Earlier tonight Alexa made some measurements of IMC transfer functions to for matching with her model, and I worked on the slow feedback to the ETM top mass. (alog about that coming in the morning, in short the L2P decoupling has helped)
Now I've made some measurements similar to what ALexa and I did on Monday night (alog 10668) of the linear coupling from the test mass angles to the out of loop frequency noise sensor (the normalized refl PDH). This time I dithered the PZTs on ISCEX, and measured the spectrum of the normalized PDH and the green QPDs on the input beam on the TMS table. I haven't calibrated the QPDs, or measured the loop gain, but these are verry slow loops and I dither the PZTs at 5Hz so that should not matter. By dithering each PZT in both pitch and yaw, and finding the peak in the spectrum of both QPDs each time, we get 2 measurements of the coupling from each QPD to frequency noise.
| Hz/ count measured by dithering PZT1 | Hz/count measured by dithering PZT2 | |
| QPDA PIT | 114 | 126 |
| QPDA YAW | 66 | 154 |
| QPDB PIT | 92 | 44 |
| QPDB YAW | 51 | 69 |
As you can see the measurements made with different PZTs aren't always consistent. Using these and measured spectra of the QPDs, we can make the projection of the linearly coupled noise from input beam pointing to ALS frequency noise, it is not a significant contribution (below 100Hz) based on any of these measurements. At around 100 Hz these projections seem to explain the noise, I believe this is just the noise of the QPDs, but will check in the morning.
I also checked the linear coupling from the ETM pitch to frequency noise, to see how much it changes with alignment. I measured 424Hz/urad, moved the ETM by 0.1urad, measured 1.12kHz/urad, tried to move it back to the original spot and measured 2.03kHz/urad. Then I tweaked up both pitch and yaw on the ITM and ETM to try to maximize the green transmission, and got 3.2kHz/urad. So we can change the linear coupling by adjusting the alingment; by hand I only seem to make it worse but hopefully the WFS will do better.
I am leaving the green arm locked, and the guardian running the ALS COMM handoff, so hopefully the IR will be resonanting in the arm some part of the night.
