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

20" F 4.6 Analysis

  • Please log in to reply
429 replies to this topic

#26 Gordon Waite

Gordon Waite

    Vendor-Waite Research

  • *****
  • Vendors
  • Posts: 431
  • Joined: 11 Jun 2007
  • Loc: New Jersey

Posted 18 February 2020 - 01:22 AM

Hi, Stargazer.  I have been watching your thread and thought I would fire up FigureXP to follow along.  I put in your diameter of 20" and radius of 184" and number of zones as 8.  Then I put in standard zone widths and immediately noticed that they are not close to matching your zonal setup.  Almost everybody I know uses equal-area zones, which makes life easier.  Why did you choose something else, and what are your actual zone inside/outside measurements?  -Gordon


  • PrestonE, nirvanix and FriarPablo like this

#27 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 18 February 2020 - 11:19 AM

other tests are needed... to verify your ke tongue2.gif


  • LarsMalmgren likes this

#28 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 18 February 2020 - 11:50 AM

Hi, Stargazer.  I have been watching your thread and thought I would fire up FigureXP to follow along.  I put in your diameter of 20" and radius of 184" and number of zones as 8.  Then I put in standard zone widths and immediately noticed that they are not close to matching your zonal setup.  Almost everybody I know uses equal-area zones, which makes life easier.  Why did you choose something else, and what are your actual zone inside/outside measurements?  -Gordon

Oh Frack,

 Hi Gordon, my mask is a 7 zone with a central obstruction, I opened the middle to make things simple (I thought)

now I've learned the errors of my way, dayum...

my measurements are as follows

 

zone 1= 2,222 - 4.283, then 5.634, 6.719, 7.652, 8.482, 9.239 and 9.938

 

I will have to correct the error of my ways and see what it shows this evening, sorry guys, my bad,I will do it right this evening and post, hope its not a mess but likely is..



#29 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,855
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 18 February 2020 - 02:12 PM

Ok. You might find this interesting.

I put your 7zone mask data into Foucault Test Analysis and displayed (half) profiles from your posts.  All graphs are at the same wave error scale so progress is easy to see in the graphs.  Graph 2A is the one that had knife reading error for zone 5 with that reading corrected to .071". 

 

20f4r6FTAprofiles1.jpg

20f4r6FTAprofiles2.jpg

20f4r6FTAprofiles3.jpg


  • LarsMalmgren likes this

#30 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 18 February 2020 - 06:44 PM

Hi All,

Redid the setup posted below...

Rick20su.gif

 

tested  with this setup...

Rick20_Surface1.gif

 

basically the same, phew.. really thought I messed thing up,so maybe another trip around

more smoothing but a shade longer to bring out zone back down a tad?.

will do one and set up for test tomorrow.

 

 

 

 



#31 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 19 February 2020 - 07:42 PM

Okay, test done, results are in...

 

Rick20_Surface2.gif

 

 

It may not appear so with this but I have dug a hole in the center of the mirror,

 

inside focus...

rick20image2.jpg

 

sorry, took defusser off for knife edge tests and forgot to install again. but wanted you to see 

the central zone with Ronchi.

Now... how to fix without messing it up, think I may be headed back towards the sphere until

this hole is fixed.

, other then that those curves look real nice do they not smile.gif



#32 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 19 February 2020 - 08:17 PM

On second thought,

 

  Will continue with smoothing strokes using 10" lap for now and see where things stand once hole is gone,why head back to the sphere, fix the hole first and go from there.



#33 dave brock

dave brock

    Mercury-Atlas

  • -----
  • Posts: 2,519
  • Joined: 06 Jun 2008
  • Loc: Hamilton, New Zealand

Posted 19 February 2020 - 08:40 PM

Why is the mirror image not circular? Something's tilted.

Edited by dave brock, 19 February 2020 - 08:41 PM.

