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

Figuring first mirror - 8" f/6

Mirror Making ATM Beginner
  • Please log in to reply
296 replies to this topic

#51 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 19 October 2021 - 06:56 AM

Wait before working.

 

Looking at your lap... It needs to be trimmed.

I am marking some things on your image of the lap to show how to improve it when you trim.

Not sure about the temperature. The roughness, which isn't all that bad, is almost certainly from working with the lap that needs trimming.

 

Don't trim until I post the marked up picture.

There is a small section of channel that is about the right depth and width.  I'll mark that area with a green circle. The idea will be that when trimming you try to make all the channels as much like that all over the lap.

 

You could be only a few short (~5 minute) sessions from a very good parabola.

 

As for heating the lap and mirror, I don't do that but I work in a temperature controlled space.

I think for one session you should continue without warming the lap and mirror.  After trimming the lap rinse the pitch chips off and apply slurry and press. 

 

How much have you been pressing until now?  How much work have you done since the last time you trimmed the lap?  I can get a pretty good idea how hard the pitch is from that.  For now I will say that you have a pretty good surface so the pitch is pretty good.  With the lap trimmed it should work better.

 

Picture coming... not to long from now.

I have had issues with the trimming and rechanneling of the lap, so I try to avoid it. I reckon that my pich is a little on the hard side, and hence the problems that I see due to trimming and/or rechanneling. Temperatures have gone down a bit so perhaps what was a good smooth lap a few weeks back is not as smooth.

 

Here's a photo of what happened on the mirror the last to last time I trimmed the lap. This was on 2nd Oct. I have done trimming and rechanneling once more after that. Not quite sure what I did to avoid these sleeks.

 

IMG 20211019 171206

 

Your tester look nice.

And very practical... you can polish your shoes while you test.  :-)

Hahaha. I quit wearing leather and hence won't need it for polishing unless I buy synthetic...I am too lazy so I just wear sneakers with suit. cool.gif cool.gif



#52 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 19 October 2021 - 07:04 AM

You have done good not beveling the edge of the lap. That is why you aren't having trouble with tde. Reducing the diameter of the lap to less than the mirror diameter is a major cause of tde. Some of your lap defects also can cause edge troubles.  Getting you channels opened up, press well and when you work the edge might get even better than it is.

 

 

I have marked some, but not all of the problem areas.

Three red ellipses show some of the worst examples of closed channels.

All except a few short segments of the channels have narrowed too much.

Those sort of things cause bad thing on the surface.

I'm surprised you mirror is not a real mess.

I think you pitch hardness is ok.  If it was too hard the mirror would be much rougher and the curve would have ugly zones.

 

KnightSkyLap.png

 

The red lines don't show how wide to make the channels, they show which side of the channel to trim to open it.

choose the side that will keep as much pitch as possible on the small pitch facets and removes some of the pitch of the fat facets, or that moves a channel toward equal spacing of the channels.  I didn't mark all of them, use your judgement to improve the channel grid within the limits of the suggested channel width.

 

The channels are trimmed to about the depth and width of the channels in the green ellipses. Make them like that all over the lap.  Yes I understand the channels were probably all the same size when you started and the outer ones have not closed as quickly.  When you trim try to make channels uniform over all the lap, and evenly spaced.

I know you can't get it perfect in one trimming.  When you trim plan how you trim so it gets better.

And I understand the lap was in better condition for much of your work and maybe only a little while like in the picture.  Don't work at all with the channels in the condition as in the picture.

 

You will probably finish this mirror without trouble before you get everything perfect with the lap.These are details that can make things about keeping the lap in good condition and action more predictable.

 

Trim channel more frequently.  Less trimming will be needed to remove less pitch from channel walls. It is easier to correct channel creep.  

Less trimming more frequently.

My old saw...  The mirror maker's secret sauce is the lap.

 

 


  • brave_ulysses and Knight Sky like this

#53 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 19 October 2021 - 07:12 AM

@dogbiscuit, here's a proper top-down photo of the lap.

 

IMG 20211019 173745


#54 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 19 October 2021 - 11:55 AM

Redone lap, @Pinbout style, just that I have done a terrible job grin.gif

 

IMG 20211019 221750


#55 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 19 October 2021 - 04:23 PM

now press well, extra time before the next work session.


  • Don H and Knight Sky like this

#56 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,042
  • Joined: 22 Feb 2010

Posted 19 October 2021 - 04:36 PM

I would take sometime to build the scope and star test it.  Even if it’s not “finished” you’ll learn some stuff about star testing while comparing with your ronchi to the star test. 


