Increased noise on Varian TCD

[Johnny Rod]

I suspect I'm missing something obvious so advice please! We have a Varian 430 with 2x columns feeding a TCD (one to each side). The baseline P2P noise has increased fourfold since I had a fiddle with it. One column (rarely used) wasn't producing any peaks (turned out to be the gas sampling valve actuator wasn't attached) and in the process I swapped the column tail ends over to rule out the TCD as the problem. I refitted them in their original sides in the TCD and now we have more noise. I've checked the fit and they seem fine, and a helium leak probe doesn't show any leaks (anywhere). Column flows are as they were before (these are both 1/8" packed columns 2m/7ft long), rear is 25ml/min and front is 15ml/min (not very balanced but the same as has been operating for years). Everything points at something I've done but I'm not sure what to do next. Any thoughts? How easy it is to damage these things?

[Johnny Rod]

Anyone got any ideas?

[Johnny Rod]

Fixed! There were three column leaks - two minor head-end ones and a larger tail-end one. There appears to be a knack to using our helium detector, so my colleague found the leaks. Broke my own problem-solving rules of not missing the obvious and re-checking thigns when you hit a dead end.