The Trackball Telescope Optics


The trackball telescope itself is basically a Newtonian reflector with the mirror inside a ball instead of a tube or a box. Trusses hold the secondary the correct distance away from the primary mirror, and a counterweight below the mirror balances the whole works so it will stay put wherever you aim it. The beauty of this design is that you can aim it anywhere simply by pushing it there (you're not locked into altitude and azimuth or right ascension and declination motions), and you can rotate it so the eyepiece is in a comfortable position no matter where in the sky you're pointing. Even if you don't build the tracking mount, the spherical base in a simple cradle is much more fun to use than an alt-az or an equatorial mount.


First, decide how big you want it to be

A spherical scope has to have its balance point in the middle of the sphere. That means the longer the focal length, the more counterweight you have to put in the bottom to balance the weight of the secondary and focuser and finder. F/6 or so is probably the practical limit. My scope is f/3.8, which is probably a bit shorter than it needs to be, but that's where my mirror wound up when I was done grinding and polishing it. (No, I didn't start with a salad bowl, but I followed what can only be described as a steep learning curve.)

There's probably a practical limit to the mirror diameter, too. Mine is 10" and the finished scope weighs 35 pounds. It's easy enough to carry without taking the secondary cage off, but I'd bet a 12" or larger would only be movable in pieces. That's not necessarily a big deal, since the trusses are simple to remove, but I do enjoy being able to carry the thing outside already assembled and start observing within a couple of minutes. Decide what's most important to you: aperture or portability, and plan accordingly.


Jerry applying fiberglassStart with a sphere

I built my first one by coating a big rubber ball with fiberglass. That was a messy job, but it made a sphere you could drive over without hurting. (I made it 1/4" thick.) I tried making another one out of paper mache and coating it with epoxy to stiffen it, but getting paper mache thick enough was as much work as fiberglass and the result wasn't as strong. (I wound up putting fiberglass over it). If you don't want to mess with fiberglass, you could go to,,,,,, Rainbow Net & Rigging (harbor buoys) or any of a dozen other places like them and buy a ready-made sphere. Several of the places will even cut the opening to your specifications. The downside is that it's hard to find a ball that's stiff enough, especially in the larger sizes. Check them out in person before you buy if at all possible.

IMPORTANT INFORMATION ABOUT MATERIALS: It's becoming apparent from several people's accounts that acrylic spheres are too flexible. Polycarbonate spheres work better, but are more expensive. I've also heard good reports from people who have used harbor buoys (or net floats). Those have a seam that needs to be filled (or sometimes it's raised, so it can be sanded down), but a little body filler will do the job. Rainbow Net & Rigging has a 16" mussel float that's good and stiff and it's only $17 Canadian (as of early 2009).

If you need to sand your sphere smooth, here's an easy way: Tape strips of sandpaper to the rim of a bucket, then rub the bucket rim around at random over the ball. The sandpaper will knock off the high spots and the circular bucket rim will keep you from digging flat spots, so you'll wind up with a near-perfect sphere.

Other materials: I've experimented with paper mache, which is surprisingly strong when you make it 1/4" thick or so, but I needed a stiffening ring inside the ball to make the opening rigid. Plus it dents where it rests on the bearings unless you soak it with wood hardener and/or coat it with epoxy or something similar. I couldn't get it durable enough without covering it in epoxy, which is halfway to fiberglassing the thing, so I wound up putting a layer of fiberglass over the paper mache.

I also experimented with fiberglass-infused body filler so I wouldn't have to deal with wet strips of fiberglass mat, but body filler comes in two types: short-strand and long-strand, and both have problems. The long-strand is pretty hard to work with because it clumps up, and the short strand isn't strong enough. It would make a good filler on top of long strand or on top of mat-and-resin, but it's too brittle by itself. Medium-strand would be perfect if there is such a thing, but I haven't found any yet. After three scopes, the best method I've come up with is to do a few layers of regular fiberglass-and-resin mat, then sand the worst of the bumps off and use short-strand body filler to smooth it out. I plan to try a polycarbonate globe for my next scope, and if I need to reinforce it I'll put the fiberglass on the inside so I don't have to sand it smooth.

How big a sphere do you need? As mentioned above, the scope's center of gravity has to be in the center of the sphere, which means the mirror has to rest below the center to help counter the weight of the secondary and the focuser and the finder. 1.5 times the mirror diameter will let you place the mirror pretty low in the sphere and still leave room for more counterweight below if you need it.

