Circular Flat Plate under stress calculations

[bonnienut]

I have been asked to look into a basic study of circular flat plates under loading (pressure). I have calculated basic formula's for a circular plate under pressure loading and obtained results using Roark's Formula's for Stress and Strain 7th Edition. After the hand calculations were done i modelled the circular flat plate in Inventor 11.0 Professional then used the Stress Analysis feature in this obtained results with same loadings as hand calculations, in the results from AIP 11.0 compared to hand calculations there is a big difference, does anybody know why?
Kieran

 

[GregLocock]

It would probably help if you identifed your problem more exactly.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GBor]

The list of potential problems is long: proper application of Boundary Conditions, mesh sensitivity, proper application of load, material property input, thick plate vs. thin plate evaluation, etc.

 

[vonlueke]

bonnienut: For the simply-supported case, make sure your boundary conditions allow the plate to slide freely in the in-plane direction (except at one point), and also make sure your mesh isn't too coarse. Then post your percent error relative to the Roark result.

 

[iainuts]

I'd agree with the above two responces. The list of potential issues is huge, and perhaps the best way of narrowing it down is to be very specific.

Inherently, there is no reason why Roark's and and FEA analysis should give significantly different results.

Roark's is not the easiest thing to use the first time you try it. If you've made a program or spreadsheet to do the cases in Roark, you might want to verify you did them correctly by posting a test run. If you purchased software, obviously they should have already been tested.

 

[vonlueke]

bonnienut: I meant to say, except at, e.g., three points in the tangential direction at 0, 90, and 180 deg on the plate edge.

 

[jte]

I tend to agree with the above responses. One issue which hasn't been mentioned is how quickly the problem goes nonlinear. The hand calc's are derived based on a linear response both from a geometric and a material behavior. If the pressure which you are applying is calculating out as causing a deflection more than about 1/2 the plate thickness, then you need to use the nonlinear (iterative) calc's provided in Roark's Formulas. As pressure is applied, the plate handles the load primarily through bending. As the presure is increased, the plate dishes out and begins to carry some of the load through membrane behavior. The linear hand calc's do not account for this. As pressure is further increased, the bending at the edges exceeds yield and the material nonlinearity begins to further complicate the problem.

As Greg and others pointed out above, a bit more background on your particular problem will allow us to help out in more detail.

jt

 

[arto]

If it's a thick plate, don't forget to include the shear deflection effect

 

[Beggar]

If the sress package in Inventor is anything like the teaser package of Cosmos included with SolidWorks, I would pretty much ignore the output.


--------------------
How much do YOU owe?

http://www.brillig.com/debt_clock/

--------------------