Can anyone actually explain how any ambient pressure contaminant can get into otherwise sealed, pressurised systems?
Theory alert could be bollocks.
Surface tentsion allows fluids to do some very strange things when in capillarys liquified gases are famous for it. They will allow fluid to travel uphill so to speak.
My theory is that they have a slight leak and there is Oxygen fluid forming, if it manges to bridge the hole the surface tension will start doing its work and cappilary action will work pulling it up stream. Now because the fluid will be travelling into a cooler temp it will aid heat transfer to the outside it will cause even more oxygen to become liquid and supply the capilary action. Once its inside if there is any sort of height difference between top and bottom of the seal you will have started a syphon. Liquid oxygen being relatively heavy compared to He will have no probs overcoming the 4 psi differential. And thus you have an up hill (in pressure differential terms) leak. Look on the bright side though it means your not loosing highly expensive He through a seal leak.