Pointing corrections jun10
09jun10
A broken beam on the platform was fixed on 24mar10.
We were then limited to a za range of 3.5 to 12 deg za until the
reinforcement of the beams were completed. x102 calibration data was
taken from 24mar10 thru 03jun10 (morning) to check the telescope
pointing. These measurements were also limited to the 3.5 to 12 deg za
range.
The plots show the pointing errors
mar10-jun10(.ps)
(.pdf) (24mar10 thru 03jun10
morning)
- Page 1: The az,za coverage for the data.
- Page 2: The pointing errors
- top,2nd : za error vs za and az
- 3rd,bottom: az error vs za,az.
- The rms errors are: za 9.73 asecs, az 10.38 asecs
- You can see the 1az term on the plots vs az.
- Page 3: the errors and the fits.
- The errors were fits to yfit=c0 + c1*cos(az)
+c2*sin(az)+c3*cos(2az)+c4*sin(2az)+c5*cos(3az)+c6*sin(3az)
- Top,2nd: the za errors and az errors with the fits over plotted
in red.
- 3rd: (data - fit) vs azimuth. Black is az errors, green is za
errors.
- Bottom: (data-fit) vs za.
- Page 4: Replot errors removing fit.
- This is the same as page 2 with the error fits removed.
- The rms error is now: zaErr=2.86, azErr: 3.16.
The results of the data fitting were:
|
AMPLITUDE (asecs)
|
PHASE (deg)
|
constant
|
|
|
1az
|
|
|
2az
|
|
|
3az
|
|
|
The 1az term for az errors and za errors both have
close to the same amplitude and their phase difference is about 90
degrees. This is a tilt in the platform. The actual values used for the
model update are listed below (the model uses cos,sin rather than
amplitude,phase)
-
Values to SUBTRACT
from current model
Model Term
|
az Err
Asecs
|
za Err
Asecs
|
constant
|
-3.68
|
1.93
|
cos(az)
|
-7.93
|
-9.93
|
sin(az)
|
8.43
|
-7.94
|
cos(2az)
|
-6.41
|
-0.31
|
sin(2az)
|
-0.94
|
-1.65
|
cos(3az)
|
-3.03
|
-0.59
|
sin(3az)
|
-0.81
|
-1.63
|
Data taking after the model was updated:
The model for all receivers was updated on 03jun10
17:00. Data was taken on 03,04,05,09 jun10. When the model was
updated, i accidentally used the wrong value for the za sin(3az) term.
I used -7.94 (the 1az term) rather than -1.65 (the 3az term). This gave
a large 3az za error. This was corrected on 09jun10 at 10am.
The plots show the pointing errors
after the model update (.ps) (.pdf)
- Page 1: The az, za coverage.
- Page 2: the pointing errors with the za sin(3az) error included
- top,2nd: za errors
- 3rd,bottom: az errors
- the 4.82 za error is from the large 3az term from the incorrect
model parameter
- Page 3: the pointing errors with the correct za sin(3az) model
term.
- I went thru and removed the incorrect model term and added in
the correct value. The plots are what we would have measured if i
hadn't screwed up updated the models.
- the za error mean and rms are: -1 +/- 2.76 asecs
- the az error mean and rmsa re: 1 +/- 2.81 asecs
- Notes:
- Tiedown tracking was turned off when this data was taken (the
temperature was above 75F). The platform was at 1256.16 to 1256.18
feet. This is off by -1.8 to -1.2 cm.
Summary:
- The pointing errors mar10 to 03jun10 were
- zaErr: 9.73 asec rms
- azErr: 10.38 az rms (with 3.7 asec offset).
- This is mainly a 1az tilt of the platform.
- After fitting the 1,2,3 az terms to the errors, data taken after
the model update(correctly) showed rms's of
- zaErr 2.76
- azErr: 2.81 asec
- The changes were installed correctly in all of the receiver
models on
09jun10 10:00 ast.
- The pointing errors have been corrected, but the platform is
probably still tilted. High frequency gains can be affected by this.
- Need to do some az swings with the tilt sensors to see if we
can measure the actual tilt.
- After the beam reinforcement (jun,jul 10) we will need to redo
this.
processing: x101/100602/chkpnt.pro
page up
home ~phil