I've reduced the gain of CSOFT pit by 3dB, to reduce the gain peaking at 2.8 Hz, which was seen to cause some glitches (see comments and alog 37705)
The RMS control signal is reduced by about a factor of 2. The attached OLTF was measrued before the gain reduction.
I've updated the FRS Ticket 8591 to reflect these changes. @DetChar -- let us know if this helps!