Leaking O-rings explained

I spent all day Saturday working on the o-ring seal problem with my tanks.
I'm 90% sure I know what the problem is.
After trying everything under the sun with respect to o-rings sizes grove depths etc...

I did an experiment, I built a tapered hollow aluminum plug and an aluminum ring that goes on the outside. I heated the PETG tubes and slid them over this tapered plug. I then filled the tube and a plug with Ln2 and let them cool. (To make them shrink)

I boiled the ring that goes over the outside in water. (To make it expand)

I then drove the heated ring over the outside of the assembly.

This explanation sounds simple, but I machined the plug and ring diameter 10 or 15 times to get all the tolerances right.

I then let every thing return to room temperature.

I filled it with water and it still leaked.

Examining the plastic tubes under a magnifying glass I see that the extruded tubes have fine ridges or groves and these seem to make little capillary tubes that lets water seep through.

I'm now once again looking for clear, smooth plastic tubes.