Simulating bolted joint using ANSYS Workbench 2

[buggles]

Hi,

I'm trying to simulate the effects of a bolted joint withstanding a pressure of 15000psi using workbench.

Here are the details: I have an 12mmthick annular ring, 70mm ID, 130mmOD which is held by 6 off M8 bolts which are counter bored into the annular ring at 6mm deep. The counterbore holes are 24mmDIA.

The pressure is applied at the opposite side to the counterbore holes and is applied evenly from the inner radius of the ring up to an 80mm DIA, i.e. an area of approximately 1178.1mm^2.

All are modelled using volumes, the ring is one volume and the bolts are 6 further volumes.

I've tried using contact elements in ANSYS Classic with the contact manager but not sure if I'm applying my constraints correctly since I had to apply zero DOF at the contact surface for the analysis to converge. This gave me a max stress of 142N/mm^2 which WOULD appear satisfactory if I was happy with my constraints.

I've now imported the same geometry into ANSYS Workbench. By holding the bolts at the shear area of the shaft I obtain a stress of in excess of 2000N/mm^2 if I use bonded or no separation contact?!?!? Also, if I use frictional contact with a co-efficient of friction of 0.16 the analysis does not converge!?!?!?

Can anyone please advise me on the correct way to analyse this bolted connection?

Apologies for the long thread but if anyone could advise I thought I would give as much information as I could to begin with.

Thanks

 

[DLTguy]

Best advise.. Please get a consultant who will do this and give you a customized training. BTW, I am not one.

Regards,
DLT

 

[Transient1]

I agree with DLTguy, contact a consultant or be prepared to take a long time learning a lot. Here's the long list of things to know:

(1) The various methods of modeling a bolt.
(2) What a contact really means, and how varying between a bonded contact and frictional contact in a bolted connection analysis changes the meaning of the analysis.
(3) Non-linear analysis.....how to get an analysis to work based on mesh size, time step, rate of load application, solution methods, etc.
(4) How your mesh affects your answer. The necessity of a mesh convergence analysis.
(5) What are the pertinent ASME codes that apply to your situation? Do you need to consider fatigue?
(6) Calculating actual, bolt pretension.

Searching these forums and ansys.net will give you information on modeling bolted connections, and other related topics.

ANSYS theory guide and ANSYS structural guide have a plethora of information about convergence.

Make sure you understand the principles of bolted calculations, you should be able to get a good idea of this solution by hand....check references Roark's and Shigley's have solutions to similar problems. For the finite element method, try to start with a simpler problem, maybe a 2-D problem with only 2 bolts.

I probably missed a few things, its doable but it will take time.