle; pressure angles at $45^\circ$ and $315^\circ$.
Modified-trapezoidal RISE program (normalized): displacement $s$, velocity $v$, acceleration $a$ (trapezoidal, lowest possible peak), and jerk $j$ (bounded, finite) versus cam angle 0→β.
Approach. Among finite-jerk double-dwell programs the modified-trapezoidal acceleration curve has the smallest peak acceleration, so it is the correct choice for “minimize maximum acceleration.” Use its standard coefficients, then note that a flat-faced follower always has zero pressure angle, so the design constraint is convexity (base-circle), not pressure angle.
Program choice. The modified-trapezoid builds the acceleration from three sine-ramp blends (rise-hold-fall of $a$) so that jerk stays finite while the acceleration plateau is as low as possible. Its peak factors are
$$C_a=4.888,\quad C_v=2.000,\quad C_j=61.43$$
with $a_{\mathrm{max}}=C_a\dfrac{h\omega^2}{\beta^2},\ v_{\mathrm{max}}=C_v\dfrac{h\omega}{\beta},\ j_{\mathrm{max}}=C_j\dfrac{h\omega^3}{\beta^3}$. In normalized form $x=\theta/\beta\in[0,1]$ the acceleration is $a(x)=A\,p(x)$ with $p$ = $\sin(4\pi x)$ on $[0,\tfrac18]$, $1$ on $[\tfrac18,\tfrac38]$, $\cos\!\big(4\pi(x-\tfrac38)\big)$ on $[\tfrac38,\tfrac58]$, $-1$ on $[\tfrac58,\tfrac78]$, $-\cos\!\big(4\pi(x-\tfrac78)\big)$ on $[\tfrac78,1]$; $v$ and $s$ are its successive integrals scaled so $s(1)=h$. The fall uses the mirror ($s_{\text{fall}}=h-s_{\text{rise}}$).
Peak velocity. $v_{\mathrm{max}}=2.000\dfrac{(2)(157.08)}{1.5708}=400\ \text{in/s}=\boxed{10.16\ \text{m/s}}$ (sets the flat-follower face width, $\pm v_{\mathrm{max}}/\omega=\pm 2.55$ in).
Peak jerk. Because $\beta_{\text{rise}}=\beta_{\text{fall}}=90^\circ$, the rise and fall jerks are equal in magnitude:
$$j_{\mathrm{max}}=61.43\frac{(2)(157.08)^3}{(1.5708)^3}=1.23\times10^{8}\ \text{in/s}^3=\boxed{3.12\times10^{6}\ \text{m/s}^3}$$
Base circle (convexity). For a flat follower the radius of curvature is $\rho=R_b+s+s''$ (with $s''=d^2s/d\theta^2$); a cusp/undercut occurs if $\rho\le0$. The most negative $s+s''$ over the cycle is $-2.51$ in, so choose
$$R_b\ge 2.51\ \text{in}\ \Rightarrow\ \boxed{R_b=3.0\ \text{in}}\ (\rho_{\min}\approx0.5\ \text{in}\gt 0)$$
The resulting contour is sketched below.
Pressure angles. For a flat-faced follower the contact normal is always parallel to the follower axis, so the pressure angle is identically zero at every cam angle:
$$\phi(45^\circ)=\phi(315^\circ)=\boxed{0^\circ}\ (\lt 30^\circ,\ \text{satisfied automatically})$$
Hence no pressure-angle modification is needed. (If contact stress or a cusp were the concern, the remedy would be a larger base circle, as sized above — not a follower offset.)
Generated cam profile for the flat-faced follower on the chosen $R_b=3.0$ in base circle (dashed): the contour is the envelope $x=(R_b+s)\cos\theta-s'\sin\theta$, $y=(R_b+s)\sin\theta+s'\cos\theta$ of the $90^\circ$ modified-trapezoidal rise, $180^\circ$ dwell at 2 in and $90^\circ$ fall. It is convex everywhere — the tightest curvature is $\rho_{\mathrm{min}}=0.50$ in in the fall — so there is no cusp or undercut, and the follower is shown at the start of the rise.