Skip Navigation Links.
The book takes the subject from an introductory level through advanced topics needed to properly design, model, analyze, specify, and manufacture cam-follower systems. Presented from the book:
Cam Design and Manufacturing Handbook
(Cam Systems Design)

Buy this book
   by Robert L. Norton
Published By:
Industrial Press Inc.
Up-to-date cam design technology, correct design and manufacturing procedures, and recent cam research. SALE! Use Promotion Code TNET11 on book link to save 25% and shipping.
Add To Favorites!     Email this page to a friend!
 
<-- Previous Page
Page   of 6   
Next Page -->

18.2 RULES OF THUMB FOR CAM DESIGN

Rules of thumb are meant to be broken! They are only general guidelines and should not be taken as gospel. Nevertheless, they can be of some value in guiding the new cam designer as they tend to capture the results of successful experience. With that caveat in mind, we present the following short list of rules that have proven themselves over time.

 

1 Do not arbitrarily insert dwells in a cam motion function unless they are really needed. Any stopping of the follower will exact a dynamic penalty in terms of reduced time available for the required motions. It may be better to overtravel the follower in a gently curved path that reverses its direction for the return motion than to stop and restart it.

 

2 For conventional double-dwell motions, the modified sine or 3-4-5 polynomial functions are good all-around “workhorse” choices, and are better than modified trapezoidal motion when vibration of the follower is considered.

 

3 For low vibrations in conventional double-dwell motions, the 4-5-6-7 polynomial is superior and the cycloidal curve is an acceptable substitute as long as their somewhat higher accelerations do not overstress the parts. Differences in theoretical peak accelerations between various functions is less important than the actual peak value when follower vibration is factored in.

 

4 To obtain extremely low vibrations in applications where camshaft speed is essentially constant, a polydyne or splinedyne approach is best.

 

5 For nonconventional (i.e., not double dwell) motions, especially when intermediate follower positions or velocities are specified, a polynomial or B-spline function will usually give the best design. If many boundary conditions are specified, then a polynomial may not work (or will be suboptimal) and a B-spline will be needed.

 

6 Pressure angles should be limited to about 30 ° with conventional translating followers, though if a low-friction ball-slide is used to guide the translating follower, then pressure angles up to about 35 ° may be tolerable. For oscillating arm followers, a maximum pressure angle of about 35 ° is generally acceptable.

 

7 When a roller follower is used, the absolute value of the radius of curvature ñ of the path of the roller centerline (pitch curve) must be kept larger than the radius Rf of the roller at all points to avoid undercutting. A ratio of | Rf | / ñ > 2 is a good target, though a ratio of 1.5 has been used successfully in some applications. The Hertzian surface stress should definitely be checked when this ratio is small.

 

8 When a flat follower is used, there can be no negative radius of curvature allowed in the cam surface contour.

 

9 When air cylinders are used as follower return springs, an accumulator should be used and the fittings, hoses, valves, etc., connected to the cylinder should be as large in diameter and as short as possible to minimize impedance.

 

10 “Full-complement” needle roller bearings used as cam followers may be shorter lived than caged roller bearings despite the superior load capacity of the former type. This is due to their poorer grease storage capacity and the fact that the uncaged needles rub on one another.

 

11 While a cylindrical roller follower has theoretically lower surface stress than a crowned follower in the same application, unless parallel alignment of the axes of the cylindrical roller follower and the cam is accurately and stiffly maintained, a cylindrical roller follower can actually have potentially higher surface stress than a crowned roller. If the alignment between cam and roller axes cannot be made accurate (including dynamic deflections), then a crowned roller may be needed, though its gentle crown radius can accommodate only slight axial misalignment.

 

12 Track cams with single roller followers will experience crossover shock, and the roller will have to reverse direction at the load reversal points, causing slip and wear. Some designers of track cams provide a follower spring to load the roller follower always against one side of the track. The other side of the track then becomes essentially an “insurance policy” against possible gross follower jump in the event of a follower spring failure or a tooling jam. Another approach to cure crossover shock and roller reversal is to use two rollers in the track, each contacting only one side of the groove and spring-loaded apart to accommodate slight deviations in groove width. A ribbed cam with two rollers pinching across the rib gives the same effect.

 

13 Topical grease lubrication of open cam surfaces in a dirty environment can be “worse than nothing” as the grease will trap airborne dirt particles and increase abrasive wear rate. Lubrication of a cam-follower joint is best done by enclosing both in an oil-filled box.

 

14 Yoke mounted roller followers are preferred to stud mounted rollers because of their axle’s smaller bending deflections and the possibility of centered loading that can reduce or eliminate torsional moments on the follower arm.

 

15 Attention should be paid to the design of follower linkages and camshafts to minimize or balance applied bending and torsional moments and their resulting deflections.

 

16 Stiff, light cam-follower trains will have superior dynamic performance with reduced vibration and improved fidelity to the theoretical motion program specified.

 

17 Avoid large amplification ratios between the end effector stroke and that of the cam follower motion as the reflected inertia (effective mass) of the end effector and its intervening links is increased by the square of this ratio. It may be better to suffer a larger cam follower motion and its concomitantly larger maximum pressure angle in return for a reduced amplification ratio.

 

18 Cam-follower systems typically have very low structural damping (are high “Q”) having dimensionless damping ratios æ of between 0.05 and 0.10, and so are prone to vibration problems.

 

19 Care should be taken to reduce clearances in follower train pivots and joints to minimize within-joint impacts on load reversal, which will exacerbate vibration. It is possible to use springs at the joints between links to take-up the joint clearance in one direction with sufficient force to eliminate impact and improve accuracy. Placing the follower return spring at the end effector as in the automotive valve train, rather than near the cam, can have the same effect.

 

20 When high accuracy is required of the end effector position in a cam-follower train, it is common practice to design the system with deliberate overtravel (sometimes called lost motion) and provide a “hard stop” on ground that the end effector contacts before the cam motion is complete. Some deliberate compliance in the form of a spring must then be provided within the follower train to avoid jamming. This technique also creates an impact event. To minimize the force and vibration of the impact, the cam can be designed (using polynomials or B-splines) to provide a low magnitude, constant velocity ramp at the end of the stroke to control impact velocity as is done in automotive valve cams.[5]

 

Copyright 2004, Industrial Press, Inc., New York, NY

 

<-- Previous Page
Page   of 6   
Next Page -->
er