Sheila, Dan, Evan

ETMY ring heater on, operating at 22 Watts input power

I turned on the ETMY ring heater with the same settings as the ETMX one, 0.5 Watts on the top and bottom.  Since then we've been locking with 22 Watts, we haven't had any long stretches but that is probably because of other activities.  

A2L tuning for test masses L2

I injected an excitation at 12 Hz into L2 of each test mass pitch and yaw, and tuned the A2L coefficents until the coupling became bilinear.  The gains that I ended up with are shown in the attached screenshots.  I also chanmged the ramp times on all of these filter modules and accepted all of these changes into SDF.  The gains may need to be retuned in the future but they seemed to stay stable over several locks today.  Screenshot attached. 

ETMX roll mode could be damped with AS 45Q Yaw

With the improvement in the sensitivity below 20 Hz, the ETMX roll mode was again more than an order of magnitude above the DARM noise floor.  We cannot improve anymore by increasing the damping gain using the AS 45Q pitch signals, the mode is now in the noise of those signals.  However, the AS 45Q YAW signals both see this mode with an SNR of better than 100.  Durring the vent we can expand the matrix used to send ASC signals to the suspensions so that we can try using YAW.  

SRCL FF

With the reduced noise from A2L the coherence with SRCL increased, so I made another attempt at SRCL FF.  Evan called and told me about a filter that he made last night, which is intended for use in feeding SRCL to ITMY L2.  I used this filter, and found that with a gain of -0.5 this works well. The second attached screenshot shows the coupling from SRCL out to DARM IN with no feedforward, and with the feedforward engaged.  The coupling is reduced by 6dB at 20 Hz, and more at ghier frequencies although the measurement lost coherence.  The last attached spectrum shows the sensitivity improvement due to both A2L and SRCL FF.  The calibration may be off here because the ESD drive strength changed.

The bad news is that we seem to loose lock a few minutes after engaging this feedforward, and we are having a good deal of difficulty relocking today. 

New nominal state for guardian,

I added a new state to the gaurdian, which does nothing other than check for lock.  It is called NOMINAL_LOW_NOISE, the index is 600, and I hope that we can add any additional states before it from now on, to minimize confusion.  Also, it is handy to have a state which does nothing.  

No inital alingment needed for 24 hours.

I came in this morning to find the IFO still locked, it had been locked for nearly 12 hours.  After it dropped lock (probably my fault) I didn't do any inital alingment, and I haven't done it since, so the alignment has not been done for about 20 hours now and things seem OK.  I think that over the engineering run we were probably doing inital alingment much more often than necessary, it would be good to keep track of how often we really need to do it.  

Locklosses in switching to DC readout

 We've had several TR_CARM locklosses and a new type of lockloss that seems to have shown up in the last 2 days, durring the switch over to DC readout. Dan and I commented out the DARM boost that gets tured on as the first thing in the DC_READOUT state of the ISC_LOCK guardian, we are engaging it by hand after the transition for now.  unclear...