  • Don H, Jeff B and Knight Sky like this

#57 Jeff B

Jeff B

    Cosmos

  • *****
  • Posts: 8,172
  • Joined: 30 Dec 2006

Posted 20 October 2021 - 10:03 AM

I would take sometime to build the scope and star test it.  Even if it’s not “finished” you’ll learn some stuff about star testing while comparing with your ronchi to the star test. 

Always good to have two "points of view".


  • Knight Sky likes this

#58 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 20 October 2021 - 03:46 PM

Seems like the last session has not gone my way. I had two sessions with ToT W (3/4D) suggested by dogbiscuit and one session of smoothening short Z (1/4D). Mirror now has TDE, I am not too sure about the figure and the surface seems worse than before. Seems like I was not able to control ToT W properly and may have overshot multiple times and hence the TDE.

 

Here's a 65 lpi outside Ronchi.

 

IMG 20211021 014917

Edited by Knight Sky, 20 October 2021 - 04:12 PM.


#59 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 20 October 2021 - 03:55 PM

With Ronchi matching, figure appears not to be too bad. To my naked eyes it seems however that the bottom part of the edges seem to have got tiny bit more correction than the top part. Could just be an optical illusion though.



#60 Steve Dodds

Steve Dodds

    Apollo

  • *****
  • Vendors
  • Posts: 1,148
  • Joined: 02 Apr 2007
  • Loc: Utah

Posted 20 October 2021 - 04:06 PM

That turned edge is nothing to worry about, a few passes with a beveling stone and it will be gone.


  • Knight Sky likes this

#61 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 26,042
  • Joined: 22 Feb 2010

Posted 20 October 2021 - 05:05 PM

 

To my naked eyes it seems however that the bottom part of the edges seem to have got tiny bit more correction than the top part.

probably a tillt error in the grating to the mirror surface

 

can't use ronhi to read tde... use a ke to look at the edge, bright ring around the disc

 

 

observing planets with an uncoated primary always gives pleasing results.


Edited by Pinbout, 20 October 2021 - 05:07 PM.

  • Knight Sky likes this

#62 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 20 October 2021 - 07:38 PM

I don't understand why it is so much trouble to take a Foucault image.


  • Knight Sky likes this

#63 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 20 October 2021 - 08:31 PM

Seems like the last session has not gone my way. I had two sessions with ToT W (3/4D) suggested by dogbiscuit and one session of smoothening short Z (1/4D). Mirror now has TDE, I am not too sure about the figure and the surface seems worse than before. Seems like I was not able to control ToT W properly and may have overshot multiple times and hence the TDE.

 

Here's a 65 lpi outside Ronchi.

 

When you say "two sessions with ToT W" do you mean you did 1 trip around the mirror and tested? Then another session of the same and tested. Then the smoothing stroke? Was there roughness seen by testing  after the TOT W's that caused you to do the short 1/4 Z?

Was that Z MOT or TOT.  I'm not familiar with that stroke.  I guess it is stroking a Z left to right, or else it would be an N.  If so how is that different in effect from an N stroke.  Please describe that stroke.  

 

The 65 lpi Ronchi looks good, would like to see what it looks like with the 133 lpi grating. The 65 lpi grating is not as sensitive as the 133 lpi.  65 is ok but it is difficult to compare smoothness of the curve to the previous Ronchi with the 133 lpi.  I think you have improved the curve but I can't be sure it is not only that the latest test was not as sensitive.

 

Foucault will show the edge better than Ronchi.


  • Pinbout and Knight Sky like this

#64 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 21 October 2021 - 12:48 AM

probably a tillt error in the grating to the mirror surface

 

can't use ronhi to read tde... use a ke to look at the edge, bright ring around the disc

 

 

observing planets with an uncoated primary always gives pleasing results.

I hope that issue is with the tilt and not on the mirror. I will check it again with 133 lpi ronchi screen. waytogo.gif

 

I don't understand why it is so much trouble to take a Foucault image.

I did take a few Foucault and those were a little disappointing frown.gif, hence I didn't post them. Attaching one now, at about 75% zone nulled. My "improved" tester is having some trouble it seems, I am not getting the similar quality Foucault images that I was getting with the old one.

 

When you say "two sessions with ToT W" do you mean you did 1 trip around the mirror and tested? Then another session of the same and tested. Then the smoothing stroke? Was there roughness seen by testing  after the TOT W's that caused you to do the short 1/4 Z?

Was that Z MOT or TOT.  I'm not familiar with that stroke.  I guess it is stroking a Z left to right, or else it would be an N.  If so how is that different in effect from an N stroke.  Please describe that stroke.  

