Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Plop and pt load

  • Please log in to reply
2 replies to this topic

#1 hakann

hakann

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1742
  • Joined: 06 Aug 2015

Posted 27 May 2019 - 03:31 AM

The Plop program show us the points, but it is a direct nail hit or do has a std pad diameter programmed ?
( if so, what diameter is that ?
I’mean load is different from a 1 mm pt to 25 mm pt.
Issue is that even if bigger is better but then we don’t know if we are at pads centre or if it is on one side of pad or two.
This will has a difference depending on cell.

I had a FEM/FEA done and idea was a 6 pt with big pads ( equal side triangles with 5 mm contact pads at CC 52 mm ) but actually it is a 18 pt with a balance bar between.
Radii factor was 0.75 ( 458 mm dia )
This triangles in 10 mm CF is 65 grams each.

Readings was in FEM for mirror data was around 6 nm RMS error.
From AMT I got it was really bad, so better use the regular Plop as it could get down 6 times lower vs bigger non equal side triangles. ( Sag was told not to worry in thin aluminum, as it sag ’equal’ )

Now seen the new idea of a floating swing balance arm, rest on 2 bearings, I has look back to the FEM readings.

Between this I got feed back that over golden 5 nm RMS or 10 nm RMS is really nothing in visusl use, so many older 18” used a 9 pt.

One 18” I heard of ( factor 10;1 ) at f/4.5 had a 9-pt ( cloosed back ) but they was not happy so they builded a 18 pt whit a beefy steel frame ( open back ) and very thick fiber/CF triangles. ( A nice solid cell.
It got better but what they learned was not the cell, it was heat in tube from the closed back.

#2 Arjan

Arjan

    Viking 1

  • -----
  • Posts: 941
  • Joined: 21 Jan 2009
  • Loc: Netherlands

Posted 27 May 2019 - 07:42 AM

The PLOP model assumes point forces.
Default all forces are taken equal, and it is up to the ATM to make a support that actually provides this. You can also make the forces unequal, which results in a different optimization of the locations and also a different cell construction.

The common way is to make a whiffle tree, as you know, and the art is in guaranteeing that these forces really are what the model works with.

Introducing finite size supports will complicate the modelling, PLOP does not include such feature.

#3 MitchAlsup

MitchAlsup

    Soyuz

  • -----
  • Posts: 3794
  • Joined: 31 Aug 2009

Posted 27 May 2019 - 08:20 PM

This is not the first conversation we have had on this subject.

 

And while none of us has access to a big FEM program, or the necessary time; we generally come to the same conclusion:: the point forces used in PLOP over-represent the surface error on a cell using actual pads. That is if you build the cell to the exact dimensions given in PLOP, and with very low friction; your mirror will be distorted by less than what PLOP indicates.

 

We see this (a bit) in the Z88 runs where most of the time Z88 surface error lying flat is a bit lower than PLOP.

 

{Now if there were a way to intercept the text between PLOP and Z88 so we could install proper edge supports.....}




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics