### OBSDATE ###
20110922

### LOGDATE ###
110922

### OBSERVERS ###
BAW (at home)

### WEATHER ###
Clear and dry

### LOCATION ###
REMOTE

### PROGRAM ###
Test S2 on sky. See if disabling periscope affects "twitching." Get some M giant data if possible.

### SIDEROSTATS ###
n1    s2  

### TARGETS ###
hr 7790 (alf pav) - system check
hr 7848 (phi1pav) - meant to be a calibrator for science

### QUALITY ###
 Average 

### PROBLEMS ###
N1 performed very well with little to no noticeable twitching. S2 "twitched" in both azimuth and elevation. Disabling the periscope had no noticeable effect, so that is unlikely to be the cause of the problem. Given that N1's performance seems to keep getting better, I am getting more and more convinced of a mechanical/grease explanation. I will do some more software work and try to test again tomorrow while logging unusually large timing glitches and encoder jumps if I can figure out how to do that.

Despite the twitching, data on alf pav (magnitude 1.9) was quite easy. Fringes were kept for the majority of time although the S/N varied greatly and rapidly (which may contribute to fringes disappearing on less bright sources). Fringes found near 0 offset.

Phi1pav (magnitude 4.8) was much harder as the twitching interfered with the servoing on the dimmer star. The good news is that the pointing offsets that have now been implemented on the ark controllers helped bring the star back to pavo when it jumped. I did not really have to go get it manually. Fringes were found near 40um offset. Fringes were only visible a small fraction of the time (largely because the south beam kept jumping away), so it took quite a while to get 100 s of integration. As my science target is a similar magnitude, I decided not to pursue it tonight.

S2 pointing is OK. Final offset was -6.25", 14.35". I recall seeing a number as large as 30".

N1 pointing is quite poor. Final offset was 100.5", -31.6". I had to redo the autocol at the start of the night even though I had done it two nights ago. It seems to change by half a degree each time I do it which is worrisome as the other siderostats seem stable for months at a time.

Anyway, automatic pointing logging is now occurring, so I can try to make new models once I get a good build up of data.