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 a 6-in f/8

mirror making
  • Please log in to reply
149 replies to this topic

#1 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 15 August 2020 - 09:09 AM

I'm tackling my first mirror to figure. It is a 6-inch f/8 that I started at the Delmarva workshop years ago. Due to the long drive home I had to leave the workshop after polishing out but before finishing the figure. I've now fabricated a tester (moving light source), collected most of the stuff I need, and made my first pitch lap (Cold pressed it with nylon window screening.) In the picture it doesn't look ideal (it didn't press into contact all the way to the edge). I'm sure you'll let me know what you think. I've also got a set of Ronchi images I'll link here.

 

My inexperienced impression is that I've got a TDE and hole in the center. What to do next is of course the main question. Bear in mind that I don't know the figuring strokes by name yet. I do have a number of published resources here if referring to figures in them is helpful (Brown, Texerau 2nd ed, Berry 1985, Kriege and Berry, Howard).

 

Thank you - I've already learned a lot reading through threads here.

 

Ronchi images 14 Aug 2020

 

Lap 14 Aug 2020


  • LarsMalmgren, davidc135, chantepierre and 1 other like this

#2 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 15 August 2020 - 11:52 AM

You need to open up those channels

 

but looks like a really good sphere, just smooth it out with narrow figure 8’s 

 

then do narrow w’a to add some correction - you don’t need much at f8.



#3 dogbiscuit

dogbiscuit

    Apollo

  • *****
  • Posts: 1,238
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 15 August 2020 - 12:09 PM

It's a pretty good sphere with slight zones. Looks like there is only slight tde.  Edge condition will show best with Foucault.  Shouldn't take much to smooth it out and make the edge good.

 

Press to get full contact all the way to the laps edge.

Probably best to press TOT.

 

If the lap is thin and you are worried about squeezing pitch out at the edge as you press put a turn or two of  non-stretch plastic packaging tape around the edge of the lap extending up the edge of the lap to about .5 or 1 mm from the top surface of the lap. This will hold the pitch in the lap while it flows into full contact all the way to the edge.

 

If polishing is complete and you are now figuring, particularly since you are already at a pretty good sphere, don't work unless the lap is right. 

 

Proper channels and pressed well are very important.  Channels don't need to be too deep an wide.  About twice as wide as what you have and about the same depth as width, and same depth and width over the entire lap.



#4 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 15 August 2020 - 12:30 PM

That's very helpful, thanks both of you. I attempted to make the lap pretty thick - I checked and it is 3/8 inch all the way around. So I'll work on good contact over all of it and open up the channels.

 

You need to open up those channels

 

but looks like a really good sphere, just smooth it out with narrow figure 8’s 

 

then do narrow w’a to add some correction - you don’t need much at f8.

Is this TOT, MOT, or alternating TOT and MOT?



#5 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 15 August 2020 - 01:59 PM

MoT from now on



#6 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 18 August 2020 - 11:20 AM

I pressed the lap a number of times until it had full contact. I cut the channels wider (and will continue to keep them open). I patched back in the little chips that jumped out, as described in earlier threads. Did narrow figure-8 (1/3) stroke eight times around the table, and made the photos linked below. One is with the knife edge (entering from right) near RoC, which may help show the figure (I'm still learning to use the tester). Close to a sphere perhaps, but you can see the zone and where I offset the Ronchi grating, the TDE. Not sure if I'm progressing yet...I can keep doing more like this. Advice on how to proceed is welcome.

 

Ronchi and Foucault images 18 Aug 2020

 

Lap before latest figuring session



#7 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 18 August 2020 - 02:10 PM

i cut my laps straight down, straight no chamfer

 

you need more than 8x's around

 

https://www.youtube....t7salR23dE&t=2s

 

1st stroke figure 8 could be a little wider...

 

I like drawing with sharpies on the back of the mirror...

 

you could draw a 30% circle on the back of yours and do not cross that line when doing your w's

left to right, front to back don't cross the line.


Edited by Pinbout, 18 August 2020 - 02:11 PM.


#8 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 18 August 2020 - 06:13 PM

The edge of the lap is vertical; it is hard to see that in the photo because I didn't repress the window screen pattern into the pitch this last time. It does go all the way to the edge.

