UMN CDMS Wiki archive

start › cdms › k100 › run_diary › 20170929

2017/09/29 (Friday)

0920	Tower/MC/Coldplate = 73.0/56.5/370 mK, Still = 1.4 K / 0.20 mbar, G1/G2 = 217/194 mbar, 1K Pot = 1.66 K / 40 mbar. MB = 28.6%
0925	Fill LN2 cold trap bath.
1313	DCRC1 is railed, but DCRC2 is still reading out fine. We were seeing some 60Hz noise before and playing around with the biasing cables. Maybe something got bumped.
	Try restarting DCRC1 15V power, changing charge offset, but no change. Phonon channels are also railed. Connect them to LED lines to see some signal, but still railed. DCRC1 rd 0 = 0031, but DCRC2 rd 0 = updating. I hope DCRC1 isn't getting stuck output now too.
1315	Stop programs and power cycle DCRC1.
1321	Start up programs and pulse viewer. Initially DCRC1 is showing traces and rd 0 = updating. After a few min, DCRC1 has output stuck around 3V (similar to baseline) and rd 0 = not updating. These are the same things that happen to DCRC3 when it gets stuck.
1329	Turn off programs and power cycle DCRC1. Without starting programs, pulse viewer closed, DCRC1 rd 0 = updating.
1333	DCRC1 rd 0 = updating. Start dcrc_driver01. DCRC1 rd 0 = updating.
1339	DCRC1 rd 0 = updating. Start towerfe3_01. DCRC1 rd 0 = updating.
1346	DCRC1 rd 0 = updating. Start triggerfe2. DCRC1 rd 0 = updating.
1353	DCRC1 rd 0 = updating. Start run, wait 10 sec, stop run. DCRC1 rd 0 = updating.
1401	DCRC1 rd 0 = updating. Start run. DCRC1 rd 0 = updating.
1403	DCRC1 rd 0 = 0000. DCRC1 looks stuck. Turn on pulse viewer and confirm DCRC1 has stuck output while DCRC2 is fine.
1409	Power cycle DCRC1. Start programs, leave pulse viewer closed.
1444	DCRC1 rd 0 = updating. Use autoRD.sh script to rd all registers.
1545	DCRC1 rd 0 = updating. Start run. DCRC1 rd 0 = 0029 (stuck). Use autoRD.sh script to rd all registers.
	Differences in the registers in these two states. Some are expected to be constantly updating, lets get 1 more “good” set to compare.
1620	Turn off programs and power cycle DCRC1. Use autoRD.sh script to rd all registers.
	Seemingly relevant differences are shown below.

Register	Function	Good Value 1	Good Value 2	Stuck State Values
0x2,0x3	SDRam Read address	0000,0000	0000,0000	00F0,1728
0x14	SDRam data port	0000	0000	B67E

1633	Power cycle and get DCRC1 stuck again. rd 14 = 377F (stuck), but rd 2, rd 3 = updating. It looks like the SDRam address is being updated but the data coming out is not.
1700	Played a little more with this. The above “good” values are when first powered on, but without programs etc. running. It seems When first turned on, both the address (2,3) and the data(14) are 0000. Once the programs and run are started, 2, 3, and 14 are all updating. When the DCRC gets “stuck”, 2,3 continue to update, but 14 does not. If the run is then stopped, 2,3 stop updating. Looking at the description of these registers, it seems like the DAQ is setting 2,3 to request data and the DCRC fills 14 with data from the requested memory location.
1705	Maybe writing a bunch to 0x02,0x03 puts it in the stuck state? Wrote a script to write an address to 0x02, 0x03 10,000 times in a row, but DCRC1 continued to work after that.
1720	Close V6 to stop circulation and allow mix to expand into still line
1720	Turn off still heater
1730	Fill LN2 cold trap bath (will pump out trap tomorrow)
1730	G1/G2 have fallen to 90 mbar, Tower has risen to 330 mK, Still pressure has risen to 8 mbar.
1730	Close 1K Pot speedivalve to stop pumping on pot. Leaving needle valve open. Close V4a and valve to 4He pump. Open 4He outlet to “exhaust” and close recovery valve. Turn off 4He pump.
1735	Turn off 3He pump. Since v6 and v9 are closed, pump is isolated.
1740	Open V4 to let mixture on condenser side and in trap expand into dump.
1740	IVC = 16 nbar. (We have not been pumping on it this run.) Comparable to last run. During that run we pumped on the IVC continuously.
1740	Tower = 500 mK
1745	Tower = 500 mK, G1 (condenser/trap/dump) = 13.5 mbar, G2 (3He pump, off) = 3 mbar, Still = 70 mbar and rising, 1K Pot = 2.56 K and rising.
1745	Open V3 so still side can expand directly to dump too. Pressures don't change much so I guess still pressure had already equalized mostly with condenser pressure.
1750	Switch to Cernox thermometer. We're below its temperature range - T.UNDER on LakeShore bridge, about 0.1 K on LabVIEW, which is wrong. Overnight it oughta warm up into range. Set LakeShore excitation to 63.2 uV, range 632 kOhm.