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

Polishing problems

  • Please log in to reply
11 replies to this topic

#1 aalbay1

aalbay1

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 14 Jul 2020

Posted 16 September 2020 - 04:41 PM

Hi, I have some questions about polishing. This is my first telescope so I'm a novice at this stuff. I'm also posting in the forums for the first time so go easy on me :) My scope is 20cm f5.5. I was trying to make the f ratio larger but couldn't seem to make it work so I stayed at this value. Anyway, my current problem is that I've been polishing for about 4 hours and doing ronchi tests at regular intervals but I don't see any improvement. I'm following the instructions on stellafane and according to my ronchi tests, the curve is "overcorrected" meaning I'm seeing curves away from center inside ROC and into center outside ROC. I tried taking a picture but my phone's camera is pretty. I read that this means the mirror is deeper than a sphere so I thought I should raise the curve. To do that, I've been polishing with mirror on top for the last hour and a half.

Another problem I've had is that the outer 4-5 mm radius is not polishing. To make sure I didn't miss anything I went all the way to 1200 grit before starting polishing but that outer rim just doesn't seem to polish out.

I would really appreciate some help on both of these problems because I'm stumped.

Thank you and clear skies,
Alp

Ps. I couldn't paste the pictures but my ronchigrams look alomst exactly like the two under the title "Surface Roughness" in https://stellafane.o...onchigrams.html except the surface roughness part. (I haven't parabolized my mirror and I wasn't planning on it.)

#2 Al Paslow

Al Paslow

    Mariner 2

  • *****
  • Posts: 217
  • Joined: 07 Mar 2005
  • Loc: SW Pennsylvania

Posted 16 September 2020 - 06:32 PM

Off the top of my head the first thing to determine is where you are in terms of from Radius of Curvature. The Stellafane page makes good references here. If you are inside or outside will make a difference of how to interpret the bands.

 

Spent the time to be accurate here.

 

You make no mention of what polishing agent you are using nor the substrate the mirror is made of. Obviously quartz or zerodur is hard and takes forever while plate glass polishes much quicker. Since glass polishes from the very center outward there is always very much concern about being fully polished out. There are lots of things to think about here including hot pressing the lap, cold pressing etc to get the best contact.

 

Ok, so the contact between the mirror and tool could be checked before moving on to polishing. I'm old school and use the pencil test. If the lines don't rub away from both the mirror and tool after maybe a minuet or two I don't move on to making the lap and polishing. Lot's of people think grinding is just a simple procedure but there is a science about doing it right.

 

When using 1200 grit, did you look look for uniformity from center of mirror to the very edge? Did you do the same with the tool? Does the mirror and tool look fully ground out to the edge. How did you test to determine it is good to move on to polishing? It's possible your edge is not fully ground out.

 

Polishing out a mirror can take hours more time then you might want it to. I would keep working the lap if I thought I had good enough contact during fine grinding. But again in polishing the game changes, things need to be right to enter that playing field.

 

Just a few ideas.


  • aalbay1 likes this

#3 Garyth64

Garyth64

    Fly Me to the Moon

  • *****
  • Posts: 5,201
  • Joined: 07 May 2015
  • Loc: SE Michigan

Posted 16 September 2020 - 07:50 PM

Don't really worry about doing any ronchi tests until the mirror is totally polished out. 



#4 Gordon Waite

Gordon Waite

    Vendor-Waite Research

  • *****
  • Vendors
  • Posts: 413
  • Joined: 11 Jun 2007
  • Loc: New Jersey

Posted 16 September 2020 - 07:53 PM

Hi, I have some questions about polishing. This is my first telescope so I'm a novice at this stuff. I'm also posting in the forums for the first time so go easy on me smile.gif My scope is 20cm f5.5. I was trying to make the f ratio larger but couldn't seem to make it work so I stayed at this value. Anyway, my current problem is that I've been polishing for about 4 hours and doing ronchi tests at regular intervals but I don't see any improvement. I'm following the instructions on stellafane and according to my ronchi tests, the curve is "overcorrected" meaning I'm seeing curves away from center inside ROC and into center outside ROC. I tried taking a picture but my phone's camera is pretty. I read that this means the mirror is deeper than a sphere so I thought I should raise the curve. To do that, I've been polishing with mirror on top for the last hour and a half.

Another problem I've had is that the outer 4-5 mm radius is not polishing. To make sure I didn't miss anything I went all the way to 1200 grit before starting polishing but that outer rim just doesn't seem to polish out.

