John, Bubba, Richard, Gerardo
16:44 Opened 10" valve. Turbo, LD and QDP80 remained on overnight. Background at 1.1 x 10-08 torr*L/sec.
***Left system be for background signal to drop.***
We returned to the center longitudinal seam weld (previously referred as butt weld) on the North side of BT. Divided the section into 3 equal sections, and bagged them. Using a rate of 1 L/s of He, for 60 seconds dwell, background at 5.2 x 10-09 torr*L/sec, leak hunting started,:
18:00 Section 1 was sprayed with He, and the other two section were purged with instrument air, we waited for 5 minutes to see a response. Background did not change.
18:06 Section 2 was sprayed with He, and once again the other two sections were purged with instrument air, no change on background.
18:10 John bags the purge/vent valve.
18:12 Section 3 was sprayed with He, while the other two sections were purged with instrument air.
18:15 Purge/vent valve is sprayed with He, and background value starts going up. Response was to quick??
18:22 section 3 is purged with instrument air.
18:30 observed max value for background 5.2 x 10-08 torr*L/sec.
19:05 Aux cart connected and pumping on closed purge/vent valve, no change noted on BT pressure.
19:12 left for L U N C H, while Turbo pump pumped He out.
===Activity after lunch===
21:28 Background signal at 5.1 x 10-09 torr*L/sec
21:40 sprayed He at a rate of 1 L/s for 60 seconds on the longitudinal weld seam section 3.
21:43 background signal starts changing, slooooow.
21:50 background signal at 2.1 x 10-08 torr*L/sec
21:50 started purging section 3 with instrument air.
21:51 background at 3.0 x 10-08 torr*L/sec
21:52 background at 4.0 x 10-08 torr*L/sec
21:54 background at 5.0 x 10-08 torr*L/sec
21:56 background at 7.5 x 10-08 torr*L/sec
21:58 background at 9.0 x 10-08 torr*L/sec
22:00 background at 1.0 x 10-07 torr*L/sec
22:03 background at 1.2 x 10-07 torr*L/sec
22:10 background at 1.2 x 10-07 torr*L/sec *
22:28 background at 8.0 x 10-08 torr*L/sec
22:38 Turned off and removed aux cart from purge/vent valve, valve was restored as found, except now space is vented.
22:47 background signal at 4.5 x 10-08 torr*L/sec
At this point we divided section 3 into two sides.
* East side which is away from the stiffener ring, this one was sprayed first with He, while the West side, close to the stiffener ring, was purged with air, we sprayed as before (1 L/s of He for 60 sec).
22:48 started spraying He on the East side.
23:03 background at 3.1 x 10-08 torr*L/sec, background continues to drop.
23:05 background at 2.9 x 10-08 torr*L/sec
23:05 switched hoses on the 2 sides, now we are going to spray He in the West side and purge with instrument air the East side.
23:05 background at 2.9 x 10-08 torr*L/sec
23:06 background at 2.8 x 10-08 torr*L/sec
23:07 start of He spraying at West side
23:10 background at 2.6 x 10-08 torr*L/sec
23:12 background at 2.8 x 10-08 torr*L/sec ***
23:14 background at 3.2 x 10-08 torr*L/sec
23:17 background at 4.1 x 10-08 torr*L/sec
23:19 background at 4.8 x 10-08 torr*L/sec
23:20 background at 5.2 x 10-08 torr*L/sec
23:21 background at 5.4 x 10-08 torr*L/sec
23:26 background at 5.9 x 10-08 torr*L/sec
23:27 closed 10" gate valve to isolate turbo pump, LD and QDP80 from vacuum system, but they remain on.
Conclusion,
We think that the leak is somewhere in the small West side of the center longitudinal seam weld, we will continue to test to pinpoint the leak source.
Typo in above - The helium flow rate was ~ 1 l/minute not 1 l/sec.
The suspected leak at the purge valve seems to have been a false positive due to the delayed response from the seam weld. After bagging the purge valve and chamber conflat as one we then removed the bag, removed the nw50 blank, and applied helium directly into the valve, into the test port of the valve body, and into the vacuum side conflat -each for 60 seconds. No response to this so we went back to the longitudinal seam weld.
The photo shows 2 bagged portions of the seam weld - one purged/inflated with air and the other with helium. We sprayed in both directions - ie swapped air and helium - no leak detector response was seen in one test while a strong signal was seen in the other. The prior day we saw a response from this area and in violation of my own rule I sprayed the triple conflat on the neighboring valve - again with 60 sec bursts of helium - no response.
The next step should be to try and evacuate the bagged portion of weld seam. This may prove to be a quicker and more definitive test than the helium testing, although the geometry is very tricky as the the weld is hidden by the vertical support structure and the stiffener stitch weld also crosses the seam weld.