Bearing Stress and sliding

[KevinNZ]

We have a steel beam resting on another section of steel (say a pipe or rod). During temperature changes the beam slides around on the other section.

How would we calculate the allowable down force on the beam to avoid high bearing loads the would prevent the two section sliding? We can tolerance some deformation or harding and the two surfaces wearing out to a flat surface between them. What we don't want is a dent in the sections that produces a high resistance to sliding.

AISC A360-16 J7 has a formula for a rolling contact, but we suspect this has a high safety factor to prevent dents in the rolling surface. Bearing stress calculations of static applications (eg bolting) give much higher loads but we think these do allow for some location deformation.

 

[CANPRO]

To get an accurate answer you would have to model that in a way to capture how the pipe deformed while it rolled. Depending on the scale of your application, I would try to physically test that before relying on analysis.

Have you tried to mitigate the problem of the pipe deforming? What's the application? Maybe you could use a solid bar or fill your hollow pipe with concrete and cap it.

 

[KevinNZ]

Canpro

Thanks for the reply. We are not concerned with the pipe bending as the beam slides on it. We can assume the pipe is solid rod. It is the deformation at the local point of contact we are trying to get a handle on.

 

[WARose]

If it's a situation where something just has to slide to avoid catastrophe.....why not go with a Teflon slide plate system? Pretty much all the people who manufacture them guarantee not just the friction, but that friction under a certain pressure. I've used them a bunch of times for equipment support.

Either that or look at some formulas with regards to Hertz's contact law. (If you are worried about local deformation.)

 

[KevinNZ]

WaRose thanks

Hertz's contact law can give local stresses but how much is too much?

It would be good if there was formula like AISC A360-16 J7 for sliding.

 

[WARose]

Hertz's contact law can give local stresses but how much is too much?

In your case, it sounds like you are more worried about dents....there are some formulas out there that will allow you to figure how much indentation there will be. 'Roark's Formulas for Stress & Strain' is a good resource for that. There is also a discussion on stresses as well.

 

[CANPRO]

I think there are two things you need to account for - the roller could flatten out a bit making it difficult to roll, and the supporting surface could depress forcing the roller to move "up hill" a little bit. I believe there are closed form solutions for determining the contact stress, contact area, and deformations. But I think your problem is more difficult than that as the static condition is just the beginning of the problem. As your roller moves, the deformed shape of the pipe and supporting surface is going to change. Changing the shape of the two objects requires some amount of energy, which isn't as straight forward as determining static contact stress and area.

 

[wannabeSE]

I assume the pipe or rod is not a roller, rather fixed. Isn't it just a simple matter of calculating the force caused by thermal expansion and contraction of the beam. Then use the coefficient of friction between the beam and pipe to calculate the required force to prevent movement. Then add whatever factor of safety is appropriate for the application.

 

[WARose]

[blue] (CANPRO)[/blue]

I think there are two things you need to account for - the roller could flatten out a bit making it difficult to roll, and the supporting surface could depress forcing the roller to move "up hill" a little bit.

That's a good point because one of the things I left hanging with my last post is: how much of a dent is too much? I guess someone could take a shot with the "up hill" approach. Still another approach I've seen people take is that if they do not indent the surface much more than the normal surface roughness....there shouldn't be a issue. (As far as increasing the coefficient of friction/difficulty of moving.) For milled structural steel that's somewhere in the neighborhood of about 1 mil (0.001 inches) or less. Not sure how realistic that is (to achieve) though.