20220430_DX_Day DX AdSec Calculate Influence Functions, Day 11

Processed Set_4 from 20220425 where shapes (TT, Focus, Coma) were NOT removed in the 4D processing

An email was sent:

Runa, All,

  I reprocessed a set of 4D IF data collected on 20220425 with no shape

However, the processing of the data showed the same problems as
the previous analysis, a poor tweaked mask, and bad (large) fitting
errors of the KL modes.

The thought was that by removing TT, Focus and Coma from
the high order mirror modes that I was removing vital information.  I will
look at all the individual hdf5 phase map files and check that there was
no shape removal (Xianyu provided me the the header variables for
the shape removal coefficients).

I looked at most of the combined OPDs (these are the averages of
the 125 to 35 push pulls (depending on mode number) of each mirror
mode that is then used to build the interaction matrix and they looked
Great.  No phase discontinuities, no unexpected bumps, etc.

Is it possible that there was a jumping actuator, or maybe an actuator
that would push but not pull while the mirror modes were being applied
on the DX AdSec?  The result would/could be that the subtraction of
a push from a pull phase map would have the mirror mode shape plus
a “feature” of the bad actuator.

Several indications that there may be a bad actuator may be:
  - The tweaked masks images
  - the last eigenvalue image
  - the KL fitted mode images

Suggestions or ideas?


  • inner_s4.jpg:

  • last_eigenmode_s4.jpg:

  • kl_shapes.jpg:

-- DougMiller - 01 May 2022
Topic revision: r1 - 01 May 2022, DougMiller
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