Displaying report 1-1 of 1.
Reports until 04:14, Saturday 23 May 2015
H1 ISC (DetChar)
evan.hall@LIGO.ORG - posted 04:14, Saturday 23 May 2015 - last comment - 15:59, Sunday 24 May 2015(18592)
Back to dc readout

Kiwamu, Peter, Dan, Evan

We made it back to dc readout tonight. 3 W PSL power, DARM feedback to ETMX.

We are in the process of trying to engineer the new transition to ETMY. We were hampered by the intrusion of a large amount of 1/f3 noise in DARM, with ASD 1×10−14 m/Hz1/2 at 10 Hz. The initial portion of the lock was fine, and then the noise came. We trended various channels and looked at coherences, but could not nail down the cause. The noisy time starts at 2015-05-23 9:41:00 UTC. After about an hour, the noise went away again.

The ASC was a bit touchy. It could be engaged by hand, but with the guardian it seemed that engaging the PRM pointing loop caused several of the other loops to go unstable. Also, we gave up on the ITM pointing loops. They increased the recycling gain, but made the arm and sideband buildups less stable.

We found that there was no tidal signal being sent to ETMY. At first blush it appears to be an IPC issue.

Comments related to this report
evan.hall@LIGO.ORG - 06:41, Saturday 23 May 2015 (18593)

Summary

We were able to transition control of DARM to ETMY with feedback to the UIM, PUM, and test masses, with the ESD controlled by the low-noise driver. The low-pass filtering was off, but the rms drive to the ESD is low enough that we should be able to switch it on without inducing saturation.

The philosophy was to first adjust the gain and filtering of ETMY L3 and ETMY L2 so that they each actuate identically to ETMX L3. Then we engineered an L2/L3 crossover by inserting a 30 Hz lowpass into L2, and a 30 Hz highpass into L3. From 10 Hz to 100 Hz, this makes the combined ETMY L2/L3 actuator look similar to the ETMX L3 actuator. Then we transitioned control from ETMX to ETMY in the usual way; i.e., ramping on the gain in ETMY L3 LOCK L (with the offloading engaged), and then ramping down the gain on ETMX L3 LOCK L.

Details

Although we set the binary sus switches so as to have lowpass filtering on, it was evident by comparing the TFs of ETMX L3 and ETMY L3 that the filtering was off. We didn't want to spend time debugging this BIO issue, so we undid the digital LPF compensation (see below) and proceeded with the transition anyway.

The procedure was as follows:

  • Measure the transfer function of ETMX L3 by measuring the transfer function L3 LOCK L EXC → DARM IN1.
  • Measure the transfer function of ETMY L3 by an identical procedure (with L2 and L1 switched off). Adjust gain and filter shape in L3 drivealign L2L to make it match the above ETMX transfer function from 10 to 100 Hz. We found we needed a gain of −50 ct/ct.
  • Measure the transfer function of ETMY L2 by measuring L2 LOCK L EXC → DARM IN1 (with L3 and L1 switched off). Adjust gain and filter shape in L2 LOCK L to make it match the above ETMX transfer function from 10 to 100 Hz. Kiwamu had done this previously; the result is FM6 (Q), along with another filter in L1 (FM6, Q^-1) which undoes this filter (since we are using offloaded feedback for the LOCK L filters). We needed a gain of 15 ct/ct (previously it was 20 ct/ct). FM7 in L1 LOCK L is meant to undo 20 ct/ct of gain in L2 LOCK L, so this should be retuned for the new gain.
  • Add a 30 Hz pole to the L2 drivealign L2L. Add a 0 Hz zero and 30 Hz pole to the L3 drivealign L2L. We put these in the drivealign FMs so that they do not affect the offloading.
  • Then ramp up the gain of ETMY and ramp down the gain of ETMX as described above.

In L3 L2L there is also a highpass filter at 1 Hz which cuts out the appearance of the microseism in the ESD drive. This reduces the rms ESD drive from 104 ct to more like 103 ct.

In L3 L2L, FM2 is used to compensate for the antiLP filters which are engaged in the ESDOUT FMs. This FM2 was installed only because ESDOUT is currently not correctly compensating the analog driver transfer function (since the analog LPFs are off; see above). FM2 should be removed once this is fixed.

The DARM OLTF is attached. Blue shows our "best" DARM OLTF configuration (i.e., what we use in full low-noise lock), with about 50 Hz ugf. Orange shows the OLTF we measured tonight on ETMX, before doing transitioning. The gain is intentionally low at this stage, because it facilitates ramping on ETMY feedback. Red shows what we measured with ETMY controlled by the LVLN driver. Here the DARM gain has been increased slightly. But the most notable feature here is the apparent loss of phase after transitioning. It is not immediately clear to us where this phase loss is coming from.

Crossover measurements have not been taken, and certainly there is some tweaking that we can do.

Images attached to this comment
Non-image files attached to this comment
john.worden@LIGO.ORG - 06:19, Saturday 23 May 2015 (18594)

There were some severe thunderstorms and downpours last night - Any chance this could be a cause of the noise at 9:41?

nutsinee.kijbunchoo@LIGO.ORG - 11:48, Saturday 23 May 2015 (18595)

Came in to give a tour and found that it's been losing lock at Find IR. So I took the ISC_LOCK to DOWN.

peter.fritschel@LIGO.ORG - 07:18, Sunday 24 May 2015 (18597)

Great to see you got DARM control to ETMY working!

Regarding the phase loss: take a look at one of the transfer functions pointed to in entry 18579  Besides the 2.2/50Hz pole/zero pair, Rich said there is another pole above/around 100 Hz due to the output capacitance. This will need to be at least partially compensated.

kiwamu.izumi@LIGO.ORG - 15:59, Sunday 24 May 2015 (18598)

John, we don't think that the thunderstom have caused the 1/f3 noise in DARM. In the periode we had this mysterious noise, there were no significant vibration on the ground according to the seismometers.

Displaying report 1-1 of 1.