You can remove all the worry about the roll-stamped portion of the tube by adding a regulator.
I know of one 2240 tube that was tested to 10,000 psi without failure, and it even had a roll-stamp.... What I don't know is how deep the roll stamp was, and they vary a LOT.... I personally feel comfortable using 3000 psi in a 22XX tube PROVIDED the roll-stamped area is not pressurized.... but that is me, I'm not recommending it to anyone else.... 1018 DOM tubing that is 0.875" OD x 0.065" wall has a safety margin of 3.1:1 at 3000 psi, and 1026 DOM tubing is 4:1.... However, at a minimum you need to use three high-tensile 8-32 SHCSs with the head taking the load into the tube wall, just as you would with a Disco.... and 10-32 is even better.... Do NOT use the stock Disco valve screws, or unrated ones from a hardware store, at those pressures....If you have a link to Mike's article, I would love to see it.... The test I am referring to was a Hydrotest by Walter Glover (Voltar1).... The only thing noticed was the 8-32 screw heads settling into the pockets in the tube until the load was equally shared, at about 7000 psi.... He stopped the test at 10,000 psi.... Lloyd has some excellent tests on Video right here on this Forum, in the PCP Gate.... Bob
I don't mean to beat a dead horse here but I think it's worth mentioning that it's not so much a matter of if a regulator fails, rather when it fails. If you keep a regulated rig long enough, it will fail. The seat on most if not all HPA regulators is a semi-rigid plastic that will eventually develop a leak. If you're lucky, the output pressure will just creep up a bit before there's enough "squish" to get it to seal. If it goes unnoticed, it will just continue to worsen until something gives. And hopefully that something is a burst disk and not a valve or a bottle.