Similar to the work done on the BSC-ISI's, I've been working on tilt decoupling on the HAMs. So far I have most of the measurements needed for the the output HAMs, I just need a couple more measurements on the input HAMs to finiish. The first attached plot is the Z to X/Y coupling for HAM6. For this data, I took the Z tf at the isolation bank input with the ISI isolated with high blends (some measurements had sensor correction on, some off, it doesn't seem to matter much, off is probably best though). I then inverted the gs13 response (to get the units to nm/s), converted to nm and then divided by -g/w^2 to get the effective tilt. Doing this let me calculate the magnitude and sign decoupling element by getting the average of the Z to X/Y tf where the phase was either 0 or 180, usually over 10 to 80-100 mhz. I was also able to avoid getting the RX/RY tfs as I done for the BSCs by thinking a bit about the RX/RY to Y/X tfs. These transfer functions go as -g/w^2, so their signs are also possible to figure out, and after some thinking I realized I could model them by using (essentially) a +1 for the RX to Y tf and -1 for RY to X tf (something Jeff mentions in his eligo era log here). The second attached plot shows the improvements I was able to get for HAM6 Z to RX/RY decoupling.

It's interesting that while the Z tilt decoupling clearly needed to be done, the X/Y to RX/RY coupling does not seem so bad. I have taken several X or Y tfs, but all of the coefficients are pretty small (on order of .1%). In fact the coefficients are small enough that the X to RY or Y to RX coupling is hard to seen. The third attached plot is the HAM4 Y to RX/RY measurement. The blue trace is the Y to X tf and gives a coefficient of about .0015 for the Y to RY element. The red trace is the Y to Y tf and is dominated by the Y to Y closed loop tf. The first couple of frequency points around 5 mhz indicated a Y to RX coefficient of something like .00005. I don't fee like waiting around for a measurement long enough to resolve that any better. This seems to be typical of the "beam-line" coefficients for the HAMs.

So far the elements I've installed are:

HAM 4     Z        Y

    RX     .01    .0015

    RY    -.01

HAM 6     Z   

    RX     -.0181

    RY      .0178

I have a few more calculated coefficients, but I'm going to wait for a chance to do a before/after measurement to make sure I actually make things better.