2019/03/26 (Tuesday)

Previous Day

Setup	DIB1/3/4 → 3.2 MHz filters 2/3/1 → cable 3/1/2 → gender changer 3/?/2 → HVConn 1/3/2 → HV 1/2 → DCRCX(4)/DCRC50(3) DCRC33(2) → 1kV power supply → HV input Lead poly shield is erected, N/E/W walls complete, no poly or lead on S wall. PuBe source present in paraffin barrel behind lead stack.
0920	Tower/MC/Coldplate = 43.0/31.5/200 mK, Still = 1.85±0.15 K / 0.160 mbar, G1/G2 = 265/234 mbar, 1K Pot = 1.79 K / 15 mbar. MB = 32.8%
0935	Fill LN2 cold trap bath
1015	Notice that MIDAS crashed again, apparently during the end of the sequence last night during the wait after flash. Again mhttpd failed. Did another restart without killing anything, mhttpd is back and all looks normal.
1040	Generate some sequences to test leakage. Currently we are not purging the Ebox. We'll start with that, then try with Ebox purge, then if that's still bad try disconnecting various things.
1047	Start sequence to dial up HV from 0 to 400V at 1 V/s.
1057	at 10:57:10, HV hits 300 V
1100	at 11:00:30 HV hits 398.8 V, max

from around 143 V to 224 V, Vtest was mostly flat at around 80 mV. Then it starts to rise again.
At 342 V, Vtest had risen to about 96 mV, then starts rising much faster
At the maximum, 400 V, Vtest reached around 165 mV. Then as we left it at that voltage Vtest wandered a bit, between 150 and 180 mV.

1112	Start sequence to dial HV back to 0 V
1117	Start purging Ebox with N2 from the LN2 Dewar vent. Opened it just enough to feel some pressure on my thumb when I held it over the port.
1137	At 11:37:20, start sequence to dial up HV from 0 to 400V at 1 V/s.
1147	At 11:47:03, HV hits 289.9 V
1150	Just as sequence finishes at 398.8 V, leakage jumps from 150 mV to 800 mV. At this same time, tower temperature starts rising suddenly. Goes from 43 mK to 47 mK in 3 min
1151	Start sequence to dial HV back to 0 V.
1156	Leakage is decreasing linearly from 800 mV, currently 450 mV. This is much higher than the max reached in the first voltage scan today. Tower temp = 48.8 mK. Still and circulation pressures are all elevated. G1/G2 = 288/256 mbar, Still=0.190 mbar
1205	HV = 0 V
	Before this severe increase in Vtest and Tower temperature, Vtest versus HV was behaving about the same as in the previous test, so purging the Ebox didn't seem to matter much.
1206	Take a look at detector channels on pulse viewer. Some are railed, but those that aren't still show pulses. No obvious oscillatory modes, but jagged breakdown noise present.
1215	Vtest is stable but Tower temperature is still elevated, flattish at 53 mK.
1215	Turn off DAQ programs. Then power down DCRC50(3). After half a minute Tower temperature starts falling.
1245	Tower = 42.5 mK, colder than it was this morning even.
1247	Begin LHe transfer @ MB = 27.6%
1334	Cease LHe transfer @ MB = 93.1%
1340	Tower stable at 42.7 mK
1342	Power down DCRC4. Disable DCRCs 3,4 in ODB. Unplug detector cables and pi filters from ebox.
1345	We want to locate the HV leakage location. Now we've got HV connected to the HV board and board connected to cables with DCRCs 3,4 connected, but powered down. DCRC2 is still controlling the HV.
1348	Start dcrc_driver01.
1349	At 13:49:15 start sequence to dial up HV from 0 to 400V at 1 V/s.
1352	At 13:52:39, HV hits 100 V.
1356	At 13:56:00, HV hits 200 V.
1400	Tower has been falling since powering down DCRC4 and unplugging the cables. Now Tower = 40.7 mK.
1402	At 14:02:37, HV hits 398.8 V, max. Didn't see the leakage rate reach as high a level as with the cables plugged in.

from around 100 V to 185 V, Vtest was mostly flat at around 68 mV. Then it starts to rise again.
At the maximum, 400 V, Vtest reached around 93 mV. Then as we left it at that voltage fell exponentially to 57 mV by 1515.
Vtest did not wander randomly like in the first test.

Suppose we set a basic criterion that Velectrode > 90% HV. Then at HV = 300V we want Vtest < 75 mV, and at 400V we want Vtest < 100 mV.

1515	Start sequence to dial HV back to 0 V
1730	Fill LN2 cold trap full. MB = 90.2%

Next Day