Ps. I couldn't paste the pictures but my ronchigrams look alomst exactly like the two under the title "Surface Roughness" in https://stellafane.o...onchigrams.html except the surface roughness part. (I haven't parabolized my mirror and I wasn't planning on it.)

Hi, Alp!  There are a couple of important ideas here.  First, your mirror isn't polished out, so you shouldn't be doing Ronchi tests.  With a half-polished mirror there isn't anything a Ronchi test could tell you that should lead to a change in your polishing procedure.  It will cause you to start doing all kinds of bizarre corrective action when all you really need to do is polish, polish, polish.  Fight the urge to look prematurely!  It is WAY too early to think about over- and under-correction.

 

Second, you didn't give enough info unless we make a bunch of assumptions.  Are you polishing with a full-sized polisher?  Gugolz pitch?  Type of glass?  How much time did you spend at each grit level during grinding?  MOT or TOT?  With an edge that won't touch off, you have to consider that someplace in your grinding regime you didn't spend enough time for the edge to come out right.  That's always a possibility, but we can't tell from your description.

 

How much pressure are you applying during normal polishing?

 

At the end of your post, you said, "I haven't parabolized my mirror and I wasn't planning on it."  So do I understand correctly that you are going to go to all the trouble of gathering materials, grinding through all that grit, make a polisher, polish for hours and learn to work the tester.  But you have no intention to try to parabolize your mirror????  A quick run through FigureXP tells you that if you stop with a sphere, a 20 cm F5.5 mirror would rate about 1.1 WAVES, with a Strehl value of 0.11.  This is not a good mirror.  It's not a poor mirror.  It's a total stinker, as in what is the point in doing all that work and ending up with a shallow shiny ash tray?  What don't I understand here?


  • Garyth64 likes this

#5 Gordon Waite

Gordon Waite

    Vendor-Waite Research

  • *****
  • Vendors
  • Posts: 413
  • Joined: 11 Jun 2007
  • Loc: New Jersey

Posted 16 September 2020 - 08:02 PM

Another problem I've had is that the outer 4-5 mm radius is not polishing. To make sure I didn't miss anything I went all the way to 1200 grit before starting polishing but that outer rim just doesn't seem to polish out.

Forgot to ask you...  On your edge problem, is the edge not polishing at all, or is it polishing slowly?  Does the width of the region get narrower as you work, or stay consistently the same width.

 

If polishing, but slowly, you can just continue polishing.  It wouldn't be unusual for your 20 cm mirror to take 8 to 10 hours to polish out to the edge with a full-sized lap and hand work.  Don't Ronchi until you pass the laser test.

 

But if the edge isn't polishing at all, then you will need to go back and regrind.  During grinding, while it helps to go "all the way to 1200 grit" that doesn't mean the edge is well ground.  If the mirror didn't get enough edge attention at #80 or #120, you won't cure that with 1200.  You need adequate time at all grit levels, and good technique so the edge is getting the attention it deserves.



#6 aalbay1

aalbay1

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 14 Jul 2020

Posted 17 September 2020 - 07:22 AM

You make no mention of what polishing agent you are using nor the substrate the mirror is made of. Obviously quartz or zerodur is hard and takes forever while plate glass polishes much quicker. Since glass polishes from the very center outward there is always very much concern about being fully polished out. There are lots of things to think about here including hot pressing the lap, cold pressing etc to get the best contact.
 
Ok, so the contact between the mirror and tool could be checked before moving on to polishing. I'm old school and use the pencil test. If the lines don't rub away from both the mirror and tool after maybe a minuet or two I don't move on to making the lap and polishing. Lot's of people think grinding is just a simple procedure but there is a science about doing it right.
 
When using 1200 grit, did you look look for uniformity from center of mirror to the very edge? Did you do the same with the tool? Does the mirror and tool look fully ground out to the edge. How did you test to determine it is good to move on to polishing? It's possible your edge is not fully ground out.

Hi Al,
Thank you for your  reply. I'm using Cerium Oxide as my polishing agnet and the glass is basic soda-lime glass (we belive).
 
I use a hot water bath for hot pressing and I open the channels everytime using a 0.5mm diametre metal pipe. 
 
I had two curves on my mirror so I did a bucnh of pencil tests up untill 500 grit to make sure the mirror was uniform, and it was. 

I should also mention I used a sub diametre tool for rough and fine grinding but I compensated for it by adjusting stroke length. Now I'm using a full diametre plaster tool.

I moved onto polishing only after I was certain that the surface had a 1200 grit finish (I used an eyepice to check the surface finish and  eliminated pits untill there were none.) After the 1200 grit both the tool and the mirror was fully ground out to the edge.
 
