Applying fixture to pin in double shear:

[SudorracMechEng]

A cylindrical jackstand, with a cylinder adjustable rod inside - held to place by a pin.
I am treating this as double shear, however i am having trouble applying the appropriate fixtures.

I have applied the force on either side to only the top portion. When i apply the fixture restaint on the opposing side to - half - it yields Von mises stress of 1044 MPA... the yielding of the steel pin is 714MPA..
However, when i place fixture around entire pin, the max von mises becomes 680MPa

Two assumptions that affect my results:
1 - the pin is tight fitting, in which restraint can be fixed around whole surface - and shear is the only issue

2- "loose fitting", which means pin bending will be taken into consideration, and will yield higher stress.

I am assuming option 2.

any help guys?

 

[kellnerp]

Well a picture would be helpful. You haven't specified the problem very clearly.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

http://www.linkedin.com/in/engineeringtransport

Phenom IIx6 1100T = 8GB = FX1400 = XP64SP2 = SW2009SP3
"Node news is good news."

 

[SudorracMechEng]

Hi sorry for the lack of specification.
Here is a screen shot of the foundation.
Ill upload another photo of how i have placed the force, and fixture.

 

[SudorracMechEng]

here is the fixture, and load.

 

[kellnerp]

Results in this type of analysis are highly geometry and boundary condition dependent.

The restraints are not correct:
You must have a sliding condition where the two faces, the outside of the pin and the inside of the ase hole come into contact. This implies a radial constrain, not a fixed constraint.
The pin will lift off the base away from the interface edge due to pin bending. The pin is simply supported in both holes.
The distribution of loads on the pin is load dependent.

The loads are not correct:
You must have a varying pressure on the other face. Typically a cosine distribution is used.

Modeling this with contact will greatly change your results.
Varying the gap between the base and vertical post will greatly change your results.
Interpreting the results is tricky. You will get a high stress at the edge between the base and post. This high stress is not indicative of failure for four reasons.

1) There is a singularity in that area.
2) Failure happens when the pin yield through it's thickness, not in a localized area.
3) The cylinder/base area on the pin is an area of high stress gradient. That gradient is highly affected by the fit of the cylinder and the base. Small differences (0.001 inch) will cause large changes in stress.
4) The modulus of the pin and supporting material greatly affects results because compliance of the support is a very significant effect. Compliance of the support will lower stresses in the pin.

What industry is this analysis for? Some industries have standards for analysis of this type of analysis.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

http://www.linkedin.com/in/engineeringtransport

Phenom IIx6 1100T = 8GB = FX1400 = XP64SP2 = SW2009SP3
"Node news is good news."

 

[SudorracMechEng]

I am using solidworks 2011 - however, i do not have the cosine distribution option - so i selected non-uniform distribution, and under equation coefficients i selected (-1) under x^2 and selected a path from the inner cylinder to the outer cylinder
- this was my attempt at trying to replicate cosine distribution, however i failed miserably.

For analysis, i used assembled jackstand - assuming this is correct.

Under Mates: I made inner and outer cylinder concentric.
- I wasnt sure what mates to do with the pin, as i didnt want to limit its range of bending motion.

Under Fixtures: I applied a FIXED surface on the bottom of the base plate of the jack stand
- i applied a "Slider" fixture on the surface of pin and the cylinder holes
- once again, i think this is totally incorrect.

However, the actual results reflected by the Von Mises stress, seem to be only slightly under what i had expected. Coincidence, maybe.

If you would like photos of my results, please let me know.

This is technically in the bauxite mining industry, but it is purely just an "in - company" design - I was just trying to work out how safe the current jackstands actually were, however i clearly am having a difficult time applying the right restraints and loads.

I am much more sound using hand calculations, however i do want something modelled on FEA that will somewhat resemble the actual situation.

So i model this as an assembly, oppose to just the pin component. I was going to model the components seperately.
I modelled the inner and outer cylinders to look at the max permissible bearing stress.

For the

 

[kellnerp]

You picked a very difficult problem, strongly subject to as built geometry. Hand calcs are very well developed for this sort of thing. In a mining environment you probably want a FOS of 5 anyway so getting stress to within the last Pascal probably isn't a good way to spend time.

As far as the model, the other comment is that you don't really need all the detail. St. Venant's principal comes into play making a lot of the structure shown unnecessary unless there is some load path that impinges on the pin/hole area. Symmetry can be used to make the model much smaller and still get good answers.

Still the boundary conditions need a lot of work. I'm looking for an example problem I did some years ago do illustrate this.

TOP
CSWP, BSSE

http://www.engtran.com

http://www.niswug.org

http://www.linkedin.com/in/engineeringtransport

Phenom IIx6 1100T = 8GB = FX1400 = XP64SP2 = SW2009SP3
"Node news is good news."

 

[SudorracMechEng]

Thankyou for your help.

If you find that example, please attach for my perusal - it would be helpful as a reference.

 

[SudorracMechEng]

Hi Kellnerp, did you find that example problem?

 

[Metrics]

If you reduce it to a 1/4 model with 3 contacting parts, the restraints become easy, but you still need to choose the worse case for the fit up of the parts.