 

The video is great - a big help. I see what the strokes are, and that mirror sounds like it had the same shape as mine. wink.png



#9 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 18 August 2020 - 06:31 PM

There’s also a rhythm that goes along with the stroke that helps keep things smooth

 

a flowing motion, didn’t get it till I saw Dick Parker do some strokes at delmarva it


Edited by Pinbout, 18 August 2020 - 06:33 PM.


#10 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 19 August 2020 - 05:55 PM

After the smoothing and correcting strokes, it looks to my untutored eye like the figure is definitely different, and the zones have shifted. Perhaps now overcorrected? And the TDE remains. Here are Ronchi images (some sharpie marks showing through).

 

Ronchi images 19 August 2020

 

Curious which of these issues you would address next. Thanks for having a look.


Edited by kcoles, 19 August 2020 - 05:56 PM.


#11 dogbiscuit

dogbiscuit

    Apollo

  • *****
  • Posts: 1,238
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 19 August 2020 - 06:54 PM

A real Foucault knife edge image is much more meaningful than the at or near ROC Ronchi image.

"Real Foucault knife edge image" means not using a line of the Ronchi as a knife.  It's not the same.

 

What's the lap look like now and what sort of pitch are you using?

What is the workroom temperature and are you pressing at that temperature?



#12 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 19 August 2020 - 07:24 PM

After the smoothing and correcting strokes, it looks to my untutored eye like the figure is definitely different, and the zones have shifted. Perhaps now overcorrected? And the TDE remains. Here are Ronchi images (some sharpie marks showing through).

 

Ronchi images 19 August 2020

 

Curious which of these issues you would address next. Thanks for having a look.

how many lpi is your grating...



#13 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 19 August 2020 - 07:34 PM

if it is 133lpi

 

after you smooth out your lines... you'll be over... like a 0.1 over

 

if you posted a bigger pic of the 2 lines outside RoC, make the lines correctly vertical, i hate turning my head looking at those lines lol.gif

 

i'll show you how and where to smooth the flow.

 

be considerate and post the pics like

 

https://www.youtube....JWsqTOnF5o&t=3s

 

 

 

6inf8 10th over.JPG

 

 

right on the money

6inf8.JPG


Edited by Pinbout, 19 August 2020 - 07:34 PM.

  • Matthew Paul and dogbiscuit like this

#14 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 20 August 2020 - 03:01 PM

I attempted to take Foucault photos when I did the Ronchi and they just didn't come out. I'm going to move the entire testing setup to another location I can darken better. While I'm at it I will take apart the tester and get the grating in there straight.

 

My apologies for using links rather than images. I run a discussion board (not as huge as CN) for an interest unrelated to astronomy, and there we are fine with either images or links to same. Knowing your preference I won't post until I can do the images as well - my first couple of attempts to insert them failed. Thank you for the video link.

 

My appreciation for your patience with a newbie figuring out how things work. I'll be back with the rest of the details when I have the tester and photos working.



#15 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 22 August 2020 - 07:21 PM

My current question is, How do you clean a mirror for Foucault testing? I spent four hours today with the testing setup now in my basement (constant 72 degrees, no currents, windows blocked). I learned a lot. Ronchi is fine and now vertical. I can see the Foucault figures much better, especially after it is dark outside. But when I photograph the Foucault with my Canon 20D on a tripod it shows much besides the figure. I went through several cycles of cleaning, as I can see smudges, and when I clean those with isopropyl, there is dust or lint. A rinse with distilled water gets that off, but takes a long time to dry and then the photo shows...another smudge (I really am keeping my hands off there) but no water spots. Actually it looked like the same smudge as before. And so on. Perhaps the CeOx needs some extra cleaning to get it off of the glass? I rinse it in running water but don't rub the glass.

 

This mirror is long ago polished out - the folks at Delmarva certified that, and I subsequently worked half a day on the figure there. It has been carefully stored since then and I'm careful with it when I have it out. I'm tempted to think that Foucault is very unforgiving, even of dust particles. Is that the case? The photos others post on here are immaculate, so my technique is lacking something. I want to fix it before inflicting any more photos or figuring questions on all of you. Thanks.



