Help in polishing and figuring a 6" mirror

[Garyh]

Hello All,
I am rather new at this so any help would be great.
I have always wanted to make a mirror. So dicided on my first attempt to make a 6" f5.6 mirror. All has gone well through the grinding and have done about 6 hours of polishing with a resin lap made from plate glass disk glued to plywood base with no scratches etc and the surface looks quiet invisible. I built a foucault tester which is a bit on the wobbly side (needs some more modifications) with a knife edge and a ronchi grating of 83 lines per inch. Slit about .003" inch gap.
I have posted a picture of what the surface looks like with both tests just inside and outdise ROC and was wondering if I can go on and try to parabolize or are my errors to large and shall I try to get to a better sphere first? Seems that the centre is flat or a hill?. Knife edge shadow appears in less than .1" more closer to .05". I still have problems reading these shadows but learning quick!!
Also what stroke would be recommended to fix it.
All help very appreciated...
Gary

[cristian abarca]

Hi Gary. I had a similar problem with the mirror I'm making. I was told by my local ATM club that I could go ahead and figure the mirror because the stroke that I would use would eliminate the problem. I decided to fix the problem instead. I did 1/3 strokes centre on centre. Rotating the mirror and walking around as I went along. This eliminated the turned down edge that you have. I did about 15 minutes at a time and retested. This may increase the size of the hill at the centre, but this is easy to remove." Let the mirror overhang the tool and work the centre portion of the mirror near the edge of the tool". ONLY DO THIS ONCE OR TWICE before rotating as you may end up with a hole in the centre and this is harder to remove. You could try larger strokes with larger sideway movement if you are not game enough to try the first method. Remember to cold press the mirror and the lap before doing any more strokes. Overall it looks pretty good. I suggest you buy a copy of texereux it will help you greatly an consider joining http://www.atmlist.net/mailman/listinfo.cgi/atm. These guys will help you out. They have helped me. If you want to figure the mirror instesd let me know and I will show you the stroke that I have used on a previous mirror.

Regards Cristian

[Garyh]

Thanks Cristian I will do some more 1/3 coc and see if the edge gets better. I havn`t been game in working the centre as you described but might give both methods a go tonight and cold press it a few times in between. I shall post a new picture and see if I made it better. I actually borrowed a old copy of Jean Texereau a few days ago from the library to help me along Its old but I suppose that makes no difference. Theres lot to learn!! I can see why people say its quicker to make a 6" first then a 12" than just learning with a 12"
Thanks again Gary

[cristian abarca]

It doesn't matter if the texereaux book is old the technique he uses is still the same. Just remember to proceed slowly and with patience. My first mirror was a 10 inch and when I finished polishing It was nowhere near where yours is. There is a little bit more to do during the figuring process before you call it complete. Keep me posted and I'll guide you through it.

Regards Cristian

[Garyh]

Hello again, It is very much appreciated that you shall help me through this Cristian, shall save me a few hours work! and mistakes....
Here is my efforts last night. I did 1/3 coc MOT for 20 mins. Done a test after 30min cool down. Lines in ronchi test looked good as well as the knife edge with the hill in the middle much smaller as well. Repressed for 15min then done some 3/4 w MOT putting some more pressure on the edge to reduce the centre hill, done this for about 15 min. I don`t know that I have done the right thing here? but heres the picture of it. Still a small hill and a bit uneven, still think there is a slight TDE? It is hard for me to tell if this is diffraction or a real feature.? .......
Thanks Gary

[cristian abarca]

Hi Gary from what I can tell you have probably gone a little too far with the hill at the centre and if the photo is inside radius of curvature then you have actually created a slight hole in the middle. This can be of consequence depending on the size of the hole and the size of the secondary mirror you are going to use. If the size of the hole is smaller than the secondary then don't worry about it if it is bigger then you can try 1/3 strokes centre on centre again and this may eliminate it or you can leave it and start to figure the mirror. When you start doing the strokes to figure the mirror you are actually opening the ronchi lines up a little. which is how the middle picture looks. To test whether it is a TDE or difraction hold a piece of paper just below the TDE and if it looks turned the it is diffraction. What I do is always look inside ROC and try to get 4 lines in the ronchi picture.It is a more accurate measurement than outside ROC. Here are a few diagrams of something that may help. If you can't get good images send me a pm and another email adress so that I can send them at a higher resolution. Try this website this guy is good. http://bi-staff.beckman.uiuc.edu/~me...te/oblate.html
Regards Cristian

[Garyh]

