Slumping a Meniscus Mirror
A meniscus telescope mirror allows the ATM to make a proportionally lighter mirror for a given aperture, due to the fact that the glass, front and back, are shaped to the necessary curvature determine by the desired focal ratio of the telescope. This allows the glass to be uniformly thick (thin!), to support faster mirrors. And the first step in creating a meniscus mirror is to “slump” the glass blank in a computer-controlled kiln.
The process of slumping involves creating a precise form, or mold, to support the glass in the kiln, and then heating the glass through a precise schedule to shape it safely and cool it such that no residual strain is induced in the glass that might affect its performance later. As a point of terminology, I like to make a distinction between a “form” and a “mold” to avoid confusion in describing the process. A “form” is a convex shape over which something is shaped. A “mold” is a concave shape into which something is shaped. I will use these terms precisely to allow you to keep the process straight.
When slumping a mirror, there are two approaches that are available: the glass can be slumped into a mold, or the glass can be slumped over a form. The benefits of the form are that shape of the form is exactly replicated by the face of the mirror that will actually be operating the telescope, whereas using a mold means that the backside of the glass is shaped, and the working surface of the mirror is indirectly shaped as the glass deforms. The benefits of a mold are that it can be more easily made using standard ATM techniques, whereas creating a form is a multi-step process. The downside to using a mold is that the working surface of the mirror is indirectly shaped and, if your kiln is too hot and the glass flows towards the centre, the resulting shape could be significantly different from what you want.
For this project, a form-based approach was used, largely based on the recommendation of my mentors. However after my slumping was complete, I did have a chance to try the mold-based approach for another ATM colleague, and therefore have some experience to relate about that later.
Regardless of form or mold, clearly whatever is used to support the glass during slumping must be able to withstand the temperatures of the kiln. The term for this is “refractory”. The following is the process for creating a refractory form.
A mold for a form
Creating a refractory form involves two major steps: creating a precise mold to support the desired curvature of your mirror, and then casting a refractory form against that mold. The quickest way to creating a precise mold is to grind it using standard ATM techniques in a blank made from plaster-of-paris. This material is easy to obtain and make, and is soft and easy to work. This allows a very precise form to be made in a matter of a couple of hours for usual amateur-sized mirrors of 16″ or less.
To make a plaster blank, you will need a round base of appropriate diameter, and some flexible material to wrap around it to make the sides. If you choose to use cardboard for the sides and/or base, you will want to use plastic wrap to contain the wet plaster, or else it will flow out the bottom. In sizing the blank allow for an extra inch or so of diameter beyond your desired mirror diameter to ensure that when slumped, the glass will be fully supported at the edge. For smaller blanks, you can probably find a baking spring pan to contain the plaster, but by the time you get to 16″ that becomes a real challenge (and expense).
After you pour the plaster, give it time to set. It hardens pretty quickly (and can get quite warm), but I tend to leave it overnight to really cure. Once cured, you can go at it using whatever ATM technique you are familiar with. Your goal will be to hog out a spherical curve of appropriate sagitta for your target mirror, keeping in mind that your blank is oversized, so the sag will be larger. For example, a 14″ f/2.6 mirror will have a sag of 0.337″. Since the radius of curvature must be kept constant in the mold, extending this to 16″ translates to a sag of 0.441″. A spreadsheet of diameters, focal lengths (and radii of curvature) and sags helps to keep this all straight.
Casting a refractory form
Once you have a smooth sphere of appropriate sag, you are ready to cast a refractory form. Your goals at this stage are twofold:
- Accurately reproduce the curve in the plaster mold.
- Use materials that will survive multiple kiln runs.
Through a lot of experimentation, one of my mentors has come up with using a mixture of non-shrinking grout and perlite, both easily obtained from big box hardware stores and garden centres. The basic idea is that the perlite provides the thermal insulation, and the grout is used to bind the perlite into the necessary shape. Since grout is not designed to be heated to slumping temperatures, the perlite is essential, and you need to use the least amount of grout necessary to keep the stuff together. I have also successfully used refractory cement, which is designed for use in fireplace repair and is, therefore, able to withstand the higher temperatures. The function of the perlite at that point is to simply take up volume; if you cast a form from the cement only, it would be very heavy. The perlite allows the whole thing to be dramatically lighter. I have made several forms now and that would be my preferred approach. The image at right is the oatmeal-like consistency you are shooting for.
Refractory slurry
As a guide for quantities, you want to make the form at least 1 inch thick, and the ratio of perlite to binder (grout or cement) should be 2:1. Add water enough for the binder you are using. If you use more binder, it will be heavier, but smoother at the mirror surface. When you have something you are happy with, make a dam around your plaster, put down plastic wrap on the surface, and spoon your mixture in. Carefully press it down, keeping in mind that plaster is pretty fragile and too much pressure will break it. The goal, however, is to get as good a match as possible on the surface that will eventually support the glass. Trust the process, don’t add too much water, however tempting that might be. Then walk away.
