How I make Telescope Mirror Blanks
Fusing thin disks to make thick slabs

A finished telescope mirror blank. This is a full-thickness, 8 inch diameter telescope mirror blank that I made myself. This web page will show how I did it.

Why make my own mirror blank? Why not? I build telescopes. I have built a lot of telescopes over the last 20+ years. I like to fabricate as much of each scope as possible myself. The one part of the telescopes I have never made for myself is the mirror blank. I know a guy who built a large trepanning saw and cut mirror blanks out of thick scrap glass. The machine was huge, took forever to cut the glass, and finding thick enough glass was very difficult. I had always thought it might be simpler to make a mirror blank by fusing together a stack of thin glass disks in a kiln. This blank is the culmination of months of experimenting ever since I picked up my kiln cheap at a yard sale.

My Duncan kiln. Here is the Duncan kiln I got at a yard sale for only $95. A great deal! The kiln only has a 12 inch inside diameter, which will limit the size of the blanks I can make, but it can be powered by any 120 Volt outlet, which is very nice. I have no 240 Volt outlets at home or in my workshop.

I spent months after acquiring the kiln experimenting with fusing glass. I ruined a lot of glass, but I learned a lot.

A stack of glass ready for cutting. I prowl thrift stores looking for cheap glass. I'll often buy glass tabletops or shelves to cut up. Here is a huge pile of glass tabletops I bought for only $15. It is a mix of 3/8 and 5/16 glass in round and square pieces. The smallest piece is the round one at 24 inches in diameter. What a bargain. It must be over 100 pounds of glass. It will keep me happily experimenting for months. Thrift shops often get glass tables or display cases that are damaged or incomplete, but the thick glass is in reasonably good shape. The thrift stores will usually sell damaged or incomplete pieces quite cheaply. They are usually just happy to be rid of it. I always ask if they have any thick glass in the back room that they are willing to sell. The thicker the better. If it has any chips or nicks in it, that's fine, because then I know it isn't tempered glass, and it can be cut.

The glass must not be tempered, or it will simply shatter into tiny pieces if any attempt is made to cut it. If a piece of glass has chipped edges or deep scratches, you can be sure it is not tempered and can be cut. If the glass is perfectly intact and has kiln polished edges, the glass is most likely tempered. A pair of polarizing filters can be used to look for the stress patterns tempering leaves in the glass when you aren't sure about a piece of glass. Don't waste your time with the tempered stuff.

I prefer to use glass all cut from the same sheet, if possible. That way I don't have to worry about mixing incompatible glass types. If not from the same sheet, then at least from the same display cabinet or glass shelving unit. That way I can be reasonably certain the glass is at least from the same lot.

Four glass disks ready for fusing. Here is a stack of four 8 in diameter disks I cut out of 5/16 inch thick glass. I've gotten quite good at cutting circles out of glass. It's taken a lot of practice though.

Note that I am using ordinary plate glass here, rather than borosilicate (Pyrex) glass, which is preferred for telescope mirrors. People ask me why I don't melt down Pyrex cookware and cast glass blanks. There are two problems with that idea. First off, Pyrex cookware isn't really made out of true borosilicate Pyrex glass. It is made out of ordinary Soda lime glass that has been tempered. There is something of big controversy and even lawsuits about the use of the Pyrex trademark on ordinary soda lime glass. Google it if you are interested in the details. so there is no advantage to using Pyrex cookware. Also, melting and casting glass is a lot more difficult than just fusing together glass disks. Soda lime plate glass will work fine for an amateur telescope once it reaches thermal equilibrium.

I uploaded a video on how to cut circles out of thin glass to Youtube a while back. I've made blanks out of stacks of a dozen thin pieces of glass, just because thin glass is so easy to find and cut. I have even had people give me piles of thin glass for free just to get rid of it. So my first experiments were done with thin glass.

Here is another video I made after I decided to switch to using thicker glass for the blanks. I can make an eight inch blank out of four layers of 5/16 glass, rather than a dozen or more layers of thinner glass. The advantage of this is that any air bubbles trapped between layers will be at least 5/16 inside the blank, and won't be reached during mirror grinding. However, I had to learn how to cut circles all over again, since thick glass is a lot tougher to cut circles in, until you know the secrets.

A k-23 firebrick. My early attempts at fusing glass taught me that I was going to need a mold to constrain the glass and prevent it from running. I don't really want to heat the glass up until it flows like water, but just enough to soften it so the layers fully fuse together. However, at the high end of the fusing temperature, the lower layers of glass will get squeezed out like toothpaste by the weight of those above. So a mold is needed to constrain the glass and keep it in a disk shape.

I have tried making molds out of lots of things. I've had the best luck using firebricks. This is a K-23 firebrick. They are light, porous, soft, and easy to cut and shape. They will withstand temperatures far in excess of the melting point of plate glass. The bricks are 9 inches long by 4 1/2 inches wide by 2 1/2 inches thick. I use two of these bricks to make a mold for 8 inch diameter blanks.

A large hole-cutting saw. I made an 8 inch diameter hole cutting saw out of a piece of 8 inch steel AC duct and some scrap steel strapping. I welded an old drill-mount nut driver to the center of it so I can chuck it up in a drill. It only took a few minutes to build, and it works very well for cutting the firebricks.

A close-up of the saw teeth. I used a metal nibbler to cut teeth into the edge of the hole saw. This makes the saw cut through the soft firebricks like a hot knife through butter. I was amazed at how well this improvised saw cut the bricks.

Cutting a hole in two firebricks. Here I have two bricks held in place in a wooden frame, ready to cut with the saw. The saw was a little tough to start since I didn't include a pilot drill bit. Since the bit would be in the seam between the two bricks, it wouldn't be much help anyway.

Cutting a hole in two firebricks. So I cut some small pieces of plastic and screwed them into the corners of the bricks to center and guide the saw. The screws easily bite into the soft bricks. Once the saw gets a good bite down into the bricks, I take off the plastic guides so I can drill down to the full depth of the saw. It is amazing how easily the saw cuts the soft bricks.

The finished mold. This is the finished mold. As well as drilling out the center, I cut off the corners so I could get the mold in and out of the kiln without scraping the edges and damaging the kiln walls.

A kiln shelf. Here is one of the kiln shelves. It will be the base for putting all the pieces together.

Putting shelf paper on the kiln shelf. I start by putting a piece of kiln shelf paper on the shelf. I'm using Bullseye Kiln Paper. It is infused with ceramic. The paper burns away in the kiln and leaves a thin layer of ceramic powder behind that the glass will not stick to. Molten glass will stick to anything it touches, and will need to be chiseled off once it cools. So kiln wash or kiln paper is absolutely necessary to prevent the glass from sticking to the shelf. I prefer the paper. The porous kiln paper also lets any air trapped under the bottom of the glass disks escape and not rise into the soft glass as a big honking bubble. Place the paper on the shelf with the printed side down.

The glass disks stacked on the kiln shelf. Here I have stacked four 5/16 in thick glass disks in the center of the kiln shelf. The glass disks have been edged to true them up a little bit, and cleaned by running them through my dishwasher. Once cleaned, I only handle them by the edges. The idea is to keep them clean so that there is no oil on them, or specks of dust or dirt that would hold the disks apart and keep a film of air between the disks.

The disks wrapped with shelf paper. Here I have wrapped thin strips of kiln paper, printed side out, around the stack of disks. This will prevent the molten glass from sticking to the mold. The strips are held in place with tiny pieces of Scotch tape. I am always afraid that residue from the burning tape will cause problems, so I use as little as possible. Just enough to hold the strips of kiln paper in place until the mold is clamped around the glass.

The mold around the disks. Here the two halves of the mold have been put in place around the stack of glass disks.

Wiring the mold closed. I put a couple of wraps of stainless steel wire around the mold and twist them tight. This is to prevent the hydrostatic pressure of the molten glass from pushing the mold apart. I've seen it happen. It is amazing how much pressure an inch and a half of semi-liquid glass can exert on the mold. Wiring the mold shut is essential.

The mold in the kiln. The whole assembly is then placed in the kiln. It is hard to tell from the photo, but it is sitting on three kiln posts, and not on the floor of the kiln. The kiln has been carefully leveled. This is important. If the kiln is not level, the semi-liquid glass will flow until it finds its own level, and your finished blank will have a lot of wedge to it. My kiln needs to have one leg wedged up about 3/4 inch to make the kiln floor level.

