Stress Analysis on Disc Spring

[Mleo11]

Anyone has experience with flexures (thin annular plate, with an etched design) that acts as a spring? The flexure is used as a centering guide and provides a spring load.

I am not sure on how to approach a stress analysis.

Any input would greatly be appreciated.

 

[Mleo11]

This analysis is to find out how things change with thickness and load, so we can know what is acceptable. Attached is a more accurate sketch of the flexure.

We want it thick enough capable to handle 10XE6 cycles and finding the deflection with a 1 lb. load at the inner edge with the outer edge fixed.
The starting thickenss is .013' and material is 17-7 cres.

http://www.yourfilelink.com/get.php?fid=7997

 

[NickE]

Mleo11- usually stresses drop off as thickness is decreased. Thus for the same deflection distance the life goes up as thickness goes down. (This is a completely general statement however.)

 

[israelkk]

Mleo11

FEA analysis is needed to accurately find the deflection at 1 lb. At the same time the stresses need to be low enough for 10XE6 cycles which is basically infinite life.

Each point where the slot ends there is A high stress concentration which has a significant influence on the life cycle.

You didn't specify the desired slot width which affect how much wide each "spring" will have.

I assume you mean 17-7 condition C heat treated to CH900?

Modeling the many options and doing the analysis to find out how things (deflection, stresses, life cycle etc.) change with the thickness and load is a time consuming process.

I understand that you are designing a (solenoid?) valve, therefore, the spring deflection should be known from the flow requirements of the fluid. But at the same time the deflection is a function of the flow orifice diameter and the flow pressure. More than that, the valve (electromagnet) pulling force is the result of the orifice sealing diameter, the fluid pressure and the spring/s sealing forces. The valve opening and closing time response is a function of the forces, the plunger mass and distance movement. So, every thing is connected.

From your need to check the deflection at 1 lb for various metal thicknesses I understand that all other variables mentioned above are supposedly known and fixed. However, what will happen if the stresses needed to achieve 10XE6 cycles will dictate a deflection which will not allow the desired fluid flow? Then you will need to start all over again?

http://www.webspawner.com/users/israelkk/

 

[Mleo11]

Israelkk:

This is a solenoid type of valve design; however, I am not experienced with it and therefore my responses tend to be incomplete, thank you for bearing with me.

An FEA is the best way and we are pursuing that route. The slot width is .02" and the selected material is 17-7 ph cond. RH950. The flexure has an installed axial load of 0.6lbf, additional working deflection of each operating cycle .006” (10XE6 cycles). We are looking for a spring rate less than 100lb/in. It is going to take some time to play with thicknesses and material to get the desired spring rate and not fail.

The material selection is up in the air. It must be compatible with nitrogen textroxide, monoethyl hydrozine and nitric acid. Nitronic metals have been mentioned, but I am not familiar with it and need to do some more research. Any recommendations?

 

[RHABE]

Hi,

I often designed this in two steps:
first: bending only (this is a good approximation if the beams are not very short);
then: adding torsion.
Any elastic calculation is suitable. (I prefer Castigliano.)
Some designs work better with straight beams as these have high radial stiffness.

Be cautious with materiels and fatigue: fatigue limit is sensitive to very slight corrosive attack and measured in benign environments.
Also: only materials with cubic room centered crystallinity exhibit a stable fatigue limit above 1 million cycles, all other materials go down further.

RHABE