Hi again,
Thanks Cristian for the pics, I found a page online similar to what you posted from someone selling ronchi eyepieces.
You were right I did go a little to far and made a hole, I was getting it back to front. I did about 20 min 1/3 MOT last night and the hole in the middle is practically gone, I am not sure of the edge but, I think it has improved as well. Even gone slightly past a sphere and more ecliptical?
Was very difficult to get the knife edge pics as seemed real touchy.
I am so supprised how quickly the surface changes after such a small amount of polishing!! Would I be ready to try to parabilize from here? Also I have noticed that there are small depressions etc (dog buscuit?) are these worth worrying about or would these be too small to degrade the image.
I would be very happy with my first mirror 1/4 surface error or better!
Thanks Gary

[cristian abarca]

Hi Gary that looks much better. The edge is slightly turned but if it's not too wide (You can measure it with a ruler) you can always beville it out. As far as the dog buiscuit or lemon peel effect goes if it's only amall amouts I wouldn't worry about it. If it's really severe you may have to go back to the last fine grinding grit and repolish. I wouldn't do this unless it was really bad. As it's your first mirror I would concentrate more on getting the shape correct when figuring. You may end up with apperture fever and may want to build a larger one, I know I have, and getting the right shape is of much greater importance. Getting back to the dog buiscuit effect (According to the literature I have) this is mainly caused by too quick or jerky strokes and can be cured by smooth. firm and slow strokes. I would probably try a little longer trying to remove the turned edge as long as I didn't create another problem an then go onto figuring. The way it is now it is ready for figuring. My first mirror ended up with a fair bit of dogbuiscuit and I have a bit of trouble getting the image to focus exactly. (We found this on some factorymade telescopes as well and some of these are used for photography and it's not noticeable) It also ended up with a turned down edge. Regardless of this because it was my first attempt and I knew that I was going to make more mirrors I was not greatly concerned. I still get good views and is still better than1/4 wave. I was aiming for 1/10 but just mist out. I was told of the defects by more experienced observers but overall it was judged as being a pretty good effort for a first mirror. So if you are going to make more mirrors go ahead and figure. From what I can see so far yours is going to turn out better than my first effort if you get the shape correct and you should be happy with that. As for my second mirror I didn't end up with dog buiscuit or a turned down edge. it was perfectly spherical. I didn't figure it because I was just practising. I finally worked out what I'm going to do next and I'm going to make a Clasical Cassegrain that I can use as an F5 Newtonian as well. I got told on saturday at my local ATM meeting that the figures look Ok so here goes.

Regards Cristian

[Garyh]

Hi Cristian,
I hope to make a nice 10" mirror in the next few years so I need lots of practice! A nice cassegrain like what you are going to do would be tops! I think I stick with a newt. Doing the convex secondary would be beyond me and drilling a hole through the primary would be rather scarey. Maybe oneday.
Well I had a go at figuring. Overall about 4 goes at 10-12 min each. After the first 2 goes the shape started going the right way but so did a lump in the middle and a bit of a TDE . Soo I tried something out of Texereau`s book with tool on top working just around the edge. I slipped twice and put a few scratches across the lap but luckily not the mirror. No damage thank goddness!! I had to scratch my head a bit as to the bumo in the middle? After looking at the normal parabolizing stroke, I worked out that the bump was exactly the same size as the distance from the cente of the mirror stopped where the www stroke finished at the edge. Next 2 sessions I took the centre all the way to the edge and it worked. Bump in the middle just about invisible and mirror seems to have smoothed out real well. It looks very close to my printed matched ronchi images I printed the other day. I will have to fix my tester a little more and get it running smoother and do some measurents. Let me know if you think I am close (you can see those defraction rings real easy in this shot). I still have a small TDE so I might bevel it out like you suggest? What is the best way to rebevel? I will post some knife edge pics in the next few days.
Best regards Gary

[cristian abarca]

That looks excellent gary. The next part is a little tricky but I'll go through it with you in the next email tonight I'm off to work now.

Regards Cristian

[matt]

Wow!

I've never done anything like this and am unlikely to...

But i'd just like to say how impressive this is to observe from a safe distance!

I like the idea there's still folks out there practicing these magical ancient arts. I have no idea what it is you're saying to each other, but it sounds great!!!

[cristian abarca]

Thanks Matt unfortunately once you go down this path and you enjoy what you do it tends to become a little addictive and if apperture fever sets in watch out.

Regards Cristian

[cristian abarca]

Hi Gary do you know how to make a couder mask. Google couder mask program follow yhe links to download a program that will automatically print a mask for you. You will need this mask to reach your final paraboloid figure. If you are not sure just email me. The measurements that you take using the focault test alone may not be sufficient to get better than a 1/4 wave. There is also ronchi for windows. Use this meausurement 6mm or -6mm outside ROC . Make sure that it is four lines only. Print the figure cut out only the 2 centre lines. Trace the outline onto the mirror with a texta. put the mirror on the stand to test and get the centre of ROC, measure back 6mm and check to see if the patterns match. If they do then you are pretty close to what you want. All you have left to do is to measure shadows with the couder mask and you are finished. If this is not so clear let me know and I'll explain it again.

