2019/05/09 (Thursday)

Previous Day

Status	Warming after finding a fridge blockage yesterday - third one.
Setup	DIB1/3/4 → 3.2 MHz filters 2/3/1 → cable 3/1/2
0905	MC(Cernox) = 238 K
1055	Cryostat vacuum jacket = 25 mbar.

Plan: leak check the IGH

Suspect that air is leaking into system, possibly during cleanings. Water in the air causes ice blockage. Cryocleanings are not getting enough out because of the leakage in.
Leak check, where we can, the volumes that mixture occupies during cleanings.
Where we can't leak check, without a lot of extra work sequestering the mixture:
- pump volume, inlet and outlet
- dump
Where it's inconvenient to leak check:
- inside the IGH
Where we plan to leak check:
- The Main Pumping Line volume, labeled “P” in the IGH schematic diagram. This is the volume read by gauge P1.
  - extends from above V6 by the pump, to the Main Pumping Line port on the IGH, and all the way from the high through-hole in the wall to the Still Gate Valve. There are 16 KF o-ring flanges we can check.
Other volumes we could check, lower priority:
- The LN2 cold traps. These are certainly involved in cleanings and they are moved around a lot. One reason they seem less likely leak candidates is that the amount of contamination removed doesn’t seem to differ between the two.

1130	Hook Alcatel leak detector to a valved port by V6 by the pump. This is the Still pumping line, volume “P” on the IGH schematic (the volume read by gauge P1).
1135	Pump out “P” with scroll pump. Hook up N2 gas to IGH vent port. This allows us to flush with N2 by alternately closing V2a (to stop pumping) then opening the vent speedivalve for a few minutes, then pump out again.
1145	10 minutes after the first flush, P1/P2 = 2.6/0.04 mbar. P1 still falling.
1155	20 minutes after the first flush, P1/P2 = 0.65/0.019 mbar.
1155	Try opening “P” to the leak detector. Signal goes off scale. This could take a while.
1155	Do a second N2 flush. This time P1 only went up to 300 mbar max as we let in N2 - last time it pegged at 1000 mbar. Don't know why. Pump it back out with scroll pump.
1215	20 minutes after the second flush, P1/P2 = 0.50/0.018 mbar.
1220	Do a third N2 flush. P1 rises to 280 mbar. Pump it back out with scroll pump.
1240	20 minutes after the third flush, P1/P2 = 0.65/0.019 mbar.
1240	Try opening “P” to the leak detector. Signal = 7e-5 mbarL/s. Better.
1240	Do a fourth N2 flush. P1 rises to 280 mbar. Pump it back out with scroll pump.
1310	30 minutes after the flush, P1/P2 = 0.38/0.015 mbar.
1310	Try opening “P” to the leak detector. Signal = 4e-6 mbarL/s. Better.
1315	Do a fifth N2 flush. P1 rises to 280 mbar. Pump it back out with scroll pump.
1340	25 minutes after the flush, P1/P2 = 0.036/0.012 mbar. (Actually these low pressures were achieved by valving in the leak detector.)
1345	Do a sixth N2 flush. P1 rises to 240 mbar. Pump it back out with scroll pump.
1420	P1/P2 = 0.019/0.012 mbar with the help of the leak detector.
1420	Do a seventh N2 flush. P1 rises to 300 mbar. Pump it back out with scroll pump.

1435	MC = 255.3 K. Warming time constant is 14.5 hours. Should reach 10 C by early tomorrow morning.

1445	Try leak checking o-ring connections in the “P” volume. Leak detector background around 2e-6 mbarL/s and settling. Squirt at o-ring connections in the pump room - no increase in signal observed.
1510	Leak check o-rings in the lab room: from the high through-hole in the wall to the Still Gate Valve, and also the “Main Pumping Line” port on the back of the IGH. Found a leak! Significant increase - 1.5e-6 to 4e-6 - and repeatable. This is at the end of the copper tube coming through the wall on the lab side. There are three KF connections here: steel KF50 (I think), then a 50→40 size adapter, then plastic KF-40, then a short elbow, then steel KF-40, then the flexible hose leading to the fridge. Leak seems to be at the plastic flange, though it's hard to rule out the others.
1520	Fill “P” volume with N2 again. Then remove all three connections mentioned above. Wipe off flange surfaces with methanol. (Although I realize now that I did not do this to the srufaces at the end of the copper pipe or the flexible hose.) Replace all three o-rings with new ones found in the toolkit drawer. Apply a little vacuum grease to each o-ring. Reattach all connections tightly. Note: P1 read 240 mbar when it was exposed to room pressure. It was reading 300 mbar when I filled it, and it was definitely above 1 atm.
1535	Start pumping out again with scroll pump.
1625	Valve in leak detector again. Do a squirt test at the flanges with the replaced o-rings: no increase in signal seen. :) Background was 1.2e-6 mbarL/s.

Next Day