How big should the opening be? Make it about 1" bigger than your mirror diameter. That will give you half an inch of clearance all the way around for the light path. You could cut it bigger, but if you cut away too much of the sphere you won't have enough bearing surface to let you look at objects close to the horizon.

An amateur telescope maker named Mike Connelley pointed out that you can use a hemisphere if you tilt it a little. He writes, "Consider a telescope looking at the horizon. In the regular hemisphere mounted case, the face of the hemisphere is vertical. I would tilt it about 30 degrees, so that the bottom of the hemisphere is tilted towards the front of the telescope just like the D-shaped altitude bearings of a regular Dobsonian. If you do this, you can point the scope from the horizon to zenith, and all the way around in azimuth, with room to spare for tracking or rotating the tube to get a more convenient eyepiece position. The hemispherical bowl would likely require reinforcement to prevent it from flexing too much, but it ought to work." He's absolutely right!

My fiberglass spheres were strong enough to not need any stiffening material around the opening, but a factory-made sphere (or a hemisphere) might be thinner and/or more flexible. If it is, you'll want to attach a ring around the top to reinforce it.


Cutaway diagramThe layout inside the sphere

Everything inside the sphere is glued, rather than bolted into place. That leaves the outer surface smooth to ride on the bearings and the axle without bumping over screw heads. If you don't trust glue, go ahead and screw, but recess the heads and smooth them over as well as you can. Any deviation that rides over the axle or the roller bearings will mess up your tracking.

The counterweight goes in the very bottom, and the mirror sits above it. Because you can't collimate the mirror from behind the way you do in a regular Newtonian scope, you hang the mirror from three supports and collimate it from above. It's easier to see in a cutaway drawing than in a photo. I show the mirror mounts and the truss mounts on opposite sides of the ball for clarity, but they're really 120 degrees apart.

My original counterweights were two disk-shaped weights off a set of barbells, mounted on a short length of pipe that I glued to the bottom of the ball. That worked fine until one day when I tilted the scope toward the horizon and the pipe broke loose. Now I use lead shot that I mixed with epoxy. It's not going anywhere!

The shot doesn't look centered in the drawing. That's because it's not. The focuser and the finder will offset the telescope's center of gravity to the side, so you should put the counterweight a little to the other side to compensate for that. How much? You could calculate out all the moments of inertia of the various parts of the system, or you could just assemble the scope, put an average-weight eyepiece in it, and balance the scope upright on a hard surface. You will have to tilt the focuser and finder inward a little to get it to balance. The point on the ball that's touching the floor when it balances is where the center of your counterweight should go.

You probably want to go a few pounds overweight inside the ball and compensate with removeable weights on top. That way if you add a heavy eyepiece, you can just remove one of the top weights. Much easier than adding more weight inside the ball! (See below for the clamp weight design.)

Don't paint the inside of the sphere until everything is glued in! The paint bond won't be strong enough to hold. As for what kind of glue to use, I used epoxy and Shoe Goo on my scope. I don't know what the best glue for acrylic or polycarbonate or other materials would be; you'll have to experiment. If people let me know what they learn about various materials, I'll post it here. I do know that Shoe Goo only cures on contact with air, so it's not good for bonding two wide surfaces together. It will only cure around the edges.


Mirror cellMaking and mounting the mirror cell

I used a triangular piece of plywood for a mirror cell, with felt pads to support the mirror and three clips to hold it in place. The mirror hangs out over the side of the triangle and could be hit by the counterweights if they ever broke loose, so I'm thinking about making a new cell that's as wide as the mirror all the way around. Even if you do that, you'll still need three "ears" on it for the collimation screws to attach to.

Aficionados of "PLOP," the fabulous program for calculating mirror support points, will shudder when they see this support pattern. This was supposed to be temporary, and I was going to run PLOP to find out where the support points should really go, but when I took the scope outside for the first time I got such good images with it that I didn't want to mess with it.

I think that was blind luck. You should probably run PLOP to see where to put your supports. Here's the link:


Mirror clampHere's a closeup of the mirror clamp. The top part reaches over the mirror by 1/8" or so, and the bottom part rests against the mirror's edge to keep it from sliding sideways when you tilt the scope toward the horizon. To make it I bent a piece of galvanized metal strap into the right shape, drilled it so the deck screw would fit through without binding, and padded the face that touches the mirror with a piece of rubber (glued on). The screw goes straight into the wood, which provides enough friction to let me snug the clamp down without fear of it coming loose. If the screw hole does loosen up over time, a little wood glue should snug it right back up again.

