October 15th 2008
This project has been ongoing for some time. I actually started on this one first but decided against grinding it before I had experience with a smaller size.
It started as a full thickness Pyrex blank from Vancouver Telescope, which I got for $100 because someone had started it, and made a mess of it. It had an uneven grind to it, resulting in an off-center depression in it that required some 5 hours to center up. Once that was done, it only took another 5 hours of grinding with 46/70 and 80 grit to get the f ratio to around f/5, giving it a 63 inch focal length, a good compromise in shortness of tube and magnification, a wider field but still doesn’t have large amounts of coma.
There were pits left from the 46/70, so I gave it 5 more “wets” of 80 grit to level it a bit more.
The tile tool was easy to make but I learned the hard way to take my time. I poured the base out of plaster in 4 separate pours to give me a 1 3/4 ” tool. I used epoxy to fix the titles but there was one problem with them as they tended to ‘float’ a bit and slide down the slope of the tool. I solved this by using a fingertip to bed them in when placed. None moved after that. I heard later on that some have problems with the pieces falling off, but those in ‘pools’ of epoxy did not tend to do this. Given the size involved, I didn’t bother to go completely to the edge with tiles by cutting tiles, just used the broken ones. This way, the tool acts a bit like a sub-diameter tool. After 220 grit, I cleaned it up with care, using epoxy to fill in holes or fix suspicious areas.
I used a bit of extra weight on it in the form of one of my mounts counterweights, one that was around 15 pounds. Later on, I used just the weight of the tool and my hands on the back of it. I theorize that a turned edge is caused by excessive overhang on a tool that has too much weight for its thickness, which lets the tool flex slightly on the overhang, resulting in the start of a turned edge. I also took my time in doing the 1/3 strokes with the tool. I did make a mistake in this stage though. I was letting the blank rest on the bare formica of my plywood backed surface thinking there would be no trouble with astigmatism. The blank has a 1/4″ raised rim on its bottom, which supports ALL the weight of the glass blank while I work on it. I checked contact when I changed grits at 220 and found I had good contact, so I used some old rubber sheeting and cut three circles of it to rest the mirror on when working “tool on top” or TOT.
I used a straight-edge and some feeler gauges to measure the depression, but I now own a spherometer good to 5 decimal places. I found I overshot my f/5 to f/4.8, so I did 120, 220 and now 320 grit with the tool on top. This is actually easier but I am not learning what I wanted from this. By changing TOT and MOT, you get control of the depth of your curve, very important if you plan on making refractive optics.
After 220 grit, I poured plaster and dental stone for the backing of the pitch lap. The finer grits after this will not change the curve all that much, so the plaster cast will be close enough for the pitch to conform properly. I found the dental product, while being slightly more expensive, is much more fluid when mixed. I plan on using it for all my later projects.
I have also made plans for the construction of the mechanics of the scope. I really like some of the light-weight upper cages connected to the mirror enclosure with trusses. I plan that route. I found some of the upper cage assemblies a bit flimsy though. Too much empty space with the eyepiece looking at open air behind the secondary. Lots of people put up thin plastic shields to combat this issue. My plan is to use my router and a 1/4″ straight bit on 5/8″ baltic plywood. I will make two rings with a shallow cut in the middle. I will then bridge the two parts with an epoxied piece of formica (cuts well with a good blade on a table saw), and use truss poles to link upper and lower sections. I will build a small version of this for my polished 6″ f4.7 to see how well it works before commiting to a big version.
For grinding with 320 grit, I started with Al2O3 instead. I found this was a bit slow, so I returned to silicon carbide after just 4 wets. I did 15 more this way, then finished up with 5 wets of aluminum oxide. I got this in a box of grits, from Aurora Astro who got it from Bill Cook, of the ATM book fame. He’d had enough and cleared out his stuff. I did this as an experiment mostly. The Al2O3 grinds more slowly in a more controlled fashion, important for getting to and keeping a curve at the correct amount. For a tough glass like Pyrex, it is rather slow at 320 grit. It also seems to break down a bit slower, confirming what more experienced workers say about it. It has grains which have a flat red blood cell shape, not a sharp, angular form.
I now have to clean up my work area to prepare for the fine grinding stages. I don’t want to return to 220 grit because of a mistake. I set the tool aside on a shelf where it can keep until I’m certain I won’t need it again.
The plaster for the polishing tool is dry but needs some clean-up to make it nice and smooth. I nearly used SANDPAPER on it but used files instead.
I used the 25 micron on the mirror and did some 25 wets. Some people like to use more but I plan to go through 15, 9 and 5 micron as well for a smooth surface. I had to take a break and repair old holes in the tool with epoxy resin to trap any old grit in them permanently.
