Material Table Generation
 

I have defined a material with table generation and I´m sure the pressure and temperature minimum and maximum values are accurate.

However in some iterations and some points the values of pressure or temperature are out of bounds, just for a little amount of time.

Is it better to let the solver do clipping or should I activate pressure and temperature extrapolation? What is the difference between these too?

If it clips early on but as it converges it goes back inside your limits then I would not worry about it, it should be fine.

I will change some variables in my simulation in order to study the influence they have on results, if then the clipping is continous should I opt the extrapolation?

If the final result clips then you can extrapolate, but a better approach would be to extend your table to cover the range your simulation is covering.