 

The 65 lpi Ronchi looks good, would like to see what it looks like with the 133 lpi grating. The 65 lpi grating is not as sensitive as the 133 lpi.  65 is ok but it is difficult to compare smoothness of the curve to the previous Ronchi with the 133 lpi.  I think you have improved the curve but I can't be sure it is not only that the latest test was not as sensitive.

 

Foucault will show the edge better than Ronchi.

I did one session with ToT W, tested and felt that a little more correction was needed. I then did another round of the same W stroke and 3/4D overhang 1/4D stroke width N (MoT), one round around the barrel. I called it Z for zigzag, but you are right that it effectively looks like slanted connected Ns on the lap from left to right. Thought behind using this stroke was that it should not only smoothen the figure but also help in getting rid of the TDE caused in the first session of W. @Pinbout mentioned this stroke in some thread on CN. It is quite possible that I may have used it incorrectly or at an inopportune time.

 

Curve has definitely improved as much as I could see on Foucault with the naked eyes. Just that I think that ~.9-1. zone is still lacking. I will post 133 lpi as soon as I can.

 

I think TDE is least of my worries, smooth curve and surface roughness is something that's bothering me. Since I have not bevelled the mirror at all, I won't mind 1-2 mm bevelling to get rid of TDE. I have not been able to pinpoint what might be causing these sleeks that's causing a rough surface. I am beginning to think that the lap is getting corrupted with micro dust particles when I rechannel it, since I keep the room windows open for fumes (using soldering iron technique) to not flow at me, during that time there's a mild breeze that flows across the room which may be carrying these micro dust particles.

 

Foucault at ~.75 zone nulled

IMG 20211021 105827

Edited by Knight Sky, 21 October 2021 - 01:26 AM.


#65 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 21 October 2021 - 02:27 AM

133 lpi inside and outside RoC
 
IMG 20211021 125216
 
IMG 20211021 125115


#66 davidc135

davidc135

    Surveyor 1

  • -----
  • Posts: 1,875
  • Joined: 28 May 2014
  • Loc: Wales, UK

Posted 21 October 2021 - 02:27 AM

Still under-corrected by 1/4 wave it looks to me. (referring to post 58)  David


Edited by davidc135, 21 October 2021 - 02:39 AM.

  • Knight Sky likes this

#67 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 21 October 2021 - 07:27 AM

I don't think it's 1/4 wave yet, but then I'm not all that good judging Ronchi.
I would have already been taking zone readings, but that seems to have gone out of favor in recent  years.

 

Concerning the tde.
It is not necessarily the stroke that turns the edge. It can be things like the cerium oxide and water quantity and mixture, stroke speed, could be the lap being freshly trimmed.
I see you posted a Foucault after a previous lap trimming that also had some turn down.
The edge you have now is not very bad and within the bounds of fluctuations that can be seen with variations of those things mentioned. As you continue adding correction, the edge can come back as good as it was, a favorable fluctuation.


  • Knight Sky likes this

#68 davidc135

davidc135

    Surveyor 1

  • -----
  • Posts: 1,875
  • Joined: 28 May 2014
  • Loc: Wales, UK

Posted 21 October 2021 - 07:56 AM

Yes, take zonal readings- even if it is only three. Centre, .71r and edge zone.  David


  • Knight Sky likes this

#69 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 21 October 2021 - 08:24 AM

I don't think it's 1/4 wave yet, but then I'm not all that good judging Ronchi.
I would have already been taking zone readings, but that seems to have gone out of favor in recent  years.

 

Concerning the tde.
It is not necessarily the stroke that turns the edge. It can be things like the cerium oxide and water quantity and mixture, stroke speed, could be the lap being freshly trimmed.
I see you posted a Foucault after a previous lap trimming that also had some turn down.
The edge you have now is not very bad and within the bounds of fluctuations that can be seen with variations of those things mentioned. As you continue adding correction, the edge can come back as good as it was, a favorable fluctuation.

I will take the Foucault readings once I get the Couder mask made. I have been lazy about it and also indecisive as to which one to use - Everest pinstick or Couder, but I realize for my level of "expertise" Couder would be a safer choice.

 

Yeah, I think I didn't go in confidently and smoothly with that stroke, and mostly likely screwed up with swiftly switching over between Ws. I am the one to be blamed for all that happened on the mirror (and perhaps a few other potential causes you mentioned). I am confident that next time I use that stroke I'll be much better at it. waytogo.gif

 

Do you think applying a center deepening W (MoT) makes sense to bring in more correction or wait until the Foucault readings and take it from there?

 

Btw., I did have a new Ronchi screen printed locally. It is not a high quality print but hopefully it wont have as many scratches as the current one.