 10outside-rough.jpg
 What our mirror looks outside of RoC 

10inside-rough.jpg

and this is what our  mirror looks inside the RoC

 

I couldn't take the pictures of our ronchi tests but these are from Stellafane's website

 

 

Thank you again,
Alp



#7 aalbay1

aalbay1

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 14 Jul 2020

Posted 17 September 2020 - 08:24 AM

Hi, Alp!  There are a couple of important ideas here.  First, your mirror isn't polished out, so you shouldn't be doing Ronchi tests.  With a half-polished mirror there isn't anything a Ronchi test could tell you that should lead to a change in your polishing procedure.  It will cause you to start doing all kinds of bizarre corrective action when all you really need to do is polish, polish, polish.  Fight the urge to look prematurely!  It is WAY too early to think about over- and under-correction.

 

Second, you didn't give enough info unless we make a bunch of assumptions.  Are you polishing with a full-sized polisher?  Gugolz pitch?  Type of glass?  How much time did you spend at each grit level during grinding?  MOT or TOT?  With an edge that won't touch off, you have to consider that someplace in your grinding regime you didn't spend enough time for the edge to come out right.  That's always a possibility, but we can't tell from your description.

 

How much pressure are you applying during normal polishing?

 

At the end of your post, you said, "I haven't parabolized my mirror and I wasn't planning on it."  So do I understand correctly that you are going to go to all the trouble of gathering materials, grinding through all that grit, make a polisher, polish for hours and learn to work the tester.  But you have no intention to try to parabolize your mirror????  A quick run through FigureXP tells you that if you stop with a sphere, a 20 cm F5.5 mirror would rate about 1.1 WAVES, with a Strehl value of 0.11.  This is not a good mirror.  It's not a poor mirror.  It's a total stinker, as in what is the point in doing all that work and ending up with a shallow shiny ash tray?  What don't I understand here?

Hi Gordon,

 

Thank you for your advice on how to correclty utilize the  Rochi test

I'm polishing with a full sized plaster, unfortunately since we dont have gugolz  here, I'm using a homemade optical pitch made by a local telescope maker. Its no gugolz but its  better than nothing. I don't now  much about its  composition as the **** won't tell us smile.gif  (its definitely made of pine resin). The glass is basic soda-lime glass (I belive).

We didn't measure our progress by duration, periodically we checked  the surface for pits using an eyepice. We increased grit when no pits were visible (as suggested by Stellafane). We alternated between TOT and MOT. There were a few times were the glass had mutliple curves but I (tried with the best of my abilities) correct the issue. But everytime I made sure that the mirror curvature was correct by using a pencil test. 

As for  the pressure, I would say adequate. I didn't apply knuckle-whitening amount of pressure nor did I go easy on it. 

I didn't know that I had to parabolize the mirror. I didn't know that was a must. Its my first mirror and I'm very new to this. Even if this mirror ends up as C-3PO's ash tray, In the end I just hope I learn enough that it never happens again and my next mirror turns out far better (and hopefully fuctional!)laugh.gif 

As for your follow up post; I belive the edge IS polishing but very slowly. 

Thank you and clear skies,
Alp


  • Gnut likes this

#8 mark cowan

mark cowan

    Vendor (Veritas Optics)

  • *****
  • Vendors
  • Posts: 9,689
  • Joined: 03 Jun 2005
  • Loc: salem, OR

Posted 17 September 2020 - 09:11 PM

I had two curves on my mirror so I did a bucnh of pencil tests up untill 500 grit to make sure the mirror was uniform, and it was.

 

Well no, it wasn't.  Every successive grit is about half as big but all grits are much bigger than the half-micron that light is.  What you end up with early on is partially ground edges (still has substantial pits), or what is usually worse long-ground edges that veer off many waves.  That takes forever to polish out but if you can polish scratches you can snag a nail with out of Pyrex (I have) you'll be fine eventually. Whatever you're doing that's keeping it the same, just lengthen the strokes (more overhang all around) to put more polishing effort there. You can also put more pressure to the outside. It doesn't matter what the figure is if it's not polished, so  pay no attention to what the mirror looks like in a Ronchi test, only the progression of the fully polished surface.  A freshly washed mirror in the sink with water running off it will show no defects with a laser ONLY if it's fully polished out.  

 

Pine resin (please show  a picture of the polisher) and cerium oxide are pretty adequate for plate glass.



#9 ccaissie

ccaissie

    Surveyor 1

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

Posted 18 September 2020 - 11:38 AM

TOT, 1/3 W, plenty of pressure.  Persist.

 

Like it is always advised, testing before polishing 100% is instructive, but generally a waste of time.

