XANSYS Message: 92852 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Rate item:

Subject: Re: [STRUC] Bolted joint contact representation
Author: Christopher Wright
Date: 2008-04-14 21:04:07

On Apr 14, 2008, at 2:43 PM, KChodimella_at_hendrickson-intl.com wrote:

> Just a general query about what contact do you specify between the
> nut and
> flange? Is it fair to assume that if in reality there is enough
> pretension
> and the joint is tight, we can use bonded contact?
Fact is that a properly made up bolted joint will not experience
relative movement between the parts in contact. That's the function
of the preload in the bolt--to clamp the two parts together with
sufficient load that friction prevents relative motion. A fairly
simple elastic analysis shown that the bolt load doesn't change until
the external loading nearly equals the preload; in this case the
bolted parts remain in contact without slipping. If slip were to
occur, the bolt would tend to unscrew, clamping force would be lost
and the bolt would fail in fatigue.

The easy way to model a bolted joint (or a riveted joint for that
matter) is to enforce contact between the mating parts by coupling at
each bolt location to determine the external load carried by each
bolt--then set the preload accordingly and verify that the load
required to resist slip is less than the frictional force produced by
the clamping load. The necessary design condition is no-slip, so that
should be one of your modeling assumptions. Modeling the bolts
explicitly is a waste of time.


Christopher Wright P.E. |"They couldn't hit an elephant at
chrisw_at_skypoint.com | this distance" (last words of Gen.
.......................................| John Sedgwick, Spotsylvania
1864)
http://www.skypoint.com/~chrisw/
Posts possibly associated with message #92852AuthorDateScore
92850[STRUC] Bolted joint contact representationKChodimella_at_Hendrickson-intl.com2008/04/14 
92851Re: [STRUC] Bolted joint contact representationParis Altidis2008/04/14 
92852Re: [STRUC] Bolted joint contact representationChristopher Wright2008/04/14