#70 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 21 October 2021 - 08:27 AM

Yes, take zonal readings- even if it is only three. Centre, .71r and edge zone.  David

Sure David, I'll try my best to take the readings. Just that with that kind of a surface I am a little less confident of being able to judge properly. Nonetheless, something is better than nothing. waytogo.gif


Edited by Knight Sky, 21 October 2021 - 08:28 AM.


#71 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 21 October 2021 - 11:00 AM

I will take the Foucault readings once I get the Couder mask made. I have been lazy about it and also indecisive as to which one to use - Everest pinstick or Couder, but I realize for my level of "expertise" Couder would be a safer choice.

 

Yeah, I think I didn't go in confidently and smoothly with that stroke, and mostly likely screwed up with swiftly switching over between Ws. I am the one to be blamed for all that happened on the mirror (and perhaps a few other potential causes you mentioned). I am confident that next time I use that stroke I'll be much better at it. waytogo.gif

 

Do you think applying a center deepening W (MoT) makes sense to bring in more correction or wait until the Foucault readings and take it from there?

 

Btw., I did have a new Ronchi screen printed locally. It is not a high quality print but hopefully it wont have as many scratches as the current one.

You are doing very well.  Only minor troubles. 

 

Measure before deciding what to do next.

 

I didn't intend for the TOT W to be the only stroke to add correction.  It is more effective adding correction from about 65% or 70% zone outward.

Along with that the typical MOT parabolizing stroke is is more effective adding correction in the central zones.

The length and width of the Ws can be altered to have more effect closer to edge or center.

 

Using the 3 classical MOT parabolizing strokes as shown in Texereau's book, the normal parabolizing stroke, the center correcting, and outer zone correction can get  you to a good parabola.  You have to test to know which one is appropriate and how far you have to go to the parabola. 


  • Knight Sky likes this

#72 davidc135

davidc135

    Surveyor 1

  • -----
  • Posts: 1,875
  • Joined: 28 May 2014
  • Loc: Wales, UK

Posted 21 October 2021 - 11:46 AM

I think my notion of 1/4 wave was an underestimate- the Ronchi showing 50% correction looked pretty close.

 

Because the Foucault image shows a slight upturn just inside the edge you'll be able to get a good k.e measurement on the outer zone which, as you say, needs most work. Even without a mask or pin. That, with the centre measurement, will give the overall pv in two ticks.

 

David


  • Knight Sky likes this

#73 Knight Sky

Knight Sky

    Vostok 1

  • -----
  • topic starter
  • Posts: 104
  • Joined: 05 Jun 2021
  • Loc: India

Posted 22 October 2021 - 05:53 AM

Here's a first set of data taken with 5 zones and 3% central area excluded. I would double check zone radius numbers and notify in case of any anomaly from my side. But looking at these results it's amply clear that I should proceed ahead with adding more correction in the center. One thing I am in serious doubt is my ability to measure RoC precisely, I may still be off by about 1/2 an inch or so. I do not have tools to measure it very precisely. Is there a simple technique that I am missing to measure it? With the tape measure there is bound to be some slack which means my measurements are always going to be on the + side of the actual RoC.

 

IMG 20211022 155413
 
IMG 20211022 155442

  • davidc135 likes this

#74 ccaissie

ccaissie

    Vanguard

  • *****
  • Posts: 2,428
  • Joined: 13 Sep 2010
  • Loc: Whitefield, Maine

Posted 22 October 2021 - 06:13 AM

Stiffening the tape measure would eliminate the sag...tape a stick or old fishin' pole to it.  Whatever you got.

 

You could enter a few different radii and see how much the analysis changes....At this stage, your mirror is undercorrected whether you are +/- a bit on your radius.  As you say, true ROC will probably be less than your sagging tape says, so if you enter that longer measured radius number in XP, a set of "perfect" zonal measurements will give you an undercorrected mirror.

 

Think more, polish less. Refine your measurements and skill over time.  Repeat until you are sure you are converging on accuracy.  Can you see the correspondence of this analysis with the Ronchi test?  How about loading it in the scope for a star test?

 

Edit:  Oh and remember to test the mirror over different diameters and average them if it's a bit different.  If VERY different then you need to think about astigmatism creeping in.  That's where the star test "shines".


Edited by ccaissie, 22 October 2021 - 06:30 AM.

  • Knight Sky likes this

#75 dogbiscuit

dogbiscuit

    Surveyor 1

  • *****
  • Posts: 1,744
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 22 October 2021 - 10:15 AM

Don't add correction to the center now.

I'll explain in my next post.

 


  • Knight Sky likes this


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





Also tagged with one or more of these keywords: Mirror Making, ATM, Beginner



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics