craig.cahillane@LIGO.ORG - posted 14:39, Friday 24 June 2016 - last comment - 19:34, Friday 24 June 2016(27953)
ASC Soft Loops Input Matrix
Shelia Dwyer, Craig Cahillane We have changed the CSOFT and DSOFT input matrix in an attempt to isolate the CSOFT and DSOFT error signals from one another. In aLOG 27944, Shelia mentions separating the HARD and SOFT signals in the input matrix. This aLOG is dedicated to the common and differential mode separation. To get solely common and differential modes, we must have our X arm and Y arm signals scaled by factorsuandvto be of the same order:CSOFT = u * X_sig + v * Y_sig DSOFT = u * X_sig - v * Y_sigWe worked on findinguandvby making ~ 1 μ radian alignment changes in the X and Y ITMs and measuring the combination of TRX A and TRX B signals that is insensitive to HARD, according to aLOG 27944. The same was done for the TRY A and TRY B output signals: MeasurementsX_sig (X Combo) Y_sig (Y Combo) Yaw 0.03975 -0.0781 Pitch 0.0302 0.106We then found and normalizeduandvby the same scalar, separated out the TR? A and TR? B components of the HARD insensitive signals, and put them in the input matrix for both Pitch and Yaw:u (Yaw X) v (Yaw Y) u (Pitch X) v (Pitch Y) TR? A -0.3521 0.0848 0.9420 0.0201 TR? B 0.8187 -0.4456 0.1936 -0.2733In the actual input matrices we have flipped every sign in the numbers above in order to have a positive gain in the CSOFT and DSOFT control loop filter banks. The code that calculates the above matrix lives in/ligo/home/sheila.dwyer/Alignment/DSOFT/sens_TMS.m
The message- soft loops work to maintain the arm alignment with our new matrices, but this doesn't keep the recycling gain high. The recycling gain can be improved by yawing TMSX.
What Jenne and I realized yesterday was that we need to measure the soft sensing within the loop bandwidth (by making DC changes) with the hard loops closed, to avoid having our measurement of the sensing contaminated by the much larger hard signal. By measuring the soft sensing this way, we measure the sensing of the degree of freedom which is not controled by our HARD loops.
Measuring this way we found that there is better separation between the gouy phases of the hard and soft signals than we had found by dithering at 8 Hz, (alog 27935) We now see that the soft signal is separated from the hard signal, for X yaw by 75 degrees, x pit by 33 degrees, y pit by100 degrees , y yaw by 78 degrees. Based on these measurements we found combinations of the QPDs that are senstive to moving the ITMs while the hard loops are closed (sensitive to soft), and checked that the orthogonal combination was insensitive.
This morning Craig and I checked how to combine them into common and differential as Craig explained in his alog. We reset the QPD offsets after searching by hand for a good power recycling gain. We powered up to 20 Watts, and rechecked our combination of QPDs, which still seemed good.
We can see that this new error signal is keeping X arm optics in the same location as we power up better than our old combination, based on trending the green transmitted power as we increase the power. However, it doesn't keep the recycling gain high. We tried putting offsets in the POPX-> PR3 pit loop, which didn't help restore the recycling gain. We can restore the recycling gain by moving ITMX yaw with the soft loops open, but this is moving the optics relative to where they were before power up (based on green transmission). We can also improve the recycling gain (back to about 32, when it started at almost 35 before the power up) by moving TMSX yaw (-1.3urad at 20 watts).
We again commented out the SRC1 loops in full lock, because they were misaligning the IFO.