cosmoswork buckling analysis result does not match hand calculation

[Donaldlo]

I am trying to do a simple buckling analysis of a delrin pressure case using cosmoswork.

Case dimension:
OD: 8"
wall thickness: .4"
End cap thickness: .4"
Length: 2.625"

Material properties:

E: 2.83E+9 Pa
Yield stress: 8.9e+6 Pa
Poisson ratio: .3

Test load: 14.7 Psi

According to Roarks's Formulas the end cap should fail approx around 10m.

But the program give a load factor of 54 and the center of the end cap bulge outward rather than inward.

I am wondering wether I setup the analysis correctly? Any help or advice would be greatly appreciated.

Thanks

 

[ERT]

Hi Donaldlo,

I did a very quick run and go the same answer 54. Sounds good, right?

No. The problem is that you appear to be using the CW "linearized buckling analysis" that uses a linear analysis engine that does not account for the membrane stress created due to deflection. I believe this membrane stress is significant in this analysis so you should not be using this method.

Regards,

ERT

http://www.akeng.com

 

[corus]

I would doubt if Roark's formula involved any non-linear effects but probably assumes much more than an FE analysis and is probably more 'incorrect'. Can you give the case number used in Roark?

corus

 

[Donaldlo]

Thanks for your response ERT and Corus. In terms of Roark's formula I used case 1c for the cylinder and case 10a for the end cap.

Thank again.

Donald

 

[ERT]

Donald,

We must have different versions of Roark as our case numbers do not line up.

I think that you should focus on the end cap. Can you find a case that has edge compression and lateral pressure? I cannot.

Sounds like a physical test maybe a better approach.

Regards,

ERT

 

[corus]

I'm not sure which cases you're referring to either. In my version of Roark VI ed. I'd use case 20 of table 35 for the elastic stability of a thin tube with closed ends (if that's what best describes your pressrue case)

corus