Also in this photo you can see the four terra cotta pot feet I use as weights on the corners of the mold. The pot feet are quite cheap, and easily withstand the glass fusing temperature. The extra weight is needed to prevent the mold from floating up and allowing the molten glass to seep under it. The mold is made of very light and porous material. The glass is very dense by comparison. If a little of the molten glass manages to seep under the mold, the mold will float up, opening a gap which will allow more glass to flow under, causing the mold to float up even higher, until the glass is running out the bottom of the mold. The extra weight prevents this problem from happening.

You can also see the wires of the thermocouple temperature probe coming into the top of the kiln from the upper right.

The kiln at 1734 degrees. Now it is time to fire the kiln. I start off with a very slow ramp up of around 400 degrees per hour until the temperature exceeds 850 degrees. This is to prevent thermal shock from fracturing the glass disks. Then I crank up the temperature as quickly as the kiln will heat up to about 1250 degrees. I let the glass soak at this temperature for about 45 minutes. This soak minimizes the amount of air trapped between the layers of glass. Once the soak time is up, I once again crank up the temperature, as fast as the kiln will heat up, to 1700 degrees. You want to get the glass through the 1350 region as quickly as possible. This is the temperature where devitirification can occur. So I have the kiln at it's highest setting to get to 1700 as quickly as possible. As you can see by this photo, I over-shot the target temperature by a little bit. Let the glass soak at 1700 for at least 10 minutes. This will ensure a complete bond between the layers.

The kiln vent hole. Here is a peek through the vent hole of the kiln with the temperature at about 1700 degrees. I have experimented with various temperatures. 1700 may seem a little high for glass fusing, but I find it makes any air trapped between layers form into little round bubbles, rather than remain as sheets between the layers. I have experimented with trying to increase the temperature to 1900 or even higher in an attempt to reduce the viscosity of the glass to the point the bubbles will rise to the top. Unfortunately, the glass begins to run out of the mold before it gets runny enough for the little bubbles to rise very far. So I no longer go much above 1700. Using thick glass ensures that the first layer of bubbles will be deep enough inside the blank that it won't be reached when grinding the mirror.

The kiln opened while hot. This is what it looks like inside the kiln at about 1700 degrees. I am trying to cool the kiln down rapidly after the 10 minute soak at 1700 degrees. I get the temperature down to about 1200 as quickly as possible. This is again to minimize the time spent in the devitrification region of 1350 degrees, and to simply reduce the length of time the run takes. I open the kiln with a long pole and wearing elbow length thermal gloves. A face shield would also be handy, since the heat coming out of the kiln is very intense. My kiln is outside on my patio with nothing nearby that could ignite or be charred by the heat (except me). I turn the kiln off and open the kiln for 15-20 seconds (or as long as I can stand the heat blast) several times to allow it to rapidly cool down. Once it has cooled down to about 1200 degrees, I close the kiln and let it continue to cool down naturally until it reaches a temperature where the glass is cool enough to handle. This usually takes about 24 hrs.

A rough blank right out of the mold. If everything worked all right, I get a rough blank that just comes right out of the mold and doesn't stick to the kiln shelf. I've had things go horribly wrong though, with the hardened glass stuck to mold and shelf. Since I perfected the above procedure though, there haven't been any more bad incidents. This is one of my early blanks made of many thin layers. The blank is quite rough around the edges. It takes on the shape of the mold, which is rough and full of voids. A little glass always squeezes out around the bottom of the mold and leaves a ridge (seen at the top here). There are also always two vertical parting lines where the two halves of the mold come together. One can be seen to the right of center here. It doesn't look like much in the photo, but it is sticking out a couple of mm from the edge of the blank. I've thought about ways to make the inside of the mold smoother, like coating it with a thin layer of refractory cement to fill in the voids. I may try that in the future. For now though, I use a grinder to smooth out and true up the edges of the blanks. I figure that even with a smoother mold, the blanks will still need a little grinding work, especially at the bottom ridge and parting line.

A blank with the rough edges ground down. Here is a blank with the rough edge ground down and trued up. It only takes a few minutes on the grinding machine I built for the job.

A machine I built to edge the blanks. Here is the super simple grinding machine I built to true up the edges of the blanks. It is just an angle grinder with a masonry wheel mounted on the edge of a table. I also mounted a lazy-susan bearing on a piece of plastic between fixed guides. This allows me to move the bearing toward and away from the grinder. Only a few minutes of spinning the blank against the spinning grinder wheel gets rid of the rough edges and parting lines, and makes the blanks truly round.