You could embed a nut in the wood and use a bolt instead of a screw, or figure out a completely different mirror clamping system. If your mirror is much thicker than mine, you probably should, because a screw in 3/4" of wood isn't going to hold a heavy mirror if you jounce it hard. The important bit is to make sure the mirror doesn't go anywhere, but doesn't get pinched by the clamps. When you tighten them, just snug the rubber down to the mirror. Don't compress it.

Note the washer at the bottom of the photo. That's where the spring for the collimation screw rests. (See below.) On the underside of the wood I've embedded a nut for the collimation screw to thread into.

Frying pan mountMy second mirror was a lot heavier than my first, so I made a beefier cell for it out of a frying pan. I used an aluminum pan with a bottom the same diameter as my mirror (10"). I cut away the sides, leaving six tabs sticking up, then carefully bent them so three of them would hold the mirror in place and the other three would act as hangers.

You only get two or three bends before the aluminum cracks, so measure twice, bend once. (Pounding them completely flat, then clamping them in a vise and bending them up and outward works fine.) Leave a little gap for expansion. I left a fairly good size gap and put some closed-cell foam in the crack to hold the mirror snug without putting enough pressure on it to distort its shape. (The uncoated mirror you see here is 2" thick, so I could probably have wedged it in with sticks without distortion [just kidding!]).

I cut a big hole in the middle, thinking it might help the mirror cool faster. Since aluminum is such a good conductor of heat, I might have been better off leaving it solid. Who knows?

Mirror mountHere's my first mirror mount inside the ball. I shaped the back edge of the wood to match the curve of the ball and glued it in place. There's room for a triangular wedge below it, also glued in, for extra strength. This (and the other two like it) are all that holds your mirror in place, so make sure it's solid.

The hole through which the collimation screw goes should be snug, so the screw has no sideways slop. This will keep your mirror from shifting from side to side when you tilt the scope from horizon to horizon. On the other hand, the hole shouldn't be so snug that the threads bind in it. The actual collimation adjustment takes place when the screw pulls upward on a nut embedded in the bottom of the mirror cell, so you don't want it riding up and down on the upper mount as well.

You could make it work that way if you would rather, and it might even hold the mirror more firmly in position if you did. But one end or the other has to be free to slip, or you won't actually have a collimation adjustment.

Put washers on both ends of the spring and under the knob to prevent any metal parts from rubbing on wood. Glue the washers in place. It's almost impossible to assemble everything if all the parts are loose!

Paint all the wood and hardware flat black to cut down on internal reflections. Glue first before painting! Note the nifty anti-reflective texture on the ball's inner surface. It's tempting to recount how many hours of painstaking work it took me to etch that in there, but in fact it's just the texture of the rubber ball that I started with, faithfully reproduced when I coated it with fiberglass .


Collimation screwsThe collimation screws are just all-thread (I used 10 x 24 on the first scope, 1/4 x 20 on the later ones) with three nuts tightened together on one end to make a convenient knob. Put a piece of rubber hose over the nuts to improve your grip.

The springs should be relatively stiff. 5 pounds only compresses these guys about 1/4".

Make the screws long enough to engage the nuts in the mirror cell with the springs all the way extended. That will allow you to put the mirror cell in place without compressing the springs. But make sure the screws are short enough to miss the inner surface of the ball.


Truss clamp from sideThe trusses

The bottom of the trusses clamp to the outside of the ball. Truss clamp from aboveIt may seem counterintuitive, but the stationary part of the clamp goes to the outside; otherwise you won't have enough wood for the mounting screws to bite into. (If you wind up putting a reinforcing ring around the opening, you might be able to do it the other way.) It doesn't look like there's much form-fitting in the clamps, but I did actually dig grooves in the blocks for the trusses to fit into. I also put tiny nails in the stationary side to pin the trusses in place so if I lift the scope by the secondary, the ball won't fall off.

Alas, the trusses aren't interchangeable. I tried, but when you're hand-building things, there are too many variables. Do your best, but if you have to number them, don't feel bad. So did I.

The trusses should probably be painted black to cut down on glare from reflected light. I plan to do that real soon now. :-)


True Value bucketThe secondary cage

Your secondary cage can be just about anything, down to something