  • Stargazer likes this

#34 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 19 February 2020 - 08:47 PM

I only fix holes with full size laps a wax paper stars  pressed into the lap

 

but that’s a hill

 

use the edge of a tool to take it down 2x’s around and test with ronchi to prove your going in the correct direction

 

holes >< pinch in

 

your Hill is <> bowing out, pretty drastically

 

i have it backwards ha ha ha

 

https://www.youtube....9bBNBrZQ0g&t=1s


Edited by Pinbout, 19 February 2020 - 10:11 PM.


#35 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 19 February 2020 - 08:48 PM

What frequency is the ronchi, I can draw it in diffract to see if your not over corrected

 

but the lines look smooth 



#36 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 19 February 2020 - 08:49 PM

Hi Dave,

 It is just the way I took the picture off the computer to transfer here,will try for a more square on photo from now on :)



#37 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 19 February 2020 - 08:53 PM

I only fix holes with full size laps a wax paper stars  pressed into the lap

 

but that’s a hill

 

use the edge of a tool to take it down 2x’s around and test with ronchi to prove your going in the correct direction

 

holes >< pinch in

 

your Hill is <> bowing out, pretty drastically 

Hi Danny,

 the grating is 85, the image is inside focus, when I looked at it outside focus the lines pinched in, I thought that meant a hole or do I have it backward?? please advise.



#38 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 19 February 2020 - 09:37 PM

Hi Danny,

 the grating is 85, the image is inside focus, when I looked at it outside focus the lines pinched in, I thought that meant a hole or do I have it backward?? please advise.

 i had it backwards

 

you need to work out that center... spread it out to make the diffract image with the edge of the tool... 2x's test, repeat to till the black dot spreads out.

 

but this is showing stig and a over all overcorrection cause the outside lines are curving a little more than the diffract image.

 

sitll work on the center before anything.

 

20in f4.6.jpg

 

 

 

24 f4.6.JPG


Edited by Pinbout, 20 February 2020 - 09:35 AM.

  • Stargazer likes this

#39 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 19 February 2020 - 10:07 PM

I was looking at the wrong vid

 

this is how I would spread out the dot...

 

https://www.youtube....h?v=Vr5c3NL9yR4


Edited by Pinbout, 19 February 2020 - 10:07 PM.


#40 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,855
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 19 February 2020 - 11:00 PM

20f4r6FTAprofile31.jpg

Same graph scale as before.

Two humps of over-correction.



#41 Gordon Waite

Gordon Waite

    Vendor-Waite Research

  • *****
  • Vendors
  • Posts: 431
  • Joined: 11 Jun 2007
  • Loc: New Jersey

Posted 19 February 2020 - 11:56 PM

you need to work out that center... spread it out to make the diffract image with the edge of the tool... 2x's test, repeat to till the black dot spreads out.

 

sitll work on the center before anything.

 

Almost hesitate to jump in here.  The center zone on this mirror isn't high, it is low relative to the adjacent zones.  The difference on the Foucault readings between Z1 and Z2 is .060, when it should only be .037.  

 

If you compare the current test with the previous Foucault, the delta between Z2 and Z1 is now .060 where it was .042 before.  And the delta between Z3 and Z2 is now .017 where it was .029 before.  So what happened is quite clear.  He did not dig out the center!!  Instead, he flattened out (lengthened) Z2.  Thus the Z2-Z1 delta increased and the Z3-Z2 delta decreased.  It looks like a central hole, but it was caused by improper work done on Z2.

 

The "delta's" always tell the true story.  Just looking at the graph, it looks like he dug a center hole.  But the delta's show the error was made lengthening Z2.  The mirror is already a little over-corrected overall, so you don't want to work Z1 in a mistaken effort to remove a non-existing hill.  That will increase the overall correction even more, and make the situation quite a bit worse.

 

Gordon


  • Stargazer and nirvanix like this

#42 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 20 February 2020 - 09:57 AM

attachicon.gif20f4r6FTAprofile31.jpg

Same graph scale as before.

