-- RunaBriguglio - 26 Sep 2010

Verification of flattening repeatability

we want to check the repeatability of the last calibrated flattening command: the procedure is:
  1. print, fsm_set_flat()
  2. print, fsm_load_shape(wanted_shape)
  3. 4D interferogram acquisition
  4. print, fsm_load_shape(wanted_shape)
  5. repeat 2 to 4
this is a "soft" version of the test, as the shell passes from the floating status to the flattened status.

a more "hard" and realistic test procedure is the following:
  1. print, fsm_set_flat()
  2. print, fsm_load_shape(wanted_shape)
  3. 4D interferogram acquisition
  4. print, fsm_rip()
  5. repeat 2 to 4
both have been executed data are stored in the folder: flattening/20100925/re-application_flat_20100923_053430.

load_shapeXX.h5: reapplication of the shape, as in the first procedure

rip_loadXX.h5: reset and load shape of the mirror as in the second procedure

New flats calibrated

matrix used: 20100922_195050/flat_data_641modes.sav        
modes applied 600        
result 20100925_144959/flattening_result.sav        
OPD file before 4d/flattening/20100925/20100925_143727_before_flat.h5        
OPD file after 20100925_144959_flat_600_modes.h5        

matrix used: 20100922_195050/flat_data_641modes.sav
modes applied 620
result 20100925_145354/flattening_result.sav
OPD file before 20100925_145202_before_flat.h5
OPD file after 20100925_145354_flat_620_modes.h5

data to be analized to get residual RMS

Sampling of the new KL base (by Fernando), to check differences wrt the previous

TrackNum modes amp
20100925_150044 0 - 9 400nm
20100925_150206 95 - 104 150nm
20100925_150305 195 - 204 150nm
20100925_150718 394 - 404 150nm
20100925_150825 0 - 9 400nm
20100925_150918 195 - 204 150nm
20100925_151028 95 - 104 150nm
20100925_151125 395 - 404 150nm
20100925_151318 0 - 9 400nm
20100925_151849 0 - 9 400nm

Data analysis on a noise measurement

dataset: 4d/Zcopy/test_noise

it is a set of 1000 images, taken at internal interferometer trigger (28.5 Hz).

NO MODE was applied during the sampling

the data reduction has been performed using the same algorithm in if_redux.pro: given 3 images a,b,c (with +command, -command, +command) the commanded mirror shape is evaluated as MirMode=((a-b)+(c-b))/4

then MirMode(surf) =MirMode/2

data in test_noise folder have been processed in the same way. this gives us informations about the noise level during the nominal procedure of IF sampling (nominal=standard sequence; noise is of course changing on different measurements).

the resulting image (where a mode should be superimposed if commanded) is dominated by a 7nm to 25nm astigmatism. this is calculated removing piston, tip, tilt.

the residual when tip, tilt, piston and astgmatisms are removed ranges from 1 to 3 nm.

a 15x and 51x sampling is also simulated. for the results, see the document.

the astigmatism is oscillating at approx 0.4 Hz

Changes on i4d_opt_intmat

to get rid of the reading error BCU while acquiring IF data with the SwitchBCU trigger, a command of 'wait, 0.1' has been added after each script line sending commands to the switchBCU.

test to verify if this patch fix the problem:

Test mode shell status command N of iteration result
simulation RIP print, i4d_opt_intmat(0,10, /sim) 24+ No errors
Acquisition Set print, i4d_opt_intmat(0,20) 28 No errors
print, fsm_load_shape(xxxxx)
Acquisition Set print, i4d_opt_intmat(0,20) 25 No errors

during a total of 84 acquisitions, the failure was never observed.
Topic revision: r1 - 26 Sep 2010, RunaBriguglio
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback