I need to model a beam tube pump plenum which has a beam tube with slots parallel to the beam, as well as 1.5 mm thick perforated plates covering the slots. The slots would not create a large number of facets, but the holes in the perforated plates make for a very large model (see images attached). I had hoped to analyze the real configuration, and then create a simplified equivalent model to go in the accelerator ring model (the pump repeats 148 times). Can you provide some guidance as to how to approach this model? Thanks for any help- Bill Crahen

Hello Bill,

this is a typical problem found in modelling accelerators' vacuum systems.

The way I usually tackle it is the following. You take ONE slot, calculate its transmission probability, and then assign an opacity to the facet which is covered with slots (eventually more than one!) which gives the same overall probability for molecules to pass through the many slots and reach the volume on the other side, the pumping plenum.

In doing so, obviously, you introduce an approximation, since the real path and angular distribution of the "equivalent" molecules will not be the same as the real molecules. So, if this distribution is important to you, then you can't use this trick/approximation, and you must run the full geometry.

I have carried out, in the past, some benchmarking, in order to quantify the difference in equivalent pumping speed (as calculated along the axis of the chamber, where the beam is) for the two configurations, full geometry and "equivalent opacity". I did it for the pumping ports of the ESRF vacuum systems... unfortunately now I can't find the short memo I wrote... but the difference was a few percent, not more than that.

Hope it helps, if not contact me again, I'll be happy to provide an example.

Cheers.