2018/04/23 (Monday)

Previous Day

• Short was DIB3pA QET to chassis. This is Side2pB.
• Short was present at 77K, started to disappear at 100 K. Not present at room temperature.
• During tower assembly this channel also shorted to chassis but messing with the DIB connection fixed it.
• This channel showed significantly higher resistance on the bench than the last check after Run 62. All other channels were close to the old values.
• The wirebonds on this channel are quite long. In order to make this an operational channel it was bonded to two stadiums of the B-ring, jumping over the A-ring.
• In Run 62 tower assembly there were multiple times that channels shorted to chassis, though I'm not sure whether this particular one did.
• Since pushing the DIB in at room temperature caused a short, probably making the bond bow up and hit the crown, seems reasonable that something similar happens upon cooling, when the housing shrinks more than the crystal and pushes the DIB closer.
• Could try simply pushing the bonds down a bit, has worked in the past, as long as this doesn't break the bonds.
• The bonding for this channel was meant to make it a readable channel. Since we probably won't look at Side2 this run, could consider jump-bonding this to another Side2 channel, simpler and more reliable bond.

• UPDATE - it wasn't the detector, it was the side coax - see April 27

1. All mixture in dump, LN2 traps cold, condenser & still valves open (3He pump on - optional)
2. Open V9, V13b, V12b+speedivalve, V3. Wait for G1 to be close to G2.
3. Close V12b+speedivalve and V3, open V1
4. monitor G1

0945	Perform mixture throughput test at MC = 276 K using above procedure. Let G1 rise to 567 mbar before Step 3, then it falls to 41 mbar, then rises steadily. Throughput 7.1 mbar/min. A little worse than Run 63.
0935	Lower main bath
1005	IVC open, beginning disassembly. The SQUET on DIB4 has a very bent pin, Pin 42, SQUID D return. This is Side1pF, so it's necessary. This connection showed no problems in checkouts though.
1020	Fiddle with the SQUET on DIB3, which seems to have bad pins and sometimes shows bad connections in phonon-related lines. Can cause various channels to come and go, including pC, the one that we need. Maybe the solder connections of the pins are a little loose and can be improved?
1130	On the bench we found that on ZC-74, Pin 27 (SQUID B bias) was bad, moving it a little caused the resistance to jump to MΩ. Tried heating this pin to rewet the solder - after this it seemed more stable at first but a little stress caused it to go bad as before. Tried removing this pin completely and soldering in a new one - now it's completely open. Not good.
	Tested SQUET ZC-23 on bench. Same result as in the cold hardware page, SQUID A reads open but everything else OK. This SQUET has all four shunt resistances reduced to 6.7 mΩ. We'll try it as a replacement for ZC-74.
1333	Removed adaptor plate from tower
1335	Look at DIB3pA-to-chassis. No short. Try pushing in the side coax at the DIB - can make the short appear, kind of unstable.
1340	Remove side coaxes
1350	Remove detector. Inspect Side2 DIB3 (note this is the top, source-side). It's clear that the very long bond on Pin 2 is bowing up, clearly coming close to the source plate. Try pushing down on it to bend it flatter. Wound up breaking it off at the DIB. So, just pulled it off completely. The Pin 1 bond is still good - still shows continuity to DIB3pB.
1410	Reassemble tower and replace side coaxes.
1410	Tower electrical check:

Next Day