{Kyle, Gerardo, Chandra}
We spent Tuesday maintenance morning hunting for a big air leak causing the pressure in vacuum system to bottom out at mid e-8 Torr range, and finally found what we were looking for. The leak measures 7.3e-5 Torr-L/s of He at an original 12" CFF blank flange next to NEG #1. This is a new leak since the post O2 vent and we are still unsure how to started. The N2 gas load from this range of leak even with all six ion pumps ON (15,000 l/s total) may be too big to ignore beam tube gas noise (according to M. Zucker), so a vertex vent may be required before O3. If all goes well with the fix, we could vent and be pumping back down in one day and open beam tube gate valves ~ one week later.
The procedure for leak testing included soft closing GV 1,2, valving in main turbo, backed by leak checker (background measuring 1e-8 Torr-L/s of He), valving out four IPs and three NEGs, and then spraying flanges for many seconds each with helium. We started spraying the (new) NEG adaptors (BT side of their isolation GVs) since we saw an increase in signal last time we leak checked near the RGA, which is very near this 12" blank. Gerardo noticed an increase in signal when Kyle was spraying around NEG 1 and asked him to spray that 12" blank where the He signal rose significantly. In the end we bagged the conflat joint with ameristat and sprayed until the He signal maxed out at 7.3e-5 Torr-L/s. We also sprayed the weld at the joint where the port tube mates with beam tube, and also the short tube seam, and did not see a rise in signal. The leak is likely coming from the copper gasket knife edge.
We valved the four IPs back into main volume but will leave NEGs valved out due to high N2 load. We scanned the RGA the entire morning to capture He rise. Text file attached - not - file size too big.
Attached is a picture taken back in mid December of 2017 of the flange which was found to be leaking today. In it, you can see that most of its flange fasteners had been removed at that time. Our best guess is that this was done in preparation to install a NEG pump at this location but, for whatever reason, was later aborted. We are unsure as to the chain of events but there was likely miscommunication involved as we would have never, knowingly, re-flanged this joint using the original non-vacuum, zinc-plated hardware (that had been provided by the OMC mode cleaner tube manufacture) that is still present today. Considering that the right hand and left hand were working independently, it is possible that the original, compressed, gasket may still be present and that is why we have such a big leak!
FRS 11544 ticket filed.