Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations pierreick on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Singularity at Pin Connection 1

Status
Not open for further replies.

Andrew Brockbank

Mechanical
Feb 28, 2020
6
Hi All,

I am currently designing a drone frame, and performing FEA on a quarter model (see below).

Full_Model_yaaycb.jpg


The arm has a 10N load acting upwards at the end, while the top surface of the main body is fixed. This is to simulate the forces occurring when the drone accelerates upwards at maximum acceleration.

The arms are connected to the main body via bolts. For simplification these have been modelled as pin connections. See below.

Zoom_in_of_Pin_Connection_irbrou.jpg


I would like to get an estimate as to how much stress occurs at the bolted connection, to see if the design is suitable. On performing a mesh sensitivity analysis I have found that there is a singularity at the edge of the round. See below maximum mesh refinement and singularity.

Mesh_around_pin_nl3vct.jpg
Signularity_Zoom_Out_z0vw6x.jpg
Signularity_Zoom_In_hrxxjt.jpg


I believe this singularity is occurring as the hole goes from being completely in connection with the pin, to being not in connection at all, essentially creating a sharp edge.

The linear analysis is being run on SOLIDWORKS simulation. I have tried the same set up on Fusion360, getting the same results. Otherwise displacement looks good, and max stress on the arm and max displacement converged many elements earlier, while stress at this bolt hole continues to increase ~linearly.

Does anyone have any suggestions of a better way to estimate the stresses occurring at the bolted connection between the arm and main body?

Many thanks,

Andrew
 
Replies continue below

Recommended for you

The first thing to do is carry out nonlinear analysis (with plasticity included).
 
Better way? Yes. Do not model fastener holes in FEMs. Its a complete waste of time and just produces unusable results. Model the fasteners with springs or beams, extract the fastener loads, and calculate the bearing stress by hand as P/dt and compare to material bearing allowable strength, etc .......
 
FEA Way, thanks for that. The non-linear simulation looks reasonable around the pins.

NEW_b9wkz9.jpg


I am now getting a stress concentration in the corner of the upper face. This, again, is where the fixed constraint ends abruptly, essentially creating another sharp edge at the boundary.

Any further suggestions on how to remove this would be greatly appreciated!

SWComposites, I agree that modelling fastener holes isn't ideal but it can be useful in determining optimal fixing locations to minimise stress concentrations at these fixings. When running several simulations with varying fixing locations I think it should help me save a lot of time in determining optimal fastening locations (before completing a validation more similar to the one you have suggested).

Many thanks both,

Andrew
 
Just ignore those stress concentrations; the are not real or accurate.

You are wasting a lot of time and effort with that ridiculously fine mesh at the fastener holes. There is no need for that to determine fastener locations.
 
SWComposites

Sorry I'm fairly new to FEA, could you explain what you mean by modelling the fasteners as springs etc to hand calculate bearings stresses?

Many thanks
 
I think you've found a deviation within the mathematical definition of the surfaces involved. I've seen this several times, where a different way of creating the geometry works better. But I agree with SWC ... I think you're chasing phantoms. What would it take to build something and test it ? What's the scale of the first pic ? If you're looking at the arm, surely hand calcs will get you 90% of the way ?

another day in paradise, or is paradise one day closer ?
 
To add to SWComposites earlier comment.

There are numerous structural codes which give equations to check the specifics you are interested in. By using the modelling procedures presented by SWComposites you can perform hand calculations for certain specifics where the FEA would not be accurate. It is a good way to justify your design as FEA is open to interpretation...
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor