: power up of DX and SX systems. RR cubes are mounted back on both sides.
We setup the SX AGW to be used with the internal source, FW1 has 90-10, FW2 QC mirror.
Then we center the QC putting the Z stage at 25mm, so 1" spot:
We change FW1 to dichroic 600-1000nm:
We find that a better position for the cube rotator is 91.45deg instead of the 92.2deg of the board setup.
These IMs are very different from the one recorded in December, we have almost a factor 100 stronger signals... Could be due to the use of different pyramid modulations, since the QC signals are re-normalized on Truth sensing modes to get the proper orientation wrt the ASM tip and tilt modes. So we change the SW, now thruth modes are used just to identify the direction of the modes, not their amplitude.
New board position for Z=35mm, in focus on IS:
|| wrong IM
We build a combined rec from the last QC rec:
- COMBO REC (TCoG48 + QC): 20160310_204000
That REC was acquired with LUCI rotator at 92deg.
We test in close loop using the ARGOS CalUnit
. We found a problem of gain: the REC seems a factor 100 too big! We build a new combo REC dividing the 20160310_223600 by a factor 100:
- COMBO REC (TCoG48 + QC): 20160310_213600
We test it in closed loop and it seems to work...
We start measuring the hot spot on CCD47 for QC:
White light source is set to 65 steps. This is equivalent to a mag 13.5 on PYR.
We spin the LUCI rotator while we keep closed the TT loop on QC. We track with the pupil derotator and we find the LUCI rotator angle that minimizes the pointing difference between the 2 sensors. This value is 65deg.
We measure the hot spot on CCD47 when the spot is centred on PYR: [32.6; 40.4].
A possible procedure to co-point the 2 sensors at a given rotator angle is:
- close the TT loop on PYR
- spin FW2 to minimize the QC signals (= difference in 4 channel fluxes)
- offset the pyramid modulator to null the QC signals
- close TT loop on QC and measure the residual TT from truth modes
We keep the TT loop closed on QC and we measure the piezo offset needed to null the pointing difference:
We then try to move FW2 and to minimize the pointing offset at 245deg.
In parallel to the SX QC calibrations we try to align the CalUnit
to LGSW on DX side. First alignment was done yesterday when Gustavo was looking at the Entrance windows flange of the sensor. Today we refined this position trying to reduce the overall coma. We ended up to a limit position were the decenter Y of the hexapod was hitting the limit:
| DX Hexa
At the end of daytime WG moved the upper arm of the CalUnit
gusset by ~1mm with the aim to reduce the hexapod decenter Y.
On top of the alignment work we tryed also to acquire a 10 modes LGSW IM, but just one LGS at time could be feeded inside the WFS. We build a just for the Blue pupil using the old KL_v16 as modal basis:
We tryed also to close the loop using this REC but the tilt was diverging...
- 11 Mar 2016