### OBSDATE ###
20120912

### LOGDATE ###
120913

### OBSERVERS ###
YK, BAW

### WEATHER ###
Clear
RH (open): 54.0% (SUSI), 49.0% (ATCA)
RH (close): 53.0% (SUSI), 52.0% (ATCA)


### LOCATION ###
SUSI

### PROGRAM ###
Test observation (with PAVO only).

### SIDEROSTATS ###
n1 n3  s1  s3 

### TARGETS ###
alf Pav - N3S3
alf Gru - N3S3, N1S1
tau Sco (took long time to get those fringes, settled with ~20s) - N3S3
rho Sco (could not find fringes) - N3S3

### QUALITY ###
  Quite bad (no/almost no useful data

### PROBLEMS ###
N3S3 (15m):
- Position of fringes were very unstable according to the waterfall plot. Suppose 1 refresh unit of the waterfall plot is t sec. Once the fringes are found, t sec later they were gone.

- We considered and rule out the following factors:
(1) optical alignment - we have recently aligned PAVO, MUSCA and all the usable siderostats on site and the numbers (pupil positions, etc, read the upcoming trip report) suggests that it was done properly.
(2) cart vibration or cart-related factor - we observed the star alf Gru during its transits (rate of change of OPD was 0 or less than 10um/s), the stability of the fringe positions did not change.
(3) TT lock - the TT flux is strong ~4-5K (N) and 3-4K (S) with alf Pav and alf Gru. We have also optimized the focus. We could not rule this out completely, so took data with one beam in the "Fringes" box at a time to check the stability of the pupil position in post-processing.
(4) Software issue - we made the following tweak to PAVO server to try ruling this out:
(4a) "setas 1 0 0" - this stopped PAVO from doing an auto search when fringes were lost
(4b) "dtime 4" or "dtime 5" - this set the group delay (gd) servo gain to a small value and made PAVO avoid large gd steps.

N1S1 (5m):
- Switching to this baseline and with the same settings, there were two noticeable period of times when the fringe positions:
(1) changed abruptly like the case with N3S3. It went away completely in t sec and
(2) changed less than ~5um (scaled from the waterfall plot) in t sec and its positions stayed within +/-15um most of the time.
- Based on the descriptions above, assuming t sec is 250ms or 50 frames of PAVO cycle time (~5ms) and using simple atmospheric turbulence maths the r0 during the good times was ~4cm and the coherence time (tau0) was ~2ms.
- Using the same assumption, the coherence time during the bad times (applicable to times when N3S3 was used as well) was ~1ms or less.
- The numbers are in the regime where one would expect atmospheric turbulence effect have on the seeing condition but we are unsure of the assumption of t.

Does that mean the poor tracking we saw tonight and many nights before are due to bad seeing?


NB:
- Fringes found at -660um with N1S1.