The HAM5/7 gate valve is open, so we have returned to low noise operation at 50W. We have a first look at our darm spectrum post-vent. We have returned to the 20mA darm offset, but we are still in the lower whitening OMC DCPD configuration because some violin modes are rung up and we want to avoid saturation. Evan did some retuning of SRCL and MICH feedforward.
We are attaching a spectrum taken during quiet time from UTC 23:39 to 23:49 Nov. 11. Note that we have not redone the calibration at this time. There is also no squeezing at this time.
PCALY to DARM BB injection for this lock. We are reasonably calibrated after this vent, within 5%. Control on EX. Coil drivers low passes on core optics PUM stages (IX, IY, EX, EY) were switched on (state 3). /ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2022-11-11_H1_PCALY2DARMTF_BB_after_vent.xml
Couplings with corner length degrees of freedom attached. SRCL needs a better retuning, which should probably be done more methodically than by eye. This new SRCL feedforward filter was added to the guardian (FM2 in SRCL1 FF bank). The MICH FF filters were not changed, but a slightly lower gain (0.93) was needed compared to the pre-vent setting (1).
A comparison of DARM before and after vent
I ran a Bruco during the quiet time Elenna posted above (thanks Dan Moraru for pointing out I was looking for the results in the wrong directory, d'oh).
Some interesting things:
- Big line at 28.25 Hz coherent with DHARD, AS_A_RF45, ASC-REFL_A_RF9_PIT, some HAM1 HEPI RZ channels, a PSL accelerometer..... Tagging PEM
- SRCL and MICH coherence below 100 Hz
- End X ESD power monitor up to 100 Hz, incluing 95 Hz peak (note this is still on the temporary power supply by the racks)
When comparing the new to old spectrum, there are two important caveats: we are using the OMC B chain so the calibration is wrong and worse according to a broadband pcal excitation than before the vent. Craig made some measurements of this, there are 5% errors in the calibration. We are also operating at a DC photo current of 20mA, for some portions of this lock.
both of these things mean that the feedforward and cleaning will be mistuned, and the calibration is wrong. We should be able to revert to the calibrated DCPDs and settle on a Darm offset with the new TSAMs after the violin modes are under control. So I don't think it's worth spending peoples time on retuning feedforward or calibration in this state.