Date: Tue, 7 Dec 2010 13:14:48 -0700 (MST)
From: Matt Olmstead
Subject: Re: 2.5m obslog 20101206 (MJD 55537)
Kaike,
	the small offsets are using -5000 to 5000, plus any full hartmanns
taken from the week after the triplet was replaced.  There are about 45
points used in total.  Most of the moves used for calibration were between
-5000 and 5000 with 8 moves greater than 5000 (up to about 9000 steps) but
only 2 moves below -5000 steps).  The fit has always been worse for big
moves but was usally within 0.05 pixels before.  Thanks
Matt
On Tue, 7 Dec 2010, kaike pan wrote:
>
> >
> > 11:10:47   loadCart : ct=16, pl=3811,  sd=A
> > 11:10:53  sop.gotoField:  running
> > Hartmann ff=123505,06
> > 11:14:27  hartSp1= r1=-1476 b1=-5.1 txt=Bad angle !!!!!!! move blue
> > ring -10.1
> > degrees and then run gotoField noSlew
> > 11:14:42  hartSp2= r1=-823 b1=-2.8 txt=OK
> > We observed almost all plates tonight and all of them were in foc.
> > We do not think that sp1 out of focus now.
> > We made full hartmann ff=07,08 and got similar result as
> > short hartmann on GTF. We desided do not move blue ring.
> > VBeside that, we do not have time to adjust blue ring.
> > This plate had a problem on MJD 55534 as guider could not
> > work with it.  Craig recommended us to use STUI 1.0b2
> > version for plates like this with some guider stars located
> > nearby edge of plate.
> > We could see guider stars but background flat was not
> > appropriate and contrast were not good. And we could not
> > apply "Plate" button in guider GUI. Guider was working
> > properly with this plate. There was error message
> > at the bottom status line in guider GUI:
> > "No plate view: shape mismatch: objects cannot be broadcast
> > to a single shape"
> > On STUI 1.02b we could see plate in Guider window properly.
> > Calibration 09,10
> > Science 11-14
> > ------------------------------------------------------------
> >  MJD    b1    r1    b2    r2    Texp      Sum   Plugfile
> > ------------------------------------------------------------
> > 55537   2.9   6.4   2.4   5.6   900.8     17.3    3811-55537-01
> > 55537   2.4   5.6   2.1   5.1   900.8     15.2    3811-55537-01
> > 55537   2.6   5.7   2.2   5.3   900.8     15.8    3811-55537-01
> > 55537   2.4   5.5   2.1   5.2   900.8     15.2    3811-55537-01
> > ------------------------------------------------------------
> >        10.3   23.2   8.8   21.2   3603.2  -->  63.5
> > ------------------------------------------------------------
> > Continue
> >
>
>     It is a little bit strange to me.  I tested blue ring setting
> with full hartmann tests for all
> cartridges. My data suggested that when the red camera is in focus,
> best blue focus ring
> setting are only different by +/- 2 degree for different cartridge.
> Why b1 ring movement
> requirement on other cartridges was only 2 degrees or so, but
> cartridge 16 needed more
> than 10 degree? Is temperature a factor here? I checked air
> temperature records for last
> night, and found the temperature did dropped about 1 to 2 degree from
> the previous cartridge
> (#18) to cartridge 16. But I did not think this temperature variation
> can explain it.
>
>       Science data taken without b1 focus ring movements indicated
> that focus of SP1 is as good as
> SP2 (actual S/N on SP1 are even slightly higher than those on SP2).
>
>       Actually, the full hartmann data (123507-08) suggested that SP1
> setting was indeed within
> tolerance: if we set r1 in its perfect focus, b1 is about -7 degree
> away its perfect focus setting (see
> below full and short data analyses results). I took another set of
> data (123515) this morning, and
> confirmed this.
>
>                       Mean offsets (in pixels/degree or steps)
> Sub-/Full
>   analyses      ff #,       B1,          R1,
> B2,             R2
>   Sub-array    123505-06,  -0.88/17.5,  -1.26/-8055,  -0.37/6.6,
> -0.56/-3584
>   Full         123507-08,   0.20/-4.0,  -0.18/-1152,  0.04/-0.7,
> -0.11/-725
>   Sub          123507-08    0.16/-3.1,  -0.16/-1028,  0.04/-0.7,
> -0.11/728
>
>
>      I am wondering if the scale factor in sosActor still needs some
> twinkles for
> large movements. If I recall correctly (Matt, please correct me if
> this is not the
> case), scale factors are derived from data with a  smaller piston
> movement
> range (+/- 5000 steps from the best focus). In real observations, we
> need move
> pistons as large as +10000 steps, mainly due to different cartridges,
> but also
> temperature variations. Last night, the piston movement from
> cartridge #18 to #16
> is over 8000 steps. My guess is that short hartmann did moved piston
> on both SP1 and
> SP2, but sent a error message too. Therefore, the following up full
> hartmann requested
> much small movements.
>
>     If this a correct guess, can we update scale factors in SoSActor
> with a bigger
> piston movement range data (say +/- 12000 steps)?
>
>     Cheers,
>
>     kaike
>
>
>
>
>
>