January 11th 2010
I have gone through 15u, 9u and 5u; 30 wets each with one minor mishap. The tool stuck fast to the mirror at around the 4 1/2 minute mark of a 9u wet, needing force to separate them. I had been playing with the ratio of grit to water and I believed it was due to the grit being worn too far, so I shortened the wets to 4 minutes from 5. No further problems arose. The mirror now has that buttery smooth surface to my fingers and the hint of mist it holds disappears when I breath on it, the signs of proper fine grinding. No pits seem to remain.
I built a 8″ and 12 1/2″ pitch laps. That went well, in spite of forgetting to use parting compound (an alcohol spray mix) on the mold. It came off but took a few tiles near the edge with it. There was one damaged in the interior area and I might have to replace it to avoid a zone being created.
I ran out of my old hard pitch, so will need to open a container to make other laps. I did add a few tablespoons of turpentine to each batch to soften it up a bit.
It did a variation of warm pressing by using a bowl of hot water that was steeply curved, allowing the lap to be suspended over the water, making it a bit more controlled with how heat is applied.
I did try to do some polishing and it seemed to work but with catching and uneven friction, the classic problem with incomplete contact between mirror and polishing lap. It will need more pressing to even it out. The main problem now is while humidity is high enough to slow evaporation of the rough/water mix, the temperature is too low to get effective polishing action. I will have to wait for better weather to allow the shop to be warmer. It is also quite hard to move. With 4 times the contact area of a 6″, the amount of work is that much larger. I seriously doubt I could do a mirror much larger by myself, and I was optimistic enough to buy 2 more 16″ blanks. I took some photos that show the wide polished area on the outer portion, with a central blank area that doesn’t show a reflection at all. This is the classic result of only polishing with the pitch lap in the upper position. Most books recommend you swap them to go evenly. I did it this way because as you polish pitch lap on top, the polish moves from the outside to the center of the mirror. If you check the center for remaining pits, if the center checks out fine, then you KNOW the rest has a complete polish.
The intervening weeks have been a mash-up of work, weekends away, the flu, cool conditions and now a shoulder/neck injury causing a pinched nerve. With the shoulder usable again, I restarted my efforts. I had wanted to finish in time for Table Mountain in August, but that seems unlikely now. I did the usual warming of mirror and lap, 30-45 minutes, followed by at least 30 minutes for warm pressing. I used a red plastic mesh bag for onions between them to add micro-facets to the lap. I had wanted to do this before but didn’t have a piece of mesh that could cover completely. This does two things: it forms the pitch to the mirror AND brings them both to similar thermal conditions. This will give the best polishing action and a better spherical surface. I am trying to maintain a better form for the polish stroke also: 2 inches or so back and forth, some variation in length and placement of tool, smaller number of strokes say 6 to 18 per position before moving and letting the rouge thin out and adding some water to lengthen each wet.
The lap is very hard to push around and changes to the rouge mixture does little to alter this. In most cases I only managed 1 1/2 hours of actual polishing time. It is true that the longer you go, the better the two pieces accommodate each and the better the figure you have. So far, there have been a total of 9 sessions with perhaps 12 hours of polishing. I have changed my cleanliness as well. My shop has woodworking tools in it and bits of grit on the floor. I have avoided cleaning it out, as this would raise dust contamination making the possibility of scratches while polishing. In between work sessions, I put each piece into a plastic bag folded over and weighed down to prevent breezes or drafts from getting dust inside. I am considering using mirror on top to finish but that means pushing around a 24 lb mirror, not the 12 lb tool.. Also rouge would not go as far, and since the mirror has to be lifted more often, it increases the likelyhood of dropping it.
Some recommend polishing the mirror face up and then face down. The mirror will polish from the other edge to the inner edge when the mirror is on the bottom and this will be reversed when the mirror is on top. I did it exclusively with the mirror on the bottom for this project. My rational was this: many mirrors fail due to poor or incomplete polishing. It LOOKS good but you can only know for certain once the coating is on. A bad polish is now unmistakable. if you do all your polishing with the mirror on the bottom, once the center polish has a complete polish, you know the edge parts must also be good.
Since the mirror now holds some polish across the entire surface, I have started testing it. The Foucault tester from RASC still works, using an old 12V lamp for light. It is a virtual copy on the one in Texeraeu’s book, right down to the brass pieces used to make the slit light source. I removed that to give more light to help me align the apparatus. I puzzled over how to do this and realized that I could use a sheet of paper to find where the reflection was ending up. I had to use a mirror support meant for an 8 inch mirror. I realized I could just tip the 12 1/2″ forward, rather than raise the tester up some 2 inches. There are some tricks to know: the distance between source and return is the geometric center. The image should be in focus when it gets back to the tester, and pass between the light source and 1/8″ from where your knife edge is. You also need to have the TRAVEL portion of the knife edge at 90 degrees to the returning beam or the shadows that you see will be meaningless. Once I did that, I replaced the slit mask. With a few adjustments, I could now see a squarish oval with a series of interference bands running vertically, more or less what should be visible. I need make a knife edge, as this is missing. I though I could use a piece of stainless steel for it and tired to cut it on my table saw. It went flying off like the piece of shrapnel it had become.
