Stress on pin and bracket.

[wimpog]

I am analysing a pin which rests through two holes in a bracket. The holes are .02 in wide and are slightly larger than the pin. The pin is subjected to a downward force in the middle (between the holes). I am intesrested in the stress where the pin meets the hole in both parts.

The problem I am having is that the more I refine the mesh, the higher the stress is, and it seems to be diverging at that point. Is there a way to get an accurate result for this type of problem?

I have "no penestration" contacts with surface to surface checked on all of the touching faces.

 

[corus]

I'd imagine that as the pin is smaller than the hole then the pin will make a point contact with the inside of the hole. This will tend to infinite stress as force/area tends to infinity as area of contact tends to zero.

corus

 

[wimpog]

yeah, this is what I suspected was happening. Does anyone know a way around that, or how to realistically model such a situation?

 

[corus]

You could try including a pressure distribution instead of the pin contact. In your situation you have effectively modelled the contact as a point load. If the pin were the same size as the hole then you'd have a sinosoidal pressure distribution. Somewhere inbetween the two would give a better stress distribution, but in my view the best way is to ignore the high stress at the pin contact when assessing the overall stress distribution. Localised yielding due to a point load won't be detrimental to the overall structyral integrity of the component.

corus

 

[wimpog]

I see, thank you very much for the help, that makes a lot of sense. My goal is to determine the breaking point of the pin, I can measure the stress due to bending in the middle of the pin, which I assume is pretty accurate, but I wonder if there is anyway to determine the actual stress on the pin at the junction. Perhaps this is just a limitation with SWS.

 

[kellnerp]

Is the point where the stress is diverging a corner? It likely is. What you need to do is create a small round on that corner and put a fine mesh on it, not the pin. Use a radius that is likely in real life. You are seeing the effect of a singularity at a corner.

Are you using node to node gaps or surface to surface?

TOP
CSWP
BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."

 

[wimpog]

well, a fillet would remove the point stress that is down the length of the pin, but I think the circle contacting a circle would still be a single line of contact in the other direction.

I use surface to surface, but what is the difference?

 

[corus]

Local yielding of course would give you a larger contact area and hence the real stress distribution could, in theory be calculated. However, I would imagine that the mesh required to capture that stress would make the model far too big. You'd be better looking at a theoretical solution, along the lines of herzian contact, if you wanted to know the stress at that point.

If you're only looking at the pin then why not do a simple hand calculation and treat the pin as a beam on simple supports where it makes contact with the outer lug(s). I suspect that a combination of bending and shear stresses will cause failure.

corus

 

[kellnerp]

You won't have a problem with line to line contact with a cylinder in a cylinder. Deflection of the material almost instantly transforms it to strip to strip. But at the end of the hole where the inner cylinder extends into space there is a singularity if the corner of the hole is sharp. Round that edge.


TOP
CSWP
BSSE

http://www.engtran.com

http://www.niswug.org

"Node news is good news."