Sheila, Alexa, Dan, Nic, Evan
EX/EY:
-
Sheila and I went to EX and EY to assess the state of the LSC-X/Y_TR photodiodes. At EX we didn't do anything. At EY, we turned down the preamp gain on the ThorLabs PD from 20 dB to 0 dB. We hope this will prevent saturating the ADC when there is significant IR light in the arm.
-
Over a period of 10 minutes, we can see the ETMX oplev drifting significantly; this degrades the built-up light in the arm when locked.
Setting up sqrt(TRX+TRY):
-
Our goal was to send the sqrt(TRX+TRY) signal to REFLBIAS which is connected to the second input of the CM board, and replace the COMM PLL CTRL signal which is used for ALS COMM. The COMM PLL CTRL signal passes through a low pass (40 Hz) and boost (1.6/40 Hz p/z). So we copied these filters into the digital filter of REFLBIAS (FM3). We also installed an integrator with a zero at 50 Hz (FM5), so that the sqrt(TRX+TRY) signal crosses over the ALS COMM OLG at low frequencies as we increase it's offset. We also have some calibration filters (FM4) installed. Meanwhile, the normalized sqrt(TRX+TRY) filter, LSC-TR_CARM had an offset of -1.4 and a calibration filter to pm. We had set up a trigger on REFLBIAS and the integrator, so that REFLBIAS only engages when TRX is above a threshold of 1.0.
Attempt at sqrt(TRX+TRY):
-
Several times tonight we were able to lock DRMI on 3f with the arms held off resonance by ALS. We were able to bring in the CARM resonance to about 800 Hzgreen, at which point REFLBIAS engaged and blew the lock. So we will have to try a different stragety for the handoff of ALS to sqrt(TRX+TRY).
-
In order to get DRMI to lock with arms, we needed to increase the MICH gain to 6 (up from 4 yesterday).
-
This afternoon, Nic and I realigned the ASAIR A and B diodes, and the ASAIR camera.
-
Sometimes, we lose lock because the endstation VCOs are running out of range. Some kind of offloaded slow feedback would be helpful here.
Guardian improvements:
-
There have been miscellaneous improvements to the ALS and DRMI guardians to improve the lock acquisition time. We were able to get DRMI+arms to lock more frequently tonight than in previous nights.
-
Alexa and Nic implemented a state in the ISC_DRMI guardian that offloads the ASC WFS to the M2 lock offsets in PRM, SRM, and BS.