I am using a 8″ pitch lap for parabolizing/smoothing this large mirror and have found it has poor contact. I rebuilt it using softer (added 4 Tbs turpentine) pitch for better conformity. I have used this for 4 spells of polishing totaling some 5 hours. With my semi-working left shoulder 80 minutes is the maximum I can go without serious pain or compromised polishing. Once you get to large mirror, sub-diameter pitch laps become the norm: they weight less, and need less force to push and pull around.
Tests have looked better now, with a trace of roughness present and it seems to be of the shape called an “ellispoidal”, half way from a sphere to the classic paraboloid. I am still thinking about my support structure. A classic Dobsonian would be easy but also a Porter Equatorial is workable with not that much extra labor.
November 12th 2010
I have decided a Dobsonian is the best choice for this scope. The upper cage is completed but still needs stain or paint. I build it out of two 5/8″ baltic birch plywood rings, connected by four 1/2″ fiberglass tubes I bought for some other project. The spider is a Gary Wolansky, attached to four 1/16 thick brass supports. There is a 3.1″ secondary to go with it but I am considering using a 2.6″ instead. Total weight is around 3 pounds.
The cage will connect to the rocker box/platform using three 3/4″ aluminum tubes. I am making my own split connectors but will have to modify them for a better result.
February 13th 2011
I did another spell of parabolizing using the 8″ pitch tool. The mirror appears very smooth now with a pronounced depression in the center 60% indicating I am most of the way to having the correct shape. There is May 7th star party at Aldergrove Regional Park which I hope to have this scope in attendance.
April 19th 2011
I have worked on it some more now. Our house is being demolished, so we have to find a new place to live by June 30th. Luckily, we seem to have found a house we can afford, in a place we both like. My back acted up again three weeks ago, so heavy lifting of a delicate Pyrex mirror was out of the question until it was better. I did 3 more parabolizing sessions totaling some 2 hours. The figure observed in the Foucault test is getting close an ideal appearance. There is a hint of astigmatism, but this could be due to its lack of support during testing. Here is a photo of it:
The use of the 8″ pitch lap worked well until I rebuilt it due to the pitch being too hard for proper use. The first attempt resulted in one that was too soft, so I rebuilt it after simmering the pitch for 30 minutes. I had to also fix the tester as a bulb had burned out in it. The shadows after just another 75 minutes seems rather muted, like it has been SHALLOWED out and not deepened. This may be from the soft lap or the tester being changed. I don’t know at this point. It looked very similar to a 12″ f/5 done by someone else before the lap rebuild. In this photo you can see the central depression. The stuff on the lower right and upper right are just a bit of polishing compound left behind.
A telescope of this size has an advantage over smaller ones: you CAN star test it WITHOUT putting a coating on it. Even BARE GLASS will reflect around 6%, making its light gathering power equal to a 3″ COATED mirror.
I have had the mirror cell welded up out of 1″ square 1/8″ wall thickness tubes bookended by a 1 1/2″ by 1/4″ aluminum bar. This will allow me to have the mirror ride in a swing–down tailgate. This can allow the mirror to be cleaned, even washed without it being removed from the lower plywood mirror box. With the final outside width established, I can make the box to hold it.
I have constructed the mirror box now. It has an interior dimension of 15 1/4″ just over the width of the mirror cell framework. The material was baltic birch plywood, very dense and capable of fine joinery, in this case 1/2″ wide finger joints. This joint is very strong when done tightly and only requires a table saw capable of using a dado blade. The jig is simple and can be made yourself. The only mistake making it was when I glued it up and failed to arrange it so that the tear out was on the outside of one joint. This mistake can be mitigated with the use of a wood veneer covering over the outside. Not a trivial solution, as you need to “seize” the veneer to make is pliable and flat by wettening it with a diluted glue/water mix and then pressing it to flatten it and equalize the water content. I can also add wood trim top and bottom. For a more durable surface, I could put on formica surface.
The weight is a bit higher than expected, so I may have to lighten it somewhat.
Not much remains to be done; connectors for the three tube members connecting upper ring to mirror cell, attitude bearings and a rotating platform. Maybe even this year it will be done. I do want to refine the look somewhat to make it more like a profession woodworking/mirror making project but that may need to wait.