[Takahashi, Ushiba]

Abstract:

We discussed how to unify the control schemes for the inertial damping of Type-A and Type-B suspensions.
Basically, we agreed that the inertial damping control scheme should align with the current Type-A control scheme.
One change to the Type-A control filter is increasing the pole frequency from 0.1 mHz to 0.5 mHz pole to convert acceleration and velocity to displacement.
This modification reduces the large transient signals while keeping small phase rotation above 30 mHz.

Detail:

Since SRM is not used for a long time, control strategy of SRM is completely different from the other susensions, so we started to unify the control strategy so that we can easily understand it.
For the first step, we discussed the control scheme of inertial damping controls.

Currently, each filter related to the inertial damping control is used for the following concept.
ACCINF: Calibrate raw geophone signals into velocity signals with 3-mHz second-order high pass filter.
FLDACCINF: Calibrate FLDACC local feedback signals into veocity signals with 0.1 mHz pole and 10 mHz second-order high pass filters.
ACCBLEND: Calibrate velocity signals from geophone and FLDACC into displacement with 0.1 mHz pole. Setting crossover frequency between geophone and FLDACC at 100 mHz including phase compensation and inter-calibration.

On the other hands, SRM filters are designed as follows.
ACCINF: Calibrate raw geophone signals into displacement signals.
FLDACCINF: Calibrate FLDACC local feedback signals into displacement signals.
ACCBLEND: only inter-calibration.

If we use SRM-type filter setting, we have no chance to measure the acceleration/velocity signals at DQ channels, which is very incomvenient for the suspension health check.
So, we decided to use control scheme used for Type-A suspensions as a default concept.

In addition, we slightly moodified the filter setting for Type-A suspensions; changing pole frequency from 0.1 mHz to 0.5 mHz, which is used for converting acceleration to velocity or velocity to displacement.
Owing to this modification, transient signals when starting suspension operation becomes small while maintaining the phase rotation at 30/50 mHz sufficiently small.

After the modification of the filter, I checked if the inertial damping control can be stably engaged or not.
Figure 1 shows the comarison between the spectra before (blue) and after (red) changing pole frequency.
Though I have no idea why the spectra around 0.25 Hz and at high frequency are significantly reduced in red, no significant degradation can be seen, so the modification itself should be fine.