360mm First time mirror grind.

[Rustysplit]

It's an optical illusion. the saw is laying on the edge of the glass, the wooden safety stop just happens to be the right distance in front to appear to be under it.

I think, hope, it is the shadows that make the curve look uneven Seems alright in the flesh. You had me worried there!!

[upahill]

Lol sorry, didnt mean to scare you, I went back and looked again and it is just my eyes playing up on me.

Wife has the heating up to the point where my eyebals are dehydrated lol

[Rustysplit]

4mm now!! Another hour tomorrow and the hogging out will be done and I can pour my concrete tool

[estwing]

really shifting with this now..your not missing much

[upahill]

Dont forget to take pics of your process. I started reading Texerau(sp?) book last night, how quick was the delivery of your blank?

[Rustysplit]

estwing, Sadly you are right about the clouds, not seen a dark sky for ages...

upahill, here you go.....

The blank was delivered in a couple of days from Galvoptics. I would not buy the full grinding kit from them though. Their quantities seem very small for a largish mirror and their cerium oxide price is through the roof, 3x that of what is easily available on ebay. They do not supply the coarse grits for hogging out either in the kit. Unfortunately, I found all this out after it arrived!! C'est la vie

[Rustysplit]

Finally!! Just a bit over 4.5mm

Now to mix some cement and pour a tool for the fine grinding.

[Rustysplit]

And now the cement has been poured for the fine grinding tool... first off the cardboard dam with greaseproof paper covering the glass

and with the cement poured in and levelled

[Rustysplit]

Having poured the a full size grinding tool base, I guess now would be a good time to pour a second pitch tool base. But what diameter should it be? I am thinking sub-diameter, but unsure as to exactly what diameter.

[Ajohn]

I would stick to full sized if I were you. You are much more likely to get a perfect sphere after polishing that way. Always finishing off grinding each grade with shortish varied strokes with tool on top also helps that aspect. I make small laps as needed with what ever is at hand eg metal disk of reasonable thickness.

I suspect that you will run into trouble trying to figure a mirror from anything other than a perfect sphere but if you want go any other way feel free. Knifing a sphere like that is amazing fun and good practice for later on. The idea is to make it grey out evenly - not just go pitch black.

I can't make a Texereau slit cropped up some where else so I sorted out these video's for some one. Slit and source move together and easy to make. Still uses the same sort of slide as Texereau but makers didn't buy a junior hacksaw and a bit of metal to make Texereau style V's. Tester is far too chunky and deep in my view too. I would make both ends so that they can be clamped down to save setting up each time it's used. It would also be a good idea to use a longer lever length to set the tilt. Telescope on them pass - I haven't found the need. Maybe plastic binoc's sawed in half would be easier.

http://www.youtube.c...r/carluchoparis

I would look on ebay for a cheap micrometer spindle rather than use the screw as shown or try and graft a digital calliper into it. How long a movement needed depends on the mirror.

You've convinced me shot blasting grit is a good idea by the way. This isn't bad progress especially after the way you started.

John

-

[Ajohn]

Having a bit more spare time.

The slit and the knife on this style of tester move together so the measurements are 1/2 what they would be if the slit remained stationary and only the knife moved.

Personally if I made one like this I would base it on an an L with the lower part facing the mirror and a web running up off that to stiffen up the vertical. I would also put the knife and source on the side facing the operator or better still mount these parts on a small thinnish board that fits onto the main unit again on the side nearest the user.

I would still copy Texereau's slide and add weight to stabilise it as he does.

The advantage of this method is that the source and slit can be very close to each other and there are some indications that slit and source moving together more closely matches theory. The distances that are measured are 1/2 'd though. Best thing to do is to work out what that means or get some foucault software and see how accurately the distances have to be measured. Feed in some fictitious perfect figures and vary them. Basically if some one want to make a 1/8 wave mirror there is a need to ideally measure to at least twice that accuracy.

On the other hand Texereau reckons his tester is ok down to a mirror focal length of 1 mtr. These days it would be easy to make a smaller light source. One advantage of his arrangement is that it's easy to add other gizmo's to the knife if needed at some point. He uses the same basic tester for a couple of other tests.

The formulae that is usually given for measurements is an approximation by the way. The full thing is a series. This may matter on some rather fast mirrors. I'd hope foucault software does take care of that.

John

-

[Glasspusher]

A full sized a lap is OK for a 14 inch mirror but you could use anything between 10.5 to 14 inches. A sub diameter lap has the following advantages:

1. Uses less pitch, 2. Easier to keep good contact between the lap and the mirror, 3. Requires less effort to push back and forth.

With a sub diameter lap the mirror is worked face up and needs to be rotated regularly to prevent astigmatism. The mirror does not have to be polished to a perfect sphere prior to figuring, it needs to be a smooth figure free from zones. With a 14 inch mirror I would parabolise using a 7 inch diameter lap.

John

[Ajohn]

True John but a true sphere will turn into a parabola more easily without so much risk of developing problems round the edge which can be tricky to fix. The usual method is to work the centre down - with a sub dia lap on a mirror of this size I would have thought. 50% is the usually recommended size for that but some seem to find 40% better. Next time I do one it will be 50% with the pitch tapered off near the edge so that it's a sort of 40% plus lap. I'm impressed by the way the quartz mirror guy works in the video link I posted.

Anyway if some one spends a fair amount of time on each grade of grit slowly walking round the mirror using a mix of figure 8 and W strokes plus the usual varied paces they will finish up with a perfect sphere fairly easily. The strokes are harder to do so more randomness creeps in which helps things go spherical. Same with polishing. The whole process depends on randomness really as the only shape where things can be moved in any direction while remaining in contact is a sphere. Realising when to press the lap and how much to do helps too. Some one who does lots of mirrors wouldn't really bother with the sphere aspect but will be very familiar with correcting parabolas.

It turns out that all of the Foucault software I can find doesn't give a truly accurate figure on some mirrors that people might make. On this one for instance the measurements they compare with will be nearly 0.01mm out on the extreme edge if it's an F5 mirror. Worse if faster or bigger. That's just enough to bias readings as many use an 0.0005in dial gauge for measurement. Not that this will make that much significant difference in terms of wave front error. There is also the question of if a tester can read the shadows that accurately. On the other hand professionals making observatory gear used to measure to 0.0001in and average the results of several readings.

If anyone bothers to read Texereau's book by the way in his later days it seems that he tempered his pitch with a little castor oil. I've used turps substitute and even a little tends to be a bit disastrous. He also used his pitch over and over again. Me thinks it will pay to buy some good stuff.

John

-

[Glasspusher]

Just to clarify, the 10.5 to 14 inch lap size is for polishing not figuring, that would be done with a 50% lap.

John

[Rustysplit]

Thanks John. I have poured 2 full size tools.When should I make a 50% tool ?. I am guessing now so as to avoid the risk of scratches?

[Glasspusher]

The 50% tool, to be used for parabolising, can be made on a base with a flat surface, it could be a thick piece of plywood, aluminium, glass or cement. The curve on the 50% tool is less than on a full sized tool and can therefore be poured onto a flat disk saving the need to 'cast' a tool base against the mirror. Of course, it would not do any harm to do this. Hope this helps.

John

[Rustysplit]

I will worry about the figuring tool later I think. I have cleared up all the cement crud now and don't really fancy making any more. I will machine out some ply discs and glue them up to give a nice solid base. Thanks gain for the guidance.

Ajohn, interesting little testing contraption in that vid. I will be looking into the testing issue next week sometime, when I will have a bit more time to sit down and take in what I am reading. Work is so busy at the moment, my mind is just about full.

My task for this weekend was to pour the tools, and get the tiles epoxied on.....

...task complete

[upahill]

Looks good. will you coat the cement in anything to prevent dust,chips etc from falling off? Shouldnt need it on the base that looks like a nice thick layer of epoxy

[Ajohn]

I suppose I should mention why I think you should use a full sized lap for polishing and get to a very good sphere before figuring it. If your mirror is an F5 the difference between a sphere and the parabola using a formulae just taken out of a book is about 5 waves or 0.0025mm as this one uses yellow light waves. It's too easy to polish in errors that are much greater than that with substantially smaller laps. It's best to use a heavy rigid lap for 2 reasons. The weight speeds up polishing and the lap has to pressed periodically to maintain contact so it needs to be rigid. I use a 2 gallon bucket of water for that. At your size something heavier might be a better idea or long periods of pressing. The other aspect of a heavy lap is that it's less likely to tip in use.

The whole idea behind mirror making is randomness and the fact that the only 2 surfaces that can move in any combination of directions and remain in contact is a sphere - or a flat. This can help in fine grinding as well. It's easy to try V,W and figure 8 strokes from time to time while grinding. It's even possible to walk slowly round while doing them rather than taking variable steps. The fact that you wont be able to do these strokes to start off with doesn't matter. The only important thing is low levels of mirror or tool overhang at the end of the strokes - say variable 1/3 to 1/4 and sometimes even less. More variability again.

The usual problem with polishing is that a close look with a torch at various angles shows small pits at the worst or semi polished scratches. I've been there. They are caused by not fully grinding out the previous grade of grit. It can even be the leavings from a much earlier grit. They are masked by the grit being used rubbing about in them so that they are hard to see. They also at least part polish up for the same reason. As a certain book mentions the only safe way to check needs magnification probably aided with a torch at various angles.

Figuring lap sizes people actually use vary in practice. Loosely from 30 to 50% dia. The reason is that as mirrors get faster the difference between a parabola and a sphere gets larger so the lap has more difficulty in remaining in contact / producing a smooth even surface. The trouble with small laps is that they can dig a huge hole in the middle. The problem with the full sized lap for this is over hang and turned down edges especially on mirrors like yours. It can be difficult to control. Texereau gives a good run down on ways and problems. A larger star lap might be safest. Only ever used a small one so pass on size and the amount of star.

