Bolt hole thickness and washer thickness 1

[FEAsolver]

i have a skid made of a plate with a certain thickness and modeled with shell elements (no solid element used in the model). The question is do I need to apply the thickness of the washer to the thickness of plate at bolted connections to improve the bearing strength of the skid at bolt holes? Currently the bolt holes edge fail under specified loading condition and no washer/nut/bolt head/bolt clamp force etc. take into consideration. Is it a common practice to add the washer thickness?

 

[rb1957]

it's common practice to add a washer under the head of a bolt. I don't think it's common practice to either model the washer (as a thickness pad near the hole) or to include the washer in the bearing calc.

 

[mechengdude]

Are you sure the "fail" isn't just an artifact of the modeling geometry and constraints? Typically when a bolt hole is restrained in the model you get unrealistic local stress. An experienced analyst can discard this... then again not knowning your exact case there may be a problem.

 

[FEAsolver]

Thanks for the comments, The loading is 5g in longitudinal direction, the high stress is not unrealistic as it distributes over a wide span with a peak at the hole edge elements. the bolt joint modeled using RBE3 element (if you are familiar with NASTRAN) and the element does not create artificial stiffness to be discarded. it is only a matter of the plate thickness (bearing strength) I believe the best approach will be to add a pad at holes to improve bearing strength of the plate and not to rely on the washer or any other lose component.

 

[rb1957]

increasing the dia of the bolt would reduce the bearing stress ...

or adding a Steel bush

 

[SWComposites]

1) a washer will add virtually nothing to the bearing strength

2) you should not be looking at peak stresses at the edge of the hole. Your material bearing strength allowable probably was calculated from test data as a P/dt stress. You should extract the fastener load from the FE model and calculate a P/dt bearing stress to compare with the material allowable.

 

[40818]

Well said SWComposites.
Never take the fastener hole edge stresses from a model as they are notoriously unrealistic. No matter how well modelled the mathematical processes involved within the FE leads to bad answers when comparing to reality

 

[darkwing888]

Here is an Machine Design article that might be useful in understanding when and when not to use peak stresses.

http://machinedesign.com/ContentItem/61166/HowlinearFEAhelpsinfatigueanalysis.aspx

"Sheesh. Some days it just doesn't pay to get out of bed."