sheila.dwyer@LIGO.ORG - posted 09:54, Monday 07 October 2024 - last comment - 15:21, Tuesday 08 October 2024(80508)
IMC locked with higer NPRO temperature
starting at 16:48 UTC, we have the IMC locked with the NPRO temperature at 0.3, compared to -0.18 for the last 1.5 years (the medm screen says this is in units of K). This was a suggestion from the PSL team to see if our problem is that the laser is near mode hopping.
Ryan Short noticed that this was still glitching at the higher temperature, so that hasn't solved the issue. The first two screenshots show times of the glitching, the glitches also show clearly in the PWR_NPRO channel, but they are not as clear when looking at minute trends as in the FSS channel. This test ran until 17:53 UTC.
We are now sitting with the IMC and FSS unlocked, to see if we see the glitches like this in the NPRO channel. This would rule out that the problem is coming from the FSS, and point to a laser problem. We will probably need to look at full data for the NPRO channel for this second test. We've been sitting here since 17:57 UTC.
We saw similar glitches in the NPRO power monitor with the FSS off as on, so the glitches don't seem to be coming from the FSS. (1st attachment)
Ryan next closed the shutter after the NPRO, before the first amplifier. We didn't see any glitches for nearly 2 hours, but then we saw a series of similar glitches (second screenshot). So this narrows the problem down to something in the laser or controller.
Continuing this glitch search from yesterday the PSL has been locked to the reference cavity with an NPRO temperature of -0.7 since 15:35 UTC October 8th. At that temperature, there was a glitch which looked slightly different from the usual glitches. There was also an osciallation in the FSS.
At around 9 am, I went to the diode room and turned off the noise eater, in that configuration I saw some glitches that looked fairly different from the ones seen regularly, it is mostly only visible in the FSS channel but can also be seen as a small step in the NPRO power channel. There were about 4 glitches like this in an hour.
Then we had the lower temperature (-0.7) with the noise eater on for about an hour, the glitches were not bad during this time.
Later, on a suggestion from Daniel, Rick and I went and disconnected the "diagnostic cables" which connect the power supply to the beckhoff system. To do this, we noted first the set and acutal temperatures and diode currents, as well as the A and B buttons. (I will add photos later of these).
Then we went to the diode room and followed instructions that Ryan Short gave me to turn off the two amplifiers in order, then the shutter and then we turned the NPRO off. We went to the rack, disconnected the cables, and turned the NPRO on by turning on the button on the controller box. This conntroller doesn't have a switch on the front panel for the noise eater, it was replaced by a cable which is no longer used. Filiberto looked up some information about this and tells us that the noise eater would be off in this configuration. We quickly saw that there were many glitches visible in this configuration, while we had the laser temperature back to it's usual -0.2K. This test started at 12:42 pacific-.
At 1:30 pacific we disconnected the "slow" BNC cable from the back of the controller, labeled NPRO temp, it was in this configuration from 1:30 to 2:15. We did see glitches in that time, but not the largest ones.
Now we've set the temperature back to normal, and reconnected the cables, and turned back on the amplifiers and their watchdogs. Oli and Tony are proceeding with initial alignment and Rick and I will reset the watchdogs before leaving.