Bas pointed out that drum mode tracker is not useful anymore because of the frequency shift (logbook #41732). But the temperature change in WE (logbook #41771) shifted the drum mode frequency to higher frequencies, and TCS_ETALON_XX_DRUM2_FREQ luckily could track this shift.

I did a similar analysis done a year ago (logbook #39039) to see the temperature dependence. Figure 1 is the plot comparing drum mode frequencies with NI,WI,NE,WE ring heater temperatures. You can see that TCS_ETALON_XX_DRUM2_FREQ follows ENV_TCS_WE_RH_TE. The temperature dependence is 0.85 Hz/K, which is consistent with a measurement done a year ago (logbook #39039). Other drum mode trackers are currently useless. We have to change the code to account for the frequency shift after the monolithic suspension installation.

Figure 2 is the raw time series data used in this analysis. I only used the data when the interferometer is "locked" (LSC_B7_DC > 250).