Depending on what binder you used, you will have an idea of the rough cure time, after which you can remove your dam. But don’t even be tempted to remove the form from the plaster for at least one day. Then you can carefully separate the form from the plaster and have a look at what you have done. Here are two forms I made, the first using grout, the second using the refractory cement:
Don’t worry (too much) about the pits. These are inevitable because of the perlite. If your mixture was wetter, then you will have fewer pits but, in my experience, I have not been able to do away with them entirely. The reason you don’t have to worry is because just before you put this into the kiln for slumping you will spread a generous coat of ceramic talc (or “Texas talc”) to fill it all in.
Getting ready to slump
Once you have cast your form, you need to wait a long time for it to fully cure. A week or two in a nice warm place, for sure. You want all the moisture to come out of the form. While it is curing and drying, you can check on the curve to make sure that it is a good representation of the plaster, using a cardboard cutout, for example, or a spherometer. Assuming it is correct, just wait. While you are waiting, you should order your glass if you have not done so already. In my experience, it takes a bit of time to get something cut. You also want to find some ceramic talc, readily available at pottery supply shops.
Talking about glass, you have options. You can get pyrex, borofloat, quartz…options galore. Here we are using plate glass, the dumb stuff. Simple and cheap, and totally functional for the mirrors we are making. And, we are using thin plate glass. As discussed earlier, the point of making a meniscus is so that you can use thinner glass for the same equivalent stiffness as a traditional thick mirror blank. So when you get your glass, you want plate glass at 1/2″ or 5/8″ thick. No more. Possibly less. If you are concerned, read this. When you get your glass, inspect it carefully for evenness: no bubbles. That should not be a problem if you use a quality glass shop.
Once you have cured your form sufficiently, you are ready to give it a spin in the kiln. What you are trying to achieve is:
- Drive any remaining moisture out of the form.
- Ensure your binder is strong enough to support the glass.
- Ensure that your form can retain its curve at slumping temperatures.
The last point is really important, because if your mixture is not right, your form can deform. I had an early cast that warped substantially when I used it. Because I did not “proof” the form in the kiln first, I ended up with an f/1.9 mirror, when I was hoping for f/2.8. So doing a kiln run with just the form is a necessary step. The run does not have to be long, but you want to ramp up to slumping temperature, hold it there for an hour, and then turn it off. When the form has cooled down, take it out and inspect it carefully. You’ll note that the form is a lot lighter than when it went in. I found that the extreme heat makes the whole thing feel very “styrofoam-y”, especially around the edges. So handle the form carefully, to avoid breaking it and having to start all over. If the form survived and did not deform, you are ready.
Slumping
Game day. You have your glass, you have your form. Before slumping, find the centre of your form, and your glass. Then spoon some of the ceramic talc onto the surface of the form and spread it around with your hands, filling in the gaps. Use latex gloves, and do it where you can easily clean up afterwards, because the talc is very fine (like baby powder) and gets everywhere. You don’t want mounds of talc, because that would transfer to the glass, but you do want to fill in the depressions left by the perlite and binder. Use your hands to feel the smoothness. If you want to err, do it on the side of not filling a depression entirely because that would translate to a small bump on the glass. A mound of talc would leave a pit on the glass, and that is much harder to grind out.
Once the form is prepped, carefully place it into the kiln, making sure it is not too close to any edge so the glass heats evenly. You can put it on the floor of your kiln. Then place the glass down carefully, centred on the form, and carefully close the lid. Now you need to program your controller. There is a lot of tribal knowledge around appropriate kiln schedules, and feel free to look around, but here is what has worked for me (all temperatures are in degrees C):
- Ramp from room temperature to 482C over 940 minutes (29.7C/hr)
- Ramp from 482C to 650C over 180 minutes (56C/hr)
- Soak at 650C for 240 minutes
- Ramp down from 650C to 482C as fast as possible, 30-45 minutes or better
- Soak at 482C for 120 minutes
- Ramp down from 482C to 315C over 1060 minutes (-9.4C/hr)
- Kiln off and cool naturally, opening lid below 100C.
Some key points about this schedule. First, the slumping soak is at 650C, not higher. If you go higher, the glass will start to flow and you won’t get an even thickness on the meniscus, and your edge will be rounded. Second, some controllers have code for “as fast as possible”, although mine does not. That is why I provide a guideline for timing step 4. And lastly, step 6 cannot be too long. If you have the time and patience, feel free to drag it out longer. The point here is that the soak at 482C, and subsequent cooling, are intended to ensure that there is no strain put into the glass from the process. The slower the glass cools, the better off you’ll be. Oh, and one more thing: do not open the kiln during the run. You may be tempted, but don’t.
And…?
Once the kiln has cooled and you can handle the glass, take it out and inspect your work. Hopefully you had good glass (no bubbles) and it survived the run. The surface may be a bit cobbled if your form was not super smooth. But once it is at room temperature, the most important thing to do is measure you sagitta. If all went well, it should be very, very close to your target. If it is, have a drink. You are ready to move on. Revel in how light your glass is compared to a traditional blank, especially if you have an aperture greater than 12″.
If the glass does not have the correct shape, check your form. Did it deform? Or did you mess up in your sagitta math? The “good” news is that if your glass survived the run, and you merely need to make a new form, you can re-slump your glass. I know, because I did one blank 3 three times, including flipping it over for the last run. And the benefit of using plate glass is that it, too, is relatively cheap.
