Not logged in
Mon Mar 16 01:20:40 2009
MCWFS noise filtered on the SUS-MC
, we noticed that the IOO-WFS system runs at 2048 Hz and sends its signals to the MC SUS
systems which run at 16 kHz.
There is no upsampling filter
or anti-imaging filter.
So, I've implemented an RLP666 filter as FM1 in the SUS-MCn_ASC(PIT/YAW) filter banks. This is like a 4th order
Cheby low pass with a low Q notch at 2048 Hz to catch the first image.
The attached PNG shows the ASCPIT_OUT signals before and after the filter is implemented. As you can see, the
big aliased spikes are gone. The reason that MC2 is different from MC1/3 is that they have a hardware 28Hz low pass
and MC2 doesn't. So MC2 had a 28 Hz low pass in software already to match the actuation phase between all the MC
mirrors. The apparent power law noise floor from 40-300 Hz in MC2 is not real - just the Hanning window tail.
And yes, it has been this way for several years and none of us noticed. It remains to be seen if this was causing
any noise in the MC coil drivers via slew rate limiting.