Regards Cristian

[Roger Davis]

Many good articles on making a Foucault tester. See Richard Berry's book "Build Your Own Telescope" (which is probably the easiest to understand) or Texereau. Richard uses a slitless tester whereas Texereau uses a slit. With Berry the formula changes as you are measuring the radius of curvature not the movement of the knife edge. Because the knife and slit move as one you a measuring delta R. Tex measures the movement of the slit to the light source so the measurement becomes 2 X delta R. I have been using a slitless tester for the last twenty odd years and so far so good.
The Ronchi for windows is a procedure that Ted Lumley used, but we used to use wire to take the form. See Laurie Halls webpage:
http://www.turbofast.com.au/astrotel/ronchi1.html
Ronchi tests are usually qualitative, the Foucault will give you a quantitative result.
The resin lap may be the cause of your micro-ripple (dogbiscuit) a true pitch lap minimises it. A good quality burgundy pitch (Gugolz) makes a difference, it's just hard to come by.
As Cristian said, you could mask the TE, but you will reduce your effective aperture. One 14" that I worked on had an edge that was soft annealed, I have the option of either trepanning another 15mm off the diameter or masking the edge. It's simpler to do the latter!

[cristian abarca]

Hi Gary I 've got the 2 ronchi images for your mirror at 5mm and 6 mm measurements. These are the grids that Roger was talking about except these are outside ROC. and you can cutthem out instead of using wires. The distance is the letter (o). See if you can print them to same size as mirror .

Regards Cristian

[Garyh]

Hi Cristian and guys,Sorry for the no posting but had to take my son up to Coffs Harbour for a few days so had to drop everything and just got back.
Thanks for the ronchi images Cristian, I have downloaded quiet a few different programs since and yep Ronchi for windows is one of them. Great program also downloaded a program called couder mask and cauder32. Gee this will make it easy the first will give me the figures and print the mask and the second all I have to do is put those figures in and put in my measurements and it does the rest for me. Anyone have a go at these programs are they reliable?
As to my lap and the dog buscuit, (Roger) I had such problems getting anything I was happy to get the resin. I think my lap might have been a bit on the hard side as only had to trim it every 3 hours or so and you have to push very hard with the thumbnail to get the smallest mark!! Also the pyrex disk I bought was very old and rough but cheap $12 so had to give it a go.
I shall compare the ronchi images tonight and see how close I am to them.
Would it help to get a finer grating also? any benefits?

I have been enjoying doing all this and I would get aperture fever if I had the money to back it up!! I still might tackle making the secondary as yet. Thanks for all the help so far guys , I might need some more help when it comes to reading the zones and shadows..I shall post some more pics as soon as my wife brings back the camera!!!!!
Regards Gary

[cristian abarca]

Hi Gary I haven't used couder32 but I have used both the other programs and they work well. Just remember to use 4 zones for the couder mask and to mask the middle zone with some opaque tape. Divide it in two it will make it easier to see the shadows. You'll see when you print it up. Good luck almost there.

Regards Cristian

[Garyh]

Hi Cristian,
Well I spent a hour or so on the kitchen floor doing some tests (not booze tests!!) and managed to get a few pics with my 300d which I didn`t think I would manage due to its bulkiness but they came out pretty good.
I had to redo the ronchi images as they were slightly the wrong dimensions.
ROC= 1716mm Dia=154mm. I think I am real close, I find it hard to tell exactly as the lines have a very blurred outline compared to the printouts, but in general seem to curve the same (maybe a tad over corrected). The measurements seem fairly correct too and may have varied +/- 0.5mm.
First is 14mm and the second is -6mm.
Do you think I shall continue and do the couder mask tests and see what the figures say? or can you see anything that may need attention first.
Regards Gary

[cristian abarca]

Hi Gary that looks pretty close to me go ahead and do the coudermask test. If you are really close then the measurements on the couder mask should confirm this. Take about 5 measurements of each zone and average them out. When you have the results you have to plot them in a graph and see what you come up with. Also do the focault test measuring the images from inside to outside the field of curvature (.707 I think) and if the measurements add up good(See texereaux). Once all this is done and everything measures up you are done. Once you take the measurements let me know and I'll run them on a program I have called figure 45. I don't know if it is still available but that plots the graph snd gives you the wave measurement of your mirror as well.

Regards Cristian

[cristian abarca]

Hi Gary just found a new version of figure 45 it's called figureXP google it and download. Put your figure in and let me know how it goes.

Regards Cristian