#16 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 22 August 2020 - 09:25 PM

Majestic uses cloth diapers- just saying.

 

show to dirty photos, now we want to see how disgusting they are



#17 chantepierre

chantepierre

    Vostok 1

  • -----
  • Posts: 171
  • Joined: 27 Mar 2020
  • Loc: France

Posted 23 August 2020 - 01:39 AM

To get a clean mirror for my tests, I had a little bucket with clean water just next to me, to drop the mirror (gently) inside the moment it’s separated from the pitch lap. It was the end of spring and things dried really fast.
Unless I did that, I was unable to really clean cerox under running water.

#18 ccaissie

ccaissie

    Surveyor 1

  • *****
  • Posts: 1,964
  • Joined: 13 Sep 2010
  • Loc: Whitefield, Maine

Posted 23 August 2020 - 06:17 AM

I usually have TOT with bigger disks, so for cleanup I have a couple of paper towels, one handles most of the slurry (cerox/barnesite).  As the remaining film dries, I hit it with a nearly dry/clean paper towel and it cleans up.  I see the reflection of the overhead lights, and that shows the degree of clean.  Pyrex ( I assume) is pretty hard stuff so clean paper towel will not wipe any damage into the surface. I believe.  

 

Maybe for photo purposes it has to be super clean but for testing I find a streak here and there to be unimportant. Lockwood's photos of tests have a few stray particles and such...no shame there.   http://www.loptics.c...hop/shop24.html

 

Foucault is my way for classical bench testing.  I do DPAC Ronchi, but find that the knife edge, in my hands, duplicates the star test results more closely.  Ronchi does tell the truth about the edge and general smoothness tho'.

 

A good 6f8 is a superb scope.



#19 ed_turco

ed_turco

    Mercury-Atlas

  • *****
  • Posts: 2,748
  • Joined: 29 Aug 2009
  • Loc: Lincoln, RI

Posted 23 August 2020 - 08:02 AM

I'm tackling my first mirror to figure. It is a 6-inch f/8 that I started at the Delmarva workshop years ago. Due to the long drive home I had to leave the workshop after polishing out but before finishing the figure. I've now fabricated a tester (moving light source), collected most of the stuff I need, and made my first pitch lap (Cold pressed it with nylon window screening.) In the picture it doesn't look ideal (it didn't press into contact all the way to the edge). I'm sure you'll let me know what you think. I've also got a set of Ronchi images I'll link here.

 

My inexperienced impression is that I've got a TDE and hole in the center. What to do next is of course the main question. Bear in mind that I don't know the figuring strokes by name yet. I do have a number of published resources here if referring to figures in them is helpful (Brown, Texerau 2nd ed, Berry 1985, Kriege and Berry, Howard).

 

Thank you - I've already learned a lot reading through threads here.

 

Ronchi images 14 Aug 2020

 

Lap 14 Aug 2020

I wish that MY first mirror had your spherical figure and that tiny TDE!  You are way ahead on figuring.
 



#20 dogbiscuit

dogbiscuit

    Apollo

  • *****
  • Posts: 1,238
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 23 August 2020 - 10:41 AM

Edit to add <Don't let the slurry dry on the mirror.  It's hard to get off if it dries.  When I finish a session and separate the mirror from the lap I immediately spray water on the mirror.>

 

I wash my hands with hand soap (Ivory), and then the mirror with the same soap wiping gently with the palm of my hand.  All traces of soap are washed away with flowing water, mirror is turned on edge to let most water run off, blot the remaining water off with a paper towel, maybe just a very little bit of wiping to prevent any embossed pattern on the towel from imprinting on the mirror surface. The result is a very clean smudge free mirror.  I pretty much do it this way all the time, because it's not hard to do and paper towels are cheap.

 

There is some correction in post 10 images but not very smooth.  Might be near or over 100% correction. Some irregular stuff showing in near COC Ronchi and maybe also the 2 band outside ROC Ronchi, might be air currents but likely is on the mirror surface.

 

Your images in post 1 showed a smoother figure than your images in posts 6 and 10.