Two humps of over-correction.

 

the ronchi gram does show your doing the ke correctly when you look at the percentages of correction for each zone...

 

you still need to spread out that dot in the center...


  • Stargazer likes this

#43 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 20 February 2020 - 11:23 AM

Hello All,

I treated it as a central hole and did three rounds of parabolizing and  smoothing W strokes with the 13" lap, did a quick test with Ronchi to confirm that all is going in the right direction,hole is getting smaller.

As expected the overall correction on the mirror is and will change but correcting the center is my concern at this 

moment will work on correction again once this is completed.

If nothing else I have given myself the confidence to know I can do this now, feels good too after all these years of

working this project.

Thanks

 

 


  • PrestonE and nirvanix like this

#44 Gordon Waite

Gordon Waite

    Vendor-Waite Research

  • *****
  • Vendors
  • Posts: 431
  • Joined: 11 Jun 2007
  • Loc: New Jersey

Posted 20 February 2020 - 06:13 PM

Hello All,

I treated it as a central hole and did three rounds of parabolizing and  smoothing W strokes with the 13" lap, did a quick test with Ronchi to confirm that all is going in the right direction,hole is getting smaller.

As expected the overall correction on the mirror is and will change but correcting the center is my concern at this 

moment will work on correction again once this is completed.

 

Stargazer, I am interested in hearing your reasoning for your strategy of working on the problems at the center of the mirror instead of working on the problems in the outer zones of your mirror.  Some ideas to think about...

 

1. If you were to get the Z2-Z1 delta to be perfect in your next session, why would you think you could perfect the other 5 deltas without changing the center?  What would be your strategy to perfect the rest of the mirror without changing the center two zones?

 

2. If you look at the sum of the inside three deltas, they sum to 109 out of a perfect 111.  If you look at the sum of the outer three deltas, they sum to 120 out of a perfect 108.  So the outer parts of your mirror are off by 12 parts in 108, while the center parts are only off by 2 parts in 111.  In general, why would you decide to work on the part of the mirror that has the least problems?

 

3. Also, given that is is far easier to perfect center zones rather than outer zones, strategies that work "outside->in" are generally preferable.  Given the need to define a strategy, it is almost always easier to perfect the outer zones, then perfect the inner zones without changing the outer zones, than it is to correct the inner zones and then correct the outer zones without changing the inner zones.

 

4. Consider what you should have learned in the session where you "dug the center hole."  Were you trying to work the center in that session?  No.  Instead, you said you were actually trying to change the outer zones.  But by doing that, you had the unintended consequence of ruining the delta between Z2 and Z1 at the center of the mirror.  There was a result that you did not expect, thus a process that you did not understand.  So if you go down there immediately and "fix" the center, how are you going to then work the outer zones, this time without changing the center deltas?  Do you understand what was wrong with your technique on the outside of the mirror that caused the bad result near the center?  And do you know a better technique to avoid that in all of the upcoming sessions that will be needed on the outside after you go to the trouble of improving the center?

 

Enjoy the process!  -Gordon


  • Stargazer, Pierre Lemay, LarsMalmgren and 2 others like this

#45 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 20 February 2020 - 06:58 PM

Hello Gordon,

 

 My reasoning for working the center, remember that zone one radius is 3.3% so inside that inner 2.222 of my zone 1 slit is a

fairly large area of mirror not really in the equation here (the hole) that looked absolutely terrible under Ronchi that I with my limited experience thought needed fixed first, kinda opposite a edge problem.

 

I have had no real problem working the outer zones to where they need to be (maybe just luck on my part) but the central part

of the mirror has been a real pain to me.

 Now maybe I just missed something or read the hill wrong all along and dug the center out trying to fix the hill that was not there so to speak but you are absolutely right. I do need to understand what caused the center issue going forward.

