travis.sadecki@LIGO.ORG - posted 22:36, Monday 06 February 2017 - last comment - 23:23, Monday 06 February 2017(33951)
Observing at 6:35 UTC
We are back to Observing, but PI mode 26 continues to be unresponsive. It is currently elevated but not rising. I'll continue to monitor it, but I still can't damp it.
Comments related to this report
terra.hardwick@LIGO.ORG - 23:14, Monday 06 February 2017 (33952)
Was just able to damp this remotely finally. Travis has edited, saved, and loaded the gain that worked to the SUS_PI guardian (+4000 gain). The phase that worked is -30 deg. Thanks Travis.
This mode is suddenly finnicky (it hasn't been problematic during this run before). To damp, it required low gain (~1000) immediately at power up and I was able to damp by finding the phase that most leveled out the mode and then stepping up the gain 1000 at a time to around 5000. This is the usual procedure and exactly what Travis did correctly; just seems we needed a few tries to find the set up that worked and that it required more fine-tuned gain and phase than usual. As Travis mentioned above, once PI modes get very rung up (a bit over 100 on the StripTool for this one for example) they become unresponsive. They also take awhile to damp; Travis's second lock loss was likely due to relocking quickly and the mode was still too rung up to be able to damp.
This mode experienced about 1 Hz shift a few days ago, likely due to temp changes that have been messing up other PI modes too. However, Travis pointed out that it was damped successfully for several days since then using the old settings. I'll take a look at it's frequency shift behavior over the last few days and see if anything different happened in the lock before all this started.
terra.hardwick@LIGO.ORG - 23:21, Monday 06 February 2017 (33953)
Note: while I was writing the alog above, I watched Mode26 start to rise again (about 45 min to an hour into lock). I stepped the gain along from -30 --> -90 and it turned the mode back down easily.
travis.sadecki@LIGO.ORG - 23:23, Monday 06 February 2017 (33954)
Was just able to damp this remotely finally. Travis has edited, saved, and loaded the gain that worked to the SUS_PI guardian (+4000 gain). The phase that worked is -30 deg. Thanks Travis.
This mode is suddenly finnicky (it hasn't been problematic during this run before). To damp, it required low gain (~1000) immediately at power up and I was able to damp by finding the phase that most leveled out the mode and then stepping up the gain 1000 at a time to around 5000. This is the usual procedure and exactly what Travis did correctly; just seems we needed a few tries to find the set up that worked and that it required more fine-tuned gain and phase than usual. As Travis mentioned above, once PI modes get very rung up (a bit over 100 on the StripTool for this one for example) they become unresponsive. They also take awhile to damp; Travis's second lock loss was likely due to relocking quickly and the mode was still too rung up to be able to damp.
This mode experienced about 1 Hz shift a few days ago, likely due to temp changes that have been messing up other PI modes too. However, Travis pointed out that it was damped successfully for several days since then using the old settings. I'll take a look at it's frequency shift behavior over the last few days and see if anything different happened in the lock before all this started.
Note: while I was writing the alog above, I watched Mode26 start to rise again (about 45 min to an hour into lock). I stepped the gain along from -30 --> -90 and it turned the mode back down easily.
Went out of Observe for less than 1 minute at 7:01 UTC to reload the Guardian code for this change.