Can't tell for sure but I think the post 1 edge was better too.

Analysis of Foucault knife readings might show low wavefront error and good strehl, but I would not settle for a figure that rough.  I would rather have a smooth sphere than a parabola that rough.

What sort of strokes were you using that resulting in the post 1 condition?

 

There are ways to smooth the existing correction but I think it would be easier and quicker to go back to a sphere, get it smoothed out and try to progress smoothly to a parabola.


Edited by dogbiscuit, 23 August 2020 - 11:09 AM.


#21 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 23 August 2020 - 01:06 PM

You are reading things correctly - the mirror is more rough and less smooth than before. I can see it in the tester. I would not settle for it either. And it is overcorrected, the screen is 100 lpi, not 133 (I had that info on the image but will put it in the text of the post from now on). After three sessions in which I saw virtually no change (last of these was images in post 6), I tried the figure 8 stroke and some zigzag (which I see increases correction) and clearly overdid it (post 10). In fact, in spite of the nice remarks about post 1, that was the figure after the 2006 Delmarva workshop. Everything I have done since resurrecting this project has made it worse, without exception. I'm not giving up yet, but I am ready to do something correctly. I don't think I need to share additional pictures, you've sussed it out. I will show the (full-size) lap - I've added to my previous patches of the chips that jumped out, pressed, trimmed the channels and made the edge flush vertical. (Again, the screen pattern predates a lot of pressing and gives a false impression that contact doesn't go to the edge; in fact it does, as I'm watching it closely.) This is Gugolz 64. I work in my basement, a constant 70-72 degrees, and that is where I press, figure, and test (everything except steps where a sink is needed).

 

Lap_23Aug2020.JPG

Lap_23Aug2020_top.JPG

 

It makes sense to me to go back to spherical, and I need some character building anyway. Believe me, I will go in baby steps from now on. Don't assume I know anything (what stroke, how big, mirror or tool on top), so details on how to go back to spherical would be very helpful. I'm all ears (or eyes). Thank you!



#22 dogbiscuit

dogbiscuit

    Apollo

  • *****
  • Posts: 1,238
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 23 August 2020 - 04:22 PM

Spherical is not far away and you need some work to smooth it out anyway so it is not really any extra work to get to a sphere.  When surface is smooth and edge good go slowly toward the parabola.  The parabola is very close to the sphere and a parabola can be reached by with very little work. 

 

Zones are sometimes caused by poorly pressed lap or closing channels.  If you have those things right the next major cause for zones is using the same stroke pattern and size for too long a time.  Too long time is relative.  Now that polish is complete, after one or two turns of the mirror when on top or around the mirror when TOT is usually about time to change the stroke in some manner.  Frequent change is good so I usually choose to change after each full turn of or around the mirror. For figuring I use almost exclusively W strokes of various stroke lengths and widths of the W.  The Ws usually will be either 5 or 7 back and forth strokes across the W.  Whatever size W  and number of strokes I've chosen I do it the same at about 12 (between 11 and 13) increments to a full rotation of the mirror or trip around the mirror.  Sometimes (but rarely) I will use a center over center stroke, usually either 5 or 7 strokes at about 12 positions, but sometimes for any straight back and forth stroke I will do a single back and forth stroke at each of about 60 increments of mirror rotation MOT or positions around the mirror TOT.  

 

For a 6" mirror typical stroke rate will be about 1 forward and back stroke in 1.5 seconds, so about 40 strokes (forward and back) per minute.  Shorter strokes like 1/3D might have a higher stroke rate, say around 1 per second or 60 per minute.  A good rate depends on some variables and lower or higher rates than what I suggest can work ok, but you won't go wrong at the suggested rate.

 

For figuring steady pressure can be applied by hands on the back of the mirror as you stroke.  Good pressure would be something between 3.5 and 7 pounds as you stroke. A kitchen scale or postal scale can be used to see what that pressure feels like.  I put my hands palms and fingers down flat on the back of the mirror with thumbs touching at mirror center. Thumbs are a good reference for path of the mirror's center as you stroke the desired pattern.  For a 6" mirror the little fingers will have to float in space because there is not room for them on the back of the mirror.  Very slightly damp palms will grip the back of the mirror well enough for stroking if pressure is applied evenly.  Too much moisture, hands slip. I rarely ever hook my fingers at the edge of the mirror.  The same pressure is maintained during the entire stroke pattern and at each increment of rotation, unless I have some special reason to alter the pressure.  Not likely to be much need for modulating pressure for a 6" f/8.

