[xuegy]

Hello All,

I'm trying to implement a partial slip boundary condition that takes a fixed slip length as the input.
I modified JohnsonJacksonParticleSlip, which was based on partialSlip. The key calculation step is

this->valueFraction() = 1.0/(1.0+slipLength_*patch().deltaCoeffs());

which is the same as this thread How to access gradient near the boundary for Navier slip bc?
The original partialSlip BC worked fine. It just doesn't support non-uniform meshes with a fixed fraction value. I can compile this new BC, but it will blow up in 1 or 2 timesteps.
This is very confusing. 0<1/(1+non_negative_number*positive_number)<=1 always holds and I can't understand why it's blowing up.
I've attached my source code. It compiles with OF v2012. I would appreciate any kind of suggestion.

[xuegy]

Seems like there's nothing wrong with the BC.
I mixed up 32bit solver with 64bit mesh and it blew up.