I see what you are saying,,, but how dies the mix not get stirred to an even blend when you top off with at reasonable rate being injected through the dip tube. that should create a tremendous stirring effect on its own.. I cant say it doesnt happen because there are those , including you , that say it has happened to them. You i F i UNDERSTAND YOU CORRECTLY YOU ARE SAYING THAT ONCE MIXED IT STAYS MIXED. i believe that 100%. Yes I think that is my hard spot on this. If that is the issue, Is it perhaps a matter of having too many tanks on the compressor so as to reduce the fill rate on each tank that prevents a good stirring form the incoming gas? I have always wanted to know and witness how the stratification happens to avoid it. So far I have not been able to make it happen filling one tank at a time with a 6 cuft compressor. I can visualize the stirring effect reducing as the tank pressure increases. I have witnessed how % changes in a fill process when I say start with an empty tank. I put in 25% (750psi) o2 and than immediately top off with air vs filling 25% psi o2 , letting it cool which makes the O2 psi drop to maybe 22% psi , retopping off with O2 letting it cool and then topping off with air to a final 3k. I find what appears to be the coeffecient of heat expansion of O2 is pretty great. A 3000psi tank to make a final 40% takes about 850psi to start and it cools to 750 +/-. One day I hope to to see this effect here at home.I think you might be a bit confused here. The issue with stratification occurs when blending. I have found that once the mix homogenizes, it stays that way. It does not separate over time. I played hell getting 4 cylinders of 50% to be 50%. I kept getting odd readings. I laid the down and tested all 4 a full 24 hours later and they were all dead nuts.
The dives these were to used for got canceled. They sat in the compressor room for five months before they were finally used. When analyzed, they were still dead nuts.