 

Parabolizing is well documented and should go well...once you're polished and then corrected to a sphere.


Edited by ccaissie, 18 September 2020 - 11:40 AM.


#10 PalomarJack

PalomarJack

    Lift Off

  • -----
  • Posts: 11
  • Joined: 30 Oct 2012

Posted 18 September 2020 - 08:33 PM

Hi Alp. This is going to be long, but it needs to be. This is no substitute for documentation or personal guidance, I'm just letting you know where you went wrong and what you are in for. It should be able to get you to a perfect sphere. But a proper parabolic figure would require many pages, diagrams and tables of figures.

 

First of all, what is done, is done. "You should haves..." won't mean much here except for others that are considering "pushing their own glass". You can skip it if you want. Here are some thoughts:

 

Start with something smaller, 4 1/4 or 6 inches and make your first scope f/10 to f/11 so you don't have to make it parabolic.

 

Don't worry about Gugolz or any of those other expensive pitches for now unless it comes in the kit. People have been making 10th wave mirrors with cheap "Plane Jane" pitch for over 100 years, if it was good enough for the Hale 200", I'm sure it will work for us. By the way, you can even use clean, thick paper as a lap. Don't do it, save that for experimenting later.

 

For your first mirror, maybe second or third, use red rouge, it cuts slow and makes a great polish. As a result you will be less likely to over-shoot the figure, too. Remember, "It is easier to remove glass that to put it back". No, that is not meant literally, it's more difficult to remove the glass around a "depression" than the glass comprising a "hill". If the kit came with something else, either order rouge or be very careful, the newer faster cutting compounds work, very fast.

 

Cleanliness is paramount, cover all optical and lap surfaces with a damp towel for the night and during breaks, dust is far more abrasive than polishing compound, about like the final grind compound. How many beginners have spent day after day, after day trying to polish out a mirror with a pitch lap contaminated with dust? I can't tell you, but it's a lot. Do not ever touch the lap, you just do not need to, ever. You cannot feel the millionths of an inch accuracy it needs to have, anyway. It will cause glazing. If you have to during repairs, have a paper towel between you and it. Those thick blue paper shop towels are the best. And do not use facial tissues, they have lotion in them these days.

 

Stay away from YouTube, I mean that. There is no substitute for printed material right next to you at the grinding stand and knife-edge tester.

 

Before We Begin

 

Now some thoughts that apply to the present. Let's see if we can salvage what you have.

 

If you don't have one, send for a book about mirror grinding. You will hear about all kinds, but for a first that will serve very well up to 8 inches is Sam Browns All About Telescopes for your first few mirrors. For an example amount of time for a first mirror, many years, nay, decades ago, an old ATMer I met as a teen took a year to make an 8 inch f/6. The view through it, by the way, was exquisite, with over a 20 year old coating! By the way, it took only a few weeks to get to a sphere.

 

Never try to get a figure before a sphere. A shortcut that never works and ends up wasting more time than it saves. Like unraveling the Gordian Knot, "Many men have tried, and many men have failed". And few people have the patience to figure a mirror with a star test, it takes a long, long time. Using a knife edge tester to figure is far faster and vastly more accurate.

 

Getting A Sphere

 

First thing to do while waiting for the book is get this beast a smooth spherical surface, you have everything you need now. Cold press for a night, put an 8" diameter piece of wood under the weight with a Styrofoam disk under that to spread the weight evenly, use about 5 or 6 pounds (2 - 3 K). Then, you will polish 2/3 stroke center over center, no exotic tangent, "W" or other weird strokes to speed up the process. And never put a full size tool on top during polishing. That is only used during grinding. It is what gave you the horrible figure in the ronchigram and is probably delaying the polish around the edge.

 

Now, cold press for a few hours, polish center over center an hour. Do this a lot, and see if the edge begins to polish out. If it just seems to be going no further after a few, days, you are going to need to return to the second to, and finest grind for a few hours. But, if the halo is now shrinking, check your Ronchi pattern or knife edge for nice, straight lines indicating a smooth sphere. You probably won't have them yet, but keep going with the center over center/cold press routine. After you have a sphere you can discontinue the cold press every time but continue polishing until the edge is completely polished out. And, check the ronchigram from time to time. The lines should stay straight or even straighten more as your technique improves. If they don't stay straight, resume the cold press routine a couple of times. Don't be surprised if this takes a week, minimum. The polishing technique, like grinding, should also comprise waking around the stand and turning the mirror under your hands. Either push it ahead of your walk-around or backwards, as long as it does not stay stationary in relation to the lap. And don't kid yourself about leaving a slight halo thinking parabolising will polish it out. WRONG! You virtually will polish out no glass from there during figuring, it is just not removed from there unless you have to return to a sphere.

