Tuesday May 28, 2013 at 5:23pm
It is surprising what you discover when you browse the SOLIDWORKS Knowledge Base!
Today, whilst searching for an answer to a customer problem, I came across a useful article that clarified for me how friction works in Simulation. I had sometimes wondered why I didn't always get the friction forces I was expecting - now I know why! Interested? I know it is a bit geeky but if you are using friction in a Static study you really need to know this!
We all know that a simple method of calculating friction between 2 surfaces is to use the equation:
Friction = {coefficient of friction} x {normal force}
where the coefficient is a value between 0 and 1
However, if you use this you may not get the same result as listed in a simulation study - unless the parts are actually sliding. (If you have not seen how to get the friction results on parts that have a 'No Pen' contact, all you need to do is to select 'List Result Force' from the Results folder and select the 'Contact/Friction Force' button).
The answser to this is because Simulation will use a friction force up to a maximum of the friction force calculated using the coefficient. Assuming there is no sliding the amount of friction listed in the results will be just enough to enable equilibrium.
This can be tested with a simple assy as shown below. The set up allows sliding in one direction; gravity is set at 10m/s^2 and a pull force is set to 1N. There is a No Pen contact between the plates with a friction coefficient of 0.1. The lower plate is fixed on its base. The mass of the upper plate is 1.56kg so the weight (i.e. the normal force) is 15.6N.
The results show that the normal force is exactly 15.6N in Y (vertical direction) as expected ....
But the friction force is NOT 0.1 x 15.6 = 1.56N (i.e. the maximum available) but is 1N (to 2 dec places) in the horizontal direction. The 1N is the amount of friction required to keep the upper plate from sliding i.e. the amount of friction required to balance the 1N pull load.
If the pull force was increased, the friction force would also increase - up to a max of 1.56N. If the pull was greater than 1.56N then sliding would occur, there would be rigid body movement and the solver would fail - I've tried it.
Having said all that, you always need to be cautious with friction results - especially if the surfaces are not capable of sliding smoothly. In reality friction is a complex phenomenon and takes different values depending on whether the surfaces are sliding (dynamic friction) or static (static friction). The size of surface can also have an influence and rough surfaces or materials like aluminium can suffer from 'pick up' which causes gouging.
Andy Fulcher
Technical Manager
Solid Solutions Management Ltd