Jenne, Matt, Sheila, Evan, Stefan, Lisa
Tonight we reworked DHARD YAW. We started with a new filter desgin, which avoids inverting any thing in the plant (our old design had just the 2-3Hz resonances inverted). The first screenshot shows the new control filter (which is in FM6). This filter gives us significantly better phase margin than we had before. We engage it in the state DARM WFS (right after transitioning to RF DARM) with a gain of 50, and leave the gain at 50 as the optical gain increases until the state CARM_5PM, where we reduce it by a factor of 2. This is because the sensor noise is too large at high CARM offsets to run with the final bandwidth. The attached screen shot shows the open loop gain at some different arm powers. We are not currently using a boost, but we think that we should be able to turn on the soft boost in FM3 (boostMS).
We also restored the DHARD PIT gain changes durring the CARM offset reduction that we found alst night, they seem to work fine.
After a half hour at 24 Watts with the new loop on and no ITM oplev damping, we saw a pitch instability ring up (about 6:55 UTC July 31st). This was fixed by turning back on the ITM PIT oplev damping. The oplev damping is back in the guardian for now.
This is to remind us that we haven't yet engaged a cut-off for PITCH. I made a new EPL33 filter to replace the current unused DHARD cut-off (ELP5,10) , such as this new cut-off can be engaged with the current higher DHARD ugf. I haven't loaded this filter yet.
