Particles disappear when activating background collisions even without cutoff

Hello,
I was testing the new background gas collisions feature (Molflow 2.9.22 for Windows) with a small mean free path (MFP= 0.1cm in a meter-scale structure with one pump on one side and an outgassing facet on the other), leading as expected to basically a brownian movement. However, I noticed that, despite me not activating the energy cutoff (and in fact putting the mass ratio to 0 to avoid the particles losing energy at all) nor the half-life, particles were disappearing (SUMABS<<SUMDES). Almost none of them reached the other side of the geometry from the outgassing face, leading to an exponential decrease in pressure from the outgassing facet rather than the linear decrease I expected.
After messing around some more, I noticed that particles disappeared after approximately 1.5million collisions with the background gas, no matter the MFP or mass of the molecules.

Edit : seemingly fixed in version 2.9.24, no more problem.

I was out of office, happy that 2.9.24 (with basically fixes to the collision model) fixed the issue, but report back if you find bugs! Thank you for your input, Marton