 

All that is the basics of what works for me.

 

Then there's choosing specific stroke length and width of the Ws

I'll say more about that later.

 

It should not be difficult to finish the mirror working only MOT.  I recommend working only MOT until something comes up that is not working right with MOT.

 

For now if you work another session I recommend doing  a few different size strokes,  mostly small to work toward a sphere. The mild zones shown in post 1 images are a typical result (for me) when doing only COC.

So add some side motion to the COC to make them narrow Ws, width of the W 1/4 or 1/3D.  Stroke length about 1/3 D and sometimes 1/2D long.  But remember do the same length and width for full turns of or around the mirror.  Change width and/or stroke length after each full turn.

 

It will take a while to remove the central depression in the center of the mirror, so I would say do the above for about 45 minutes and see how it looks after that.


  • Pinbout likes this

#23 dogbiscuit

dogbiscuit

    Apollo

  • *****
  • Posts: 1,238
  • Joined: 18 Oct 2018
  • Loc: Barksdale

Posted 23 August 2020 - 04:49 PM

Uniform channel depth and width, good pressing is important for getting a smooth surface.

 

The 20D should be able to get good Foucault images.  I've used one of those.

Print out something like an Eye chart on some card stock and place it beside the mirror surface and focus the camera on the text.  Flip the camera's mirror up before the exposure to avoid the vibration distorting the image.

 

Foucault will show the edge condition better than Ronchi, and better than using one edge of a grating line for a knife.


  • Pinbout likes this

#24 kcoles

kcoles

    Explorer 1

  • -----
  • topic starter
  • Posts: 65
  • Joined: 13 Jul 2020
  • Loc: western Pennsylvania, US

Posted 25 August 2020 - 10:47 AM

That guidance makes sense and is a big help. The way you describe placing hands on the mirror is about what I have been doing; it makes the most sense.

 

The 20D has also been great on a telescope over the years, where I learned to do mirror lockup and check (and do) the focus manually.



#25 Pinbout

Pinbout

    ISS

  • *****
  • Posts: 24,885
  • Joined: 22 Feb 2010

Posted 25 August 2020 - 03:49 PM

You are reading things correctly - the mirror is more rough and less smooth than before. I can see it in the tester. I would not settle for it either. And it is overcorrected, the screen is 100 lpi, not 133 (I had that info on the image but will put it in the text of the post from now on). After three sessions in which I saw virtually no change (last of these was images in post 6), I tried the figure 8 stroke and some zigzag (which I see increases correction) and clearly overdid it (post 10). In fact, in spite of the nice remarks about post 1, that was the figure after the 2006 Delmarva workshop. Everything I have done since resurrecting this project has made it worse, without exception. I'm not giving up yet, but I am ready to do something correctly. I don't think I need to share additional pictures, you've sussed it out. I will show the (full-size) lap - I've added to my previous patches of the chips that jumped out, pressed, trimmed the channels and made the edge flush vertical. (Again, the screen pattern predates a lot of pressing and gives a false impression that contact doesn't go to the edge; in fact it does, as I'm watching it closely.) This is Gugolz 64. I work in my basement, a constant 70-72 degrees, and that is where I press, figure, and test (everything except steps where a sink is needed).

 

attachicon.gifLap_23Aug2020.JPG

attachicon.gifLap_23Aug2020_top.JPG

 

It makes sense to me to go back to spherical, and I need some character building anyway. Believe me, I will go in baby steps from now on. Don't assume I know anything (what stroke, how big, mirror or tool on top), so details on how to go back to spherical would be very helpful. I'm all ears (or eyes). Thank you!

You need to keep the channels open and squares even. Or you will get those shallow zones you see when using the ronchi as a ke.

 

You need to de correct but going back to spherical, you don’t need to. 




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



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics