Preload change with temp help

[DanStro]

This seems like an easy problem but for some reason I have a mental block.

I am trying to model the change in a preloaded assembly as a function of temperature.

The basic situation is this:
A threaded retaining ring is torqued down to get a predefined preload, at room temperature. Then the temperature of the assembly is changed. I want to see how that temperature change affects the preload. What I am struggling with is how to lock the retainer to a reference point or surface. If I don't do that Abaqus will just keep the original load value. I've confirmed this by looking at the reaction force vs. temp and that is a constant value. I don't believe this is what is happening in reality so I am trying to 'fix' it

A simplified problem is this (see image below of axisymmetric case):
I have a fixed housing with a ball bearing placed in it. The ball is held in place using a threaded retainer that contacts the spherical surface. I want to look at how the initial preload changes as a function of temperature.

What I've tried and thought about trying so far:
I have a static step that applies the preload from the retainer then another step that applies the temperature.

What is happening is that when the temperature is changing, and the components are expanding/contracting, Abq is adjusting the retainer position to keep the original preload. That isn't what I want, and doesn't seem realistic. What I'd like is to link the position of the retainer to a point on the housing that in the real part will be part of the thread that the retainer is using. The retainer then would move with that point as the length of the housing is changing.

My first thought was to do a coupling but you can't activate/deactivate for specific steps, the same thing goes for a tie constraint. It doesn't seem like connectors is the right way to go but maybe that is because I just don't have any experience with them.

Any suggestions?

Thanks,
Dan

 

[FEA way]

Is preload deactivated in the second step (the one in which temperature is applied) ?

In order to link movement of the retainer to the selected point you could use equation constraint. Connectors could work too.

 

[DanStro]

Yes, the preload is deactivated before the temperature step. I've considered equations too but I need to brush up on them before I can think of how it will be done.

 

[Mustaine3]

Activate a contact in the second step, that has properties 'No Separation' and Friction=Rough. Make sure that the part is then moved a little bit sideways, so the contact is really closed. Now the part is glued at the contact location.