A face shield, gloves and high quality dust mask are required for this grinding operation. I only do it outside.

Annealing the blanks. The glass blanks have a lot of internal stress that needs to be relieved. This is accomplished through annealing the blanks. Annealing involves heating the blanks back up to 900 degrees slowly so they don't break due to thermal shock. The mold is not necessary, since we aren't getting up to temperatures where the blank might soften. The insulating properties of the mold might also inhibit proper annealing. Soak the blank at 900 degrees for about four hours. Then allow the blank to cool very slowly, about 50 degrees per hour, down to 800. Then I turn off the kiln and let it cool to room temperature naturally. This annealing cycle relieves all the internal stresses in even thick blanks.

The finished blank. This particular blank was edged, then returned to the kiln for fire polishing and then annealing. To fire polish the blank, it is heated up to about 1300 degrees in the kiln. At this temperature the glass won't melt, but it will flow enough to smooth over rough edges and sharp corners. Once the fire polish was done, I lowered the kiln temperature to the annealing temperature and did a regular annealing cycle. I don't bother fire polishing all my blanks. I just wanted to try it with this one to see how it looked. The fire polish made the rough edge nice and smooth and transparent. It is possible to look through the side of the blank and see the individual layers of glass and the planes of bubbles. It's neat.

Now that I have several blanks made, it is time to grind and polish one into a telescope mirror and test it out. I can use one blank as the mirror and another as the grinding tool, and have a couple in reserve in case something goes horribly wrong during the grinding and polishing operations. I'll post an update once a mirror is completed.

Ok, so this process isn't really quicker or easier than my friend's trepanning saw, However, my kiln is smaller and quieter than his saw, and I don't have to buy or scrounge for hard to find, full thickness glass to cut into blanks. I can use cheap and or free thinner glass.

I am on the lookout for a larger kiln that will allow me to make blanks in the 10-12 inch size range. If anyone knows where I can get one cheap, please email me and let me know.

My first laminated 10 inch mirror blank. UPDATE
I have just completed making my first 10 inch mirror blank. Here you can see it with one of my older 8 inch blanks sitting on top of it. It's not perfect. It's not full thickness, but my next one will be. The 10 inch blank also has a few rather serious chips. These are from the glass sticking to the mold. I had some other issues with the mold too. I am in the process of making a better mold before doing any more 10 inch blanks. 10 inches is absolutely the biggest blank I can make in my kiln. It was a stretch to even accomplish this. The mold fills the entire kiln with hardly any room to spare.

My new 10 inch mirror blank mold. UPDATE
Here is my new mold for making 10 inch blanks. I made it by cutting 1/2 inch wide circular sections out of the same sort of soft, K-23 firebricks I used to make the 8 inch mold above. In fact I used the scrap pieces cut out of the center of the 8 inch mold. I used my band saw to cut them out, and gave them a final shaping on my upright belt sander. I cemented the pieces together using furnace cement. The finished mold is a great fit around the kiln shelf. As before, the two halves of the mold will be held together with several wraps of stainless steel wire.

The first 10 inch mirror blank out of the new mold. Here is the first 10 inch blank out of the new mold. Much better than the one above. The blank came out very round and didn't stick to the mold causing the sort of bad chips I saw with my first mold. That first mold worked so badly that I'm not even going to show any photos or write up how I made it. I don't want to send people down a dead-end street. The mold I am using now is the way to go. It works so much better.

Believe it or not, I ran out of glass. So this second 10 inch blank still isn't full thickness either. It's hard to believe I used up all that glass I bought before. Though I did give away one piece to my sister in law for use as a tabletop.

Making a full thickness 10 inch mirror blank. UPDATE
More Glass!! I found a huge 1/2 inch thick glass tabletop at a thrift store for only $25. It measured 30 X 60 inches and weighed a ton. Too bad I didn't think to take any photos of it before I started slicing it up.

Here three 1/2 inch thick circles are stacked on the kiln shelf and about to have the mold clamped around them. I still have to wrap the glass with kiln shelf paper first. Three 1/2 inch thick layers are close enough to full thickness for a 10 inch blank as far as I am concerned.

My third 10 inch mirror blank. Here is the result of firing the above stack of glass. Once out of the kiln, I removed it from the mold, and trued up the edge on my grinder. It is a fine-looking, near full-thickness, 10 inch diameter blank that I intend to grind and polish into a 1st class telescope mirror.