 

Lap Inspections

 

Trim the lap, it needs to be kept about 4% or 5% smaller in diameter and make sure the facets are not closing. You can use a wire wheel on a grinder to put a nice bevel around the lap. Messy, but makes a fast, clean bevel. Also, occasionally trim between the facets, too. Look for large chips in the facets and edge or other damage and shiny (glazed) spots. If you keep the lap wet and clean of oils this should not develop. Patching a chipped facet just requires a clean toothpick with a drop of hot pitch on the end applied to a warm lap, repress and trim. Remember the blue shop towel or paper towel between your hand and the lap.

 

What's Next?

 

At this time I would suggest you do not work at this obsessively. You will burn out and that glass will never see light in the end. Now that you have a sphere and it is ready to figure, take a break, as in a few days, for a while and forget about mirror making. You should do this anyway before figuring the parabola to clear your mind. Everything changes here, you will have to re-learn polishing, all over again.

 

As already mentioned you need a parabola at f/5.5, even at 10X per inch the spherical aberration will be intolerable if you stay with a sphere. I am not going to give you the particulars in figuring. Like grinding and polishing to a sphere we each develop our own subtleties. But, generally as a beginner you use a modified "W" pattern. Wide "Ws" in the middle and narrow at the ends. It is usually a 2/3 stroke, center over center and narrows to a 1/10 stroke at the sides with 50% overhang, with, only the weight of your hands on the glass to stabilize it, all while continuing the walk around and and be sure of mirror rotation. See, I told you, you need to re-learn polishing. Take a while to practice with a couple of circular pieces of wood in place of the lap and mirror to develop your modified "W" pattern, with walk around, before re-figuring that perfect sphere you put on the mirror. You probably will not cold press during figuring, unless you return to a sphere.

 

For a first mirror, you may end up going back to a sphere a few times, but have patience. Just, imagine the polish you will have in the end. With an f/5.5 you will probably take several hours to begin to parabolise it. As you get close you will cut your polishing sessions to one or two trips around the grinding stand as you ease into the correct figure. One more time, do not try and figure it without printed or other displayed documentation, YouTube will never work and the printed word is not enough. For a beginner, math formulae and tables of numbers specifying knife-edge spacing's only go so far, you need diagrams, pictures and drawings in a simplified format. That is why I suggest All About Telescopes first.

 

For your first mirror, or three, stay with a 3 zone mask for figuring. It will be good enough for several years of viewing pleasure. Then, save the tool/lap in a box, inside of a tightly closed plastic bag with a hand full of rice or a desiccant bag inside of that, in the freezer. Leave it at room temperature for a few days so the rice or desiccant can absorb as much moisture as possible before freezing. Then, when it is time to re-coat, you can use a 5 zone mask, pick up where you left off, take a few more hours and your improved techniques, and re-figure it to the mirror a perfectionist would be proud of. You may even get lucky and find it is near perfect after all. Oh, when you thaw it, leave it sealed until it is up to room temperature.

 

And never forget the three "Ps", patience, patience and patience.



#11 kfrederick

kfrederick

    Skylab

  • *****
  • Posts: 4,055
  • Joined: 01 Feb 2008

Posted 20 September 2020 - 07:25 AM

 Fine grind  and polish in a way you do not change the RC .  Mirror Matic calls this the neutral spot where you can let it run for days and the RC does not change   And you have NO edge problems .  Working by hand  full sized tooling 1/3 W  and doing half tool on top and half tool on bottom . On a machine spin grinding using a .7 diameter tool spinning offset maybe 18 % .When spin grinding keep checking the RC it should not change . Spin polishing works great you can feel the edge with your finger it is sharp . Correcting a mirror with a good edge can be done by leaving the edge alone and shorten the center . Learn how to take accurate knife edge readings .I like a Couter mask . To get perfect contact leave your tool on the top of your mirror with thin plastic bag between for days . When you ready to polish trim the lap and scratch it up . you have perfect contact even with hard pitch . I never use weights when pressing this way . 



#12 ed_turco

ed_turco

    Mercury-Atlas

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

Posted 25 September 2020 - 02:56 PM

As usual, I see upteen mutually exclusive pieces of advice in addition to plain old-fashioned dogmatism that won't get anyone closer to a conclusion.  I would touch this thread with a ten-foot pole. 

 

I've brought these issues up on a number of occasions; doesn't anyone listen and learn anymore?


  • PrestonE and SandyHouTex like 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






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics