Summary:
We checked ground loop of (almost) all ISC and SUS but not SEI cables at racks or SAT amps. Everything was good except ASC_AS_C, OM3, OMC QPDs and ZM1.
ASC_AS_C and OM3 were fixed.
ZM1 is still bad and the problem should be either the in-chamber cable between the feed through and the cable bracket or the connection at the feed through.
OMC QPDs were not fixed, I'm still trying to remember if QPDs were like this in the past.
Details:
The check method we used was to remove cables at the rack, test the connection between pin 13 and the chassis ground, and between pin 13 and the cable shielding. There should be no connection to the chassis ground but there should be to the shielding.
ASC_AS_C problem was the feed through connection, which doesn't make sense, but it was fixed by undoing and reseating the connection at the feed through in chamber.
OM3 seems like our old problem (https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=12345) where bad design means that the male micro DB connector shell is too close to the surrounding metal. I undid all of the tiny connectors on four OSEMs, reseated and engaged screws again. Each OSEM connector has two screws, and when fully tightened both I got a short. I tried to make the outside screws looser than the inside screws and this seemed to have done the trick. I'm afraid this is just a bad adjustment in an attempt to get a temporary relief. TVo has a picture of the BOSEMs and it seems like there are some spaces between the shell of the connectors and it looks OK-ish.
ZM1 is still bad. Disconnecting ZM1 BOSEM cable didn't do anything, so it should be the cable from the feed through to the cable bracket in chamber, or the connection at the feed through just like ASC_AS_C. Problem is that the feed through is at the top of the chamber and it's not easy to reach inside.
OMC QPDs are strange. When OMC DCPD in-air cable is disconnected from the chassis at the same time, there's no connection between OMC QPD in-air cable pin 13 and chassis. However, when OMC DCPD cable is connected, OMC QPD in-air cable pin 13 has a connection to chassis. OTOH, OMC DCPD cable pin 13 is not connected to chassis no matter what.
I'm trying to remember if this is the same thing as https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=12349. We checked the continuity between DCPD signal ground (pin15, 16 and 19, 20 of the DCPD cable) and the shield of OMC QPD cable, and of course they're still continuous because it's the design error, but I cannot remember if the symptom we're seeing can be explained by just this. Asking Rich.
Hope this helps...
The short circuit of the chassis and pin13 of the DCQPD happens if the in-air DCPD cable is connected to the whitening chassis
So if you check Pin13 of the disconnected in-air DB25 for the DCQPD (=the shield of the cable), you find the short with the rack chassis (= the signal ground) if the DCPD chain is connected.
When the in-air cable for DCPD is disconnected at the rack, the connection 1) is cut. Therefore, you no longer see the short between the pin13 of the disconnected in-air DB25 for DCQPD and the rack chassis.
----
LHO ALOG 28969 says: "All the OMC cables (OMC DCPDs, OMC QPDs, and OMC PZT) are sharing their shield on the OMC breadboard. Therefore one needs to remove the related three DB25 cables from the flange when the shield shorting is checked for them."