A stack of four mirror blanks. Here is a stack of my home-made 8 and 10 inch mirror blanks. The second from the bottom is the one above. I can use the thinner blank below it as the grinding tool.

Grinding my 10 inch mirror blank. UPDATE
I have started rough grinding one of my home-made 10in glass blanks. Here is an action shot of me, pushin' glass. I am aiming for an f/5 mirror. I am working with the experts at the SPAC Mirror Lab on the fabrication and testing of this mirror. It will probably take me a few months at least to complete this mirror, and build a scope for it. Stay tuned for further updates.

Click the photo for a larger view.

Polishing my 10 inch mirror blank. UPDATE
I have finally moved on to polishing my 10in f/5 mirror. You can see the pitch polishing lap behind the mirror. After only about 1 hour total polishing time, the surface is starting to take on a nice shine. Working on the mirror for only a few hours, one day a week at the SPAC Mirror Lab, I probably still have a month and a half or so to go before the polishing and figuring is finished and the mirror is ready for aluminizing.

Click the photo for a larger view.

The 10 inch mirror completed and aluminized. UPDATE
Here is my finished 10 inch home-cast mirror just back from aluminizing. It is a thing of beauty, if I do say so myself. I ground and polished it myself too. The next time someone asks me if I made my own mirror, boy do I have a story to tell them.

Click the photo for a larger view.

My first 14.5 inch mirror blank. UPDATE
I have just completed my first 14.5 inch blank in my new kiln. the photos below show how I did it.

Click the photo for a larger view.

Making the 14.5 inch mold. I started by making a 14.5 inch inside diameter mold. I used a hot wire foam cutter to cut out a 14.5 inch diameter circle in styrofoam. Then I cut circular segments out of soft K-23 firebricks with a bandsaw to encircle the foam disk. Later I used furnace cement to glue all the pieces together. I then coated the inside of the mold with kiln wash to prevent the molten glass from sticking to it.

Click the photo for a larger view.

Cleaning glass circles in the dishwasher. I cut three 14.5 inch diameter circles out of a piece of 1/2 inch thick glass tabletop I bought cheap at a thrift store. I have found that the glass needs to be very clean to get a good fusion between the layers. So I always clean the glass in my dishwasher before putting it in the mold. These disks were so tall that I had to remove the upper rack from the dishwasher to get them in.

Click the photo for a larger view.

Assembling the pieces on the kiln base. I disassembled the kiln to make loading the heavy glass and delicate mold into it easier. No more leaning down into a deep kiln to assemble the mold, and scraping my knuckles on the sides as I work. I love this new kiln. I took off the two heater rings, leaving just the base. I first carefully leveled the base. I started with a shelf that was well coated with kiln wash, sitting on three feet. I added a piece of shelf paper, just to be safe. Then I stacked up the three pieces of glass and clamped the mold around them.

Click the photo for a larger view.

The kiln ready to fire. Here is a view down into the kiln after reassembling it around the mold. The object sticking in from about 7:00 is the thermocouple temperature probe. I positioned it right above the mold in an attempt to get an accurate gauge of the glass temperature. The kiln was now ready for firing. The entire firing cycle took about 7 hours. I used a very slow ramp up to prevent stressing out and breaking the large mass of glass. There was also a long soak at 1200 degrees to eliminate air between the layers, and a soak at 1700 to ensure a good fusion. Due to the very large mass of glass, cool-down took a very long time (over a day) before I could handle the blank, and remove it from the kiln.

Click the photo for a larger view.

My first 14.5 inch mirror blank. So here is the completed blank after removing it from the mold and truing up the edge on my grinder. The blank is 1.5 inches thick and weighs 20 lbs, 10 oz. This isn't the biggest blank I can make in the new kiln. I should be just able to make 16 inch blanks in it, but it will be tight. I'll need to find some more thick glass.

Click the photo for a larger view.

I will post more updates in the future. Stay tuned.



[Back to Mike's telescope workshop]    [Email me]

Other places to visit:

[Mike's Homepage]    [Mike's home-built jet engine]

[Mike's Home-Built Wind Turbine]    [Mike's Home-Built Solar Panels]    [Mike's gold prospecting page]

Copyright 2011 Michael Davis, All rights reserved.