2017/05/24 (Wednesday)
| Setup: DIB1/3/4 → 3.2 MHz filters 3/1/2 → cable 3/1/2 →gender changer 3/1/2 → HVConn 1/3/2 → HV 1/2 → DCRC55/DCRC33 | |
| Took QETbias-scan data overnight. Up to the _2320 series looks good. But in the _0020 series, at Event 134 (about 7 seconds in) all the channels rail. Maybe the driver offset settings didn't work. That means that Ib*pf = 30,25,20,15 all worked but 10,5 didn't work. | |
| 0835 | Tower = 33.0 mK, Main Bath = 69.5%. |
| 0835 | Fill LN2 cold trap bath. |
| 0910 | Looking at the DCRC1 channels, it seems as though they had fallen into s/c. After snapping and then returning them to the default Ib*pf=30 biases, it seems that the offsets have shifted compared to yesterday. The appropriate driver offsets are now 0.4 to 0.6 lower than they were yesterday. Did all three lose lock in the same way? |
| 0920 | Relocking DCRC1pA gets the offset closer to what it was yesterday. Tried the different QET biases - the Ib*pf=30 and 25 values are ok but decreasing to the Ib*pf=20 value causes it to drop into superconducting. Weird. |
| 0955 | Finish IbIs on DCRC1pA/Side1pE. The current at which it falls into s/c is 126.82 uA, very different from the last IbIs five days ago where it was 123.98. QET bias 128 previously indicated Ib*pf around 30 but this IbIs suggests it's around 16. Huge change. It indicates that P0 was 92.2 pW, now 96.5 pW. That's more power required to hold the TES at transition, so it seems as though parasitic power has decreased. |
| 1030 | DCRC1pB has a s/c falloff at about 138.3 uA, previously was 134.8 uA. Also suggests a decrease in Ppar. |
| 1035 | DCRC1pC has a s/c falloff at about 124.8 uA, previously was 122.72 uA. Also suggests a decrease in Ppar. The IbIs shifts suggest a fall in Ppar for these three channels of 4.3/5.7/3.1 pW. This is an approximation assuming zero Rpar. It's 3.1/4.2/2.2 if you assume Rpar = 9 mOhm. Regardless of the Rpar assumption it's a 4.6/5.2/3.4% change. It doesn't seem as though Rpar has changed, it's about the same for DCRC1pA as in the older IbIs. |
| 1120 | On DCRC2 there are smaller shifts, only about 1 pW / 1-2% in pB and pC, and pD is very close, in fact its falloff current decreased a little, suggesting a Ppar increase of 0.5 pW / 0.5%. |
| 1145 | Remeasure falloff current on DCRC1pA and now it's back to 123.8, similar to the old value. This is reminiscent of the last run, where I saw that P0 measurements were not stable if you let the TES go s/c. Now we do take the TES into s/c but only briefly. Maybe with the channels all in s/c overnight they went into a low Ppar mode for a while, and now they are back to normal. |
| 1150 | Similarly DCRC1pB's falloff is lower now, closer to before (although maybe even lower). Same with DCRC1pC. |
| QET bias settings for different dP/dI [μA] | |||||||
|---|---|---|---|---|---|---|---|
| QETbias * prefactor [μA] | |||||||
| channel | 30 | 25 | 20 | 15 | 10 | 5 | 0 |
| DCRC1pA/DIB1pA/Side1pE | 127.5 | 126.55 | 125.7 | 125 | 124.5 | 123.9 | 124.0 |
| DCRC1pB/DIB1pB/Side1pB | 139.3 | 138.1 | 137.05 | 136.25 | 135.55 | 135.1 | 134.9 |
| DCRC1pC/DIB1pC/Side1pD | 126.65 | 125.6 | 124.7 | 123.85 | 123.3 | 122.95 | 122.7 |
| DCRC2pB/DIB3pC/Side1pA | -117.8 | -116.45 | -115.3 | -114.35 | -113.6 | -113.15 | -112.8 |
| DCRC2pC/DIB4pC/Side1pC | -150.05 | -149.15 | -148.35 | -147.65 | -147.15 | -146.7 | -146.45 |
| DCRC2pD/DIB4pD/Side1pF | -146.2 | -145.35 | -144.7 | -144.15 | -143.75 | -143.45 | -143.4 |
The table above is a correction of yesterday's table. The DCRC1pA values are corrected. We'll use these to repeat the Ib*pf=10 and 5 data sets. Correction: let's skip the Ib*pf=5, there's too much uncertainty in that setting.
| 1235 | Although the DCRC1 channels seem back to normal, the DCRC2 falloff currents have shifted again. The QETbiases in the table above for Ib*pf = 10 are now too low for the DCRC2 channels. It seems the Ppar values are 1-3 pW lower than normal in these channels. |
|---|---|
| 1240 | And the DCRC1 channels are now all railed. They had been set to the Ib*pf = 10 biases. Now maybe they've fallen into s/c? |
| 1310 | After restoring DCRC1 channels and leaving it for a bit, recheck the DCRC1 channel falloff currents. They seem to be at the old values. Checked DCRC2 channels again as well and their falloffs also seem to be at the old values. |
| 1330 | Tower/MC/Coldplate = 32.9/29.6/193 mK, Still = 1.27 K / 0.110 mbar, G1/G2 = 219/211 mbar, 1K Pot = 1.678 K / 24.0 mbar. Main Bath = 62.4%. Boil off rate: close to slow mode (Predicted level: 63.4%) |
| 1410 | After several attempts, all 6 channels are set to there Ib*pf = 10 values from the table above. |
| 1414 | Start 2 1-hour data series (start #) of 1-sided 0V data with Frontend Gain = (15/15/5)/(5/4/4), OG = 4/2, QET biases set with the Ib*pf=10 values in the table above, with 1000 beginning randoms, no flashing, and a 60 ADC bin threshold trigger on DCRC1 pA (Side1 pE). DCRC4 is still set up to read NaI triggers but isn't relevant for this sequence. |
| 1537 | Stop sequence, because everything railed about 3 minutes into the second series. All channels rail, as though maybe one channel fell into s/c which put the DCRCs into low Pparasitic mode, which then suddenly drops every other channel into s/c (because they are close to the bottom of transition). |
| 1545 | Brought each channel, in order, into transition by increasing QET bias by 10 uA. When I got to the last one, DCRC2pD, it had already come back into transition without needing adjustment. So maybe DCRC2pC (which is an unstable channel) is the key one - maybe when it falls out of transition it brings the others down. |
| 1615 | A confirmation of the above: I set DCRC2pC higher into transition, at Ib*pf=20, then was able to set the other 5 channels to the low point of Ib*pf=10. Then I tried to lower DCRC2pC to the Ib*pf=15 level, and it went s/c - and all five other channels railed. It looks like they are all s/c. |
| 1635 | Checked the falloff currents for the DCRC1 channels while DCRC2pC was still s/c, to see if they had shifted to values characteristic of the low-Ppar state. But they seem to be at their old standard values. Same with the other two DCRC2 channels. Also DCRC2pC itself seems to have the old standard falloff value. |
| 1802 | Fill LN2 cold trap. |