 

I have done now six trips around with the parabolizing and smoothing W well trying to keep correction close, but only Foucault will tell me for sure once I am done.

 

For what it is worth..the hole is getting smaller and the bands appear to be staying close, I figure another three or four trips around will fix the center,then I shall try to understand what I did wrong, learn from my mistake and have a go at it again. 

 

you see Gordon, I have it backwards, the hard part I find easy and the easy part I find hard, just need to find that compromise to bring the two together smile.gif

 

B.T.W.  You can call me Rick, been a subscriber to your channel for years now, thanks for that.

they say mirror making is a learning process, well I am still learning but learning I am.

 



#46 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,371
  • Joined: 22 Feb 2010

Posted 20 February 2020 - 07:41 PM

lets see another ronchi gram. no one was talking about a hole until you posted the ronchigram.


Edited by Pinbout, 20 February 2020 - 07:42 PM.


#47 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 20 February 2020 - 09:18 PM

Danny and All,

 Here is Ronchi image after 10 trips around with parabolizing and smoothing W stroke to

correct the hole and try to keep parabolization close to where it was, center looks much better

not as dark and smaller then before but can use some more work, sorry for image quality best

I could do for now...

 

ronchi1.jpg

 

going down to do a Foucault test to see where things stand now before I do more

 



#48 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 20 February 2020 - 10:07 PM

Okay All,

Foucault test is in..

 

Rick20_Surface3.gif

 

no where near as bad as I would have thought it would be after 10 trips around ( over an hours worth of work) trying to repair the centeral zone and keep things close to where they where everywhere else, I must be doing something right.

 

Like I stated previous post, the center is much better, not perfect but better, should I continue going around the barrel or start correcting surface.. will think about it and decide tomorrow.

 

I am open for points of view and expert opinions if anyone cares to enlighten me because I am not even sure what to do next smile.gif.

Thanks



#49 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,855
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 21 February 2020 - 12:11 AM

20f4r6astig.jpg

Hint of astigmatism there.  It shows in the other Ronchigrams too.   Think about what might have caused that.

 

Besides the usual stuff like mirror back flat, flat turntable and pad between mirror and table, frequently clocking the mirror on the turntable, working same stroke for full turns of the mirror...

I haven't done much stroking Ws with a turntable.  I imagine that W strokes could have some tendency to make astigmatism if one wasn't careful about the ratio of table rotation speed and rate of stroking the W pattern.  It seems to me that at some speeds it would be possible that parts certain parts of the W near or far from mirror center could sync with the mirror rotation in ways to always put the lap at the same part of the W at the same mirror sector.  Very slow table rotation rate so that many full W patterns are done in each rotation would probably reduce the risk.  And maybe a high table rotation rate with near full rotation each stroke would work too, but might be difficult to control the lap by hand at high speed and the speed might not be good for making a smooth surface.

 

Fixing that should now be your first goal.  I think you will have to forget the parabola until that is taken care of.  I think it will be necessary, or at least easiest to go to a sphere, doing all those things to avoid astigmatism and hope it polishes out quickly.

Maybe someone else can tell you how to fix that without straying too far from the parabola.

 

I suppose there is some chance the mirror is bent on the test stand.  I don't think so.  Hoping I'm wrong about that and it's just test stand induced, or that somebody can tell you a quicker way to fix that.


  • Stargazer, LarsMalmgren and Pinbout like this

#50 Stargazer

Stargazer

    Messenger

  • *****
  • topic starter
  • Posts: 446
  • Joined: 30 Jun 2006
  • Loc: NewBrunswick, Canada

Posted 21 February 2020 - 06:03 AM

dogbiscuit,

 

  The table does not turn when I do the W strokes, its like walking around the barrel, do a side to side swing and back, turn mirror on table and turn lap on mirror.

will keep an eye on that next time mirror on stand and evaluate more.

looks like I will continue to fix central area after work and see how it goes.

Thanks




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