I didn’t get very far in the HVAC experiments because the EY chiller water pumps did not recover from the site-wide shutdown until I went to the chiller pad and reset them. The time lag and hot day meant that I had to wait a few hours for recovery (fortunately I reset it quickly enough that we didnt loose lock). While I was waiting for EY temperature recovery, SF-6 (lab area cold-deck) went off on its own and Richard brought Ken in to work on it - he says one of the wires connected to the breaker was burned through. Ken just got it fixed and we are both headed out.
I did learn a couple of things in the abbreviated experiments:
1) both the chilled water turbulence and the turbine plenum turbulence are contributing to the vibration background at 4 Hz at EX
2) side-wide shutdown of turbines, associated chillers, and water pumps increased the range from about 128 to about 138 Mpc
Since I couldn't complete the experiments, I tried just turning the turbine flows down to pre-pandemic levels. I was unable to do this for the CS and EX HVACs - to me it seemed like vane actuator slippage, like we have seen in the past. EY AHU-1 worked fine for me and I turned the flow down from 12k to 9k CFM as agreed with Bubba. Otherwise I have left everything the same. I did not get a chance to toggle the lead chiller at EX.
EX HVAC experiments all times May 19 UTC:
16:21:51 EX chiller un-enabled, AHU manually disabled
16:32:58 CHWP-2 disable
16:39:38 All systems back on
17:08:36 Manual chiller 2 enabled uncheck, AHU-2 Manual Disable checked
17:11:56 Chiller still on start so uinchecked chiller system enable
17:21:22 checked manual pump disable
17:35:30 system restored: AHU-2 disable unchecked, Chiller system ehable checked
18:43 AHU manual disable checked, Chiller system enable unchecked
~18:55 system restored
Site-wide HVAC off:
20:16:05 start of site off
20:23:32 back on
This plot is a bit out of date, but here is a DARM comparison for the time when the site HVAC was off. The range improvement from this HVAC shutoff that gained us about 10 Mpc appears to be a reduction in noise from 100-200 Hz in DARM. I think that region is mostly limited by jitter noise.