large spring button/support optimisation

[nr4]

I am working on a spring button assembly as shown in the sketch. It can be visualized as the same principal used to support a coil spring in a car’s coil-over shock assembly. Originally it was designed on thumb-suck values and therefore overdesigned to prevent failure. It forms part of a pneumatic actuator safety return. It is shown in figure.(A)

I need to minimize the materials used and still be on the safe side of material failure. This is done to save both weight and manufacturing cost. The rod should be minimized in diameter, but this is not my main concern. The button plate should be minimized in thickness. The button itself is machined from a large billet and the thread extends past the flat surface of the plate. Any material from the surface to the thread end is removed and wasted, so a decrease in required thread length will help considerably. This I can solve with general thread calculations.

Q.1) Now, my biggest problem is spring forces (F) on the button plate. Can this be assumed to be uniformly annular, or is there some isolated pressure area (A) where the top coil of the spring is in contact with the button plate, causing an uneven moment around the rod end in the direction of the Z-axis ? (shown in figure.(B))

Q.2) If this is the case? How do I solve the problem? Should I consider it as a simply supported beam? The problem is, it’s not so simple. Not like half the button is clamped in a wall, it’s screwed on a stick.

Q.3) If this is not the case, and the force distribution is actually annular, it seems equally hard to solve the problem. How do I approach it in this manner?
Extra info: All dimensions are metric and in millimeter. The spring exerts a maximum force 24600N. The button plate is made of 300WA steel (a South African standard). I can get more info on the springs if appropriate.
Any help will be greatly appreciated

http://files.engineering.com/getfil...a11-9976-58262fbc4e8f&file=spring_support.bmp

Regards
Nr.4

 

[nr4]

Oh yes. The loading type can be considered as static, no general impact or reflex sources will cause shock problems.

 

[desertfox]

Hi nr4

If the spring ends are squared and ground then the force can certainly by taken as annular.
I would then just take the resultant force of the spring and calculate the tension in the rod and threads as the button is purely held on by the threads.
In other words just take the preload of the spring and check the threads for stripping and just make sure the rod diameter is sufficient to cope with the tension generated by the spring force.
Assuming the spring force is annular eliminates any bending moments that would occur in the rod ie they cancel out on each side of the spring.

regards

desertfox

 

[nr4]

Thus far I have decided to do a sectional analysis like cutting a slice of cake and consider a sectional part of the annular load applied by the spring. The spring is ground by the way.
That's the simplification yes. I am currently solving the problem in another manner aswell, and that is by FEM analysis in solidwoks. The problem is I need actual results to quantify my calculations. Maybe I'll get the boss approve some destructive testing
I'll keep everyone posted. Any other help will be appreciated.

 

[btrueblood]

If you find a book on springs, such as Mechanical Springs by A. M. Wahl, or other publications from the Spring Manufacturer's Institute, you can find papers that show that all coil springs will generated some bending moment as they are flexed (and if you think about how the coil wire first contacts the end plates at one, off-axis point, it's obvious that some moment should occur). The effect is reduced as the number of coils increases (because, presumably, the spring "squirms" a bit in response to the unequal stresses, redistributing the load).

Wahl gives the formula

e/R = 1.123(Z-1)

e is the eccentricity of the resultant spring load (distance from spring axis), R is the mean coil radius.

Z = 1 + .5043/N + .1213/N² + 2.058/N³

N is the "number of solid coils", which he later defines as the number of coils between tip contact points + 1.5 turns.

 

[desertfox]

Hi nr4

Another way to look at the stresses on the flat button would be to look in Roark's book of formula's for stress and strain.
In my fifth edition page 338 shows a flat plate with the inner edged fixed and the outer edge free with an uniform annular load line at a intermediate radius.
I also have the book that btrueblood mentions, if you give full details of the spring and rod details I'll try to help further.

regards

desertfox

 

[nr4]

Kickass! This is why I am here. I have very little recourses at my disposal. The rod is 26mm in diameter but M24 thread at the button end and made of 316 stainless. For the spring, it is ground flat at both ends. The exact dimensions I’ll have to measure at our workshop a little later this afternoon. I’ll post that as soon as I can.