It's worth copying Texereau's slide what ever type of tester you build. It's dead solid and easy to make. Most people have copied it. The video shows a similar arrangement but it's easier to screw tube or rod down to a board and cut some V's with a junior hacksaw. I had brass but aluminium should work just as well. Mine has a micrometer spindle but these seem to have become rarer and far more expensive. 25-50mm and 1-2in used micrometers tend to be dirt cheap and shouldn't be difficult to adapt even if it means sawing off most of the frame. You'll find out why a micrometer spindle or screw is best when you try to make a sphere grey out rather than just go pitch black. The tilt adjustment needs to be subtle for that too. It can be done by hand with a simple wooden block carrying a knife but wish you luck. The screw spindle also helps when readings are taken as it's easier to move the knife very smoothly while watching the shadows. I suspect my tilt adjustment is going to become and M4 fine bolts with a fair sized "dial" on it. A straight 1/4 bsf bolt is tricky to use.

I had problems to start off with using a coudre mask - there are other ways of using masks that can help. Pins too but you are a some way of that stage at the moment. I think the main problem in this area is that people have little choice. They get to the stage of using a tester long before they know how to use it. I suspect most people make life difficult for themselves before they learn that aspect. Pretty sure I did.

John

-

[Ajohn]

Here's an example of what can happen - the time spent could easily be half way to a full polish or more.

This one shows shadows. Unusually it mentions slit width. He's using a wide one to weaken the shadows. The movement is a bit strange in places.

Some of these show a mirror more similar to yours - polished on a machine. It's not worth the effort of making a machine unless you want to or are going to make lot's of mirrors. The comments they have posted are of interest.

http://www.youtube.c...nsoptics/videos

The red dot tests without any info tests are most probably hartmann test or a variation of it.

It seems the elevated guide tester shown in the other video's often with the wood work the wrong way round is down to stellafane. Oddly I don't find that surprising.

John

-

[Rustysplit]

Tonight I started the fine grinding. About 3/4 hr on 180 grit and I think I have got all the previous pits out and have an even surface all the way to the edges. I will check again in good light tomorrow with fresh eyes.

The tile tool has bedded in really well. The pencil line test shows that I have very good contact between tool and mirror. The cross has completely gone in just 4 complete passes of the tool. It seems to be quite evenly worn...

The curve depth has remained unchanged at a shade past 4.5mm.

I have to say that if anyone is thinking of doing their own mirror, stop thinking and do it!! Although I have a long way to go, I am thoroughly enjoying the process. It is strangely relaxing and very rewarding at the same time. Slowly watching the curve appear is quite magical. As someone who knew absolutely nothing about mirror grinding when I started, the help and encouragement from all the people on here and the physical act of starting the grind, has really begun to make the theory of the different techniques makes sense. Something that I had thought of as being far too involved to even contemplate, now seems nowhere near as daunting.

Now watch me go and muck it up

[8kids]

This may be a silly question but what level of noise does the process generate, I fancy having a go but will I upset the neighbours.......and more importantly the wife!

Gary.

[Glasspusher]

I have to say that if anyone is thinking of doing their own mirror, stop thinking and do it!! Although I have a long way to go, I am thoroughly enjoying the process. It is strangely relaxing and very rewarding at the same time. Slowly watching the curve appear is quite magical. As someone who knew absolutely nothing about mirror grinding when I started, the help and encouragement from all the people on here and the physical act of starting the grind, has really begun to make the theory of the different techniques makes sense. Something that I had thought of as being far too involved to even contemplate, now seems nowhere near as daunting.

Alan, I think you are hooked. Above all you need to enjoy the activity which you clearly are. Looking good so far.

John

[Ajohn]

Noise - hardly any at all.

I hope my posts haven't put any one off trying. All I have tried to do is encourage people to read Texereau because if they do they will get there. If they listen to people who regularly make mirrors they might not. Simple as that.

The only problem with testing is that people make testers and then use them before they know how too. I don't think the web helps much in that respect other than video's may show you what you might see. There is some one building a stellafane type tester on a yahoo group. The table runs on wooden V's and wont settle down properly, the spring pressure of the dial guage is moving the table. Fanatastic. No doubt some wonderful person came up with the design. Some people have used a screw adjustment with it and it still wont settle down without extra springs. Personally I think people should think in terms of buying a junior hacksaw and a set of warding files. Those get called all sorts of things but the blades of the flat one in the set will be about 10-12mm wide and 100 to 150mm long. They are cheap. B&Q sell odd bits of metal. Maybe there is an old brass hinge kicking about - all sorts of things can be used. The main problem with a screw adjustment is that people are likely to have to make a dial to go with it. Micrometer spindles seem to be much rarer and more expensive these days. Used 25-50mm and 1 to 2in micrometers are dirt cheap however and it might be possible to adapt one. Then there is the marks on paper etc. There are lots of options. I reckon a dial guage is the worst of the lot for most people.

John

-

[Rustysplit]

Gary, not sure I agree with Ajohn about the noise. The fine grits, 180 and finer, are pretty quiet for sure, but the 80/120 grit I used for hogging out the initial curve made a reasonable amount of noise. The sound reminded me of a power hacksaw, the reciprocating type. Not enough to upset the neighbours unless you are doing it at unsocial hours. As to the Wife, only you know the answer to that

If you do a quick search on youtube for mirror grinding, you will get an idea of the noise level for rough grinding.