2019/07/22 (Monday)
| 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)/DCRC55(1) DCRC33(2) → 1kV power supply → HV input Three walls of the lead/poly shield are up. |
|---|---|
| Status: Taking bkg data. | |
| 0924 | Fill LN2 cold trap bath. |
| 0925 | Tower/MC/Coldplate = 39.0/30.2/199 mK Still = 1.3 K/ 0.10 mbar / 0 mW, G1/G2= 110/99 mbar, 1K Pot = 1.7 K / 90±10 mbar. MB= 47.2% |
| 1154 | Commence LHe transfer. MB=43.4% |
| 1205 | Check event rate. Count number of visible pulses in 50 traces, three trials: count 7, 6, 9 pulses. 44.75±9.54 Hz. |
| 1212 | Check event rate. Count number of visible pulses in 50 traces, three trials: count 5, 9, 9 pulses. 46.79±9.76 Hz. |
| 1226 | PuBe source in lab in paraffin barrel behind lead stack. |
| 1231 | Check event rate. Count number of visible pulses in 50 traces, three trials: count 13, 19, 11 pulses. 87.48±13.34 Hz. |
| 1235 | Check event rate. Count number of visible pulses in 50 traces, three trials: count 13, 14, 14 pulses. 83.41±13.03 Hz. |
| 1235 | Cease LHe transfer. MB=93.4% |
| We've been taking data at an HV drift rate of 2.49V/min (0.0415 V/sec). At a guess, with the PuBe source present the drift rate should be 85/45 times higher, 4.704V/min (0.0784V/s) | |
| 1550 | Write new sequence, using loops like the last one, to take data at 180V initially with a +4.83 V/min drift. Target the same maximum voltage as before, so the series are only 26.7 min long, final voltage is about 327 V. |
| 1557 | Start a sequence of 8 @ 27 min PuBe series at HV = 180→327V, rate 4.83V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl |
| 1620 | MIDAS web page is down, I guess the DAQ crashed… HV is at 179.7 V and not changing. |
| 1620 | Check the programs on cdms2 - looks like mhttpd has crashed. Try a ./start without a ./kill. |
| 1625 | Sequence starts again where it left off. It had already started taking data, looks like about 15 minutes with the HV stuck at 180V. So this series will be at too-low bias. But, we should be able to use it to see whether the drift rate needs to be modified. |
| Series should end about 16:50. We'll wait for the flash and then abort the sequence and see what we have. | |
| 1630 | Check BatViewer. Unfortunately it looks like the DAQ is triggering constantly. Maybe noise is higher with the higher rate? Let's try increasing the trigger threshold. NO! actually the trigger had not been enabled! It had been turned off in the ODB to do the rate checks earlier, and had not been turned back on. Turning it back on now. |
| 1632 | Check with pulse viewer, all the channels look good and it looks like we are getting triggers in the right place. We'll get about 16 minutes of proper data to look at. |
| The series failed to process because Dump 1 had no data. | |
| 1735 | Start a sequence of 8 @ 27 min PuBe series at HV = 180→327V, rate 4.83V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl |
| 1747 | MIDAS page is down again and mhttpd is not running. Again, use ./start without ./kill. Sequencer picks up at loop 1/160 of the HV drifting loop. |
| Must be something about the sequence that’s causing mhttpd to crash. I’ll try editing out most of the comments. | |
| Again Dump 1 is empty because of the crash. This won't process either. May as well stop the sequence and dial down the HV and flash. stupid DAQ is a piece of junk. | |
| 1815 | Stop sequence manually. |
| 1820 | Manually bring HV to 0V by stepping up DCRC2qI to 11.64 V |
| 1830 | Flash DCRC1&4 LEDs for 30 seconds |
| 1840 | Modify HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl , removing a lot of the comments. |
| 1917 | Start a sequence of 8 @ 27 min PuBe series at HV = 180→327V, rate 4.83V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl |
| 1927 | gdamn thing crashed again, pissing me off. |
| 1935 | get MIDAS started again then dial HV to zero |
| 1940 | completely kill MIDAS then restart MIDAS |
| let's try the sequence again. Not bothering to flash this time, just see if it crashes. | |
| 1945 | Start sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl again, just a test |
| 1955 | Crashed again. It happens right at “TRANSITION START”. I noticed that this is right after the line ODBSET “/Seriesinfo/Duration (s)” 1600.0 . I changed this to reflect the actual series length. But maybe this screws it up for some dumb reason. |
| 2005 | Modify HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl so it now says ODBSET “/Seriesinfo/Duration (s)“5280.0 like the previous, working sequence. It's not correct but that doesn't really matter. |
| 2005 | Step HV back down to 0V |
| 2009 | Start sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl again, just a test |
| 2020 | crashed a-fucking-gain. This time the tower_fe stopped too, for variety. restart DAQ, step HV to 0V. |
| 2033 | Start a test sequence, an edited version of the sequence we were using before today. |
| 2042 | Crashed again. Something else is apparently wrong. Maybe something in the ODB is set wrong??? |
| 2045 | restart daq, step HV to 0V |
| 2050 | edit test.msl again, try again. Another crash. In this one I removed all the bias-changing lines. It still crashes at TRANSITION START. |
| 2120 | do a series of test sequences. A minimal sequence started working. Then started adding lines, and it kept working. it sorta seems like it randomly fixed itself. |
| 2124 | Start sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl again, just a test |
| OK, this time it's working, at least the first series started. Just fixed itself. Dumb. I'll just let it go, though the first series is probably starting out with a counterbias. I'll just copy below that I started a new series with the standard format: |
|
| 2124 | Start a sequence of 8 @ 27 min PuBe series at HV = 180→327V, rate 4.83V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl |
| 2140 | crap!! Dump 1 has no data!!! That is apparently an unrelated problem to the crashing. frug. |
| 2142 | stop sequence |
| 2205 | this MIDAS daq is an unstable piece of shit. I've tried several tests again - in one case it actually produced a Dump1 with data in it. In a subsequent test the DAQ crashed again. In yet another test it is again producing empty Dump1. bullshit. |
| after some more tests, trying again, but only after a full DAQ kill and start. | |
| 2220 | Start a sequence of 8 @ 27 min PuBe series at HV = 180→327V, rate 4.83V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_4p83VPM_loopversion_prebias_10_2_dur_27min_flashDCRC_1_4.msl |
| nope, another crash. Absurd. | |
| 2230 | Restarted the DAQ, sequence restarts, and I see that again Dump1 is empty. |
| trying to decide whether to let it keep going or to kill it. Likely if I let it keep going it will get stuck at high bias. Would it be bad for it to be at high bias overnight? | |
| 1038 | stop sequence |
| ok now I'm going to try the old sequence. | |
| 2242 | Start a sequence of 8 @ 50 min series at HV = 180→306V, rate 2.49V/min, with 10% prebias for 2 minutes, flash at end with 30 min wait. CDMS detector: Driver Gains = 1x, DCRC4/1 QET biases = (490/400/495)/(420/480/450) uA, Front-End Gains = (2/2/3)/(5/5/5) 1000 BOR, 5Hz InRun randoms, trigger on DCRC1pD threshold = 6, baseline/sample length = 120/28 (standard 16/5), phonon pre-pulse 500 (standard is 500). Use sequence HVdrift_8X180V_2p49VPM_loopversion_prebias_10_2_dur_50min_flashDCRC_1_4.msl |
| another crash. so I guess it has nothing to do with the new sequence. | |
| oh but this time after restarting it Dump1 is not empty. So maybe that part has something to do with the new sequence. | |
| maybe we need to have cdms2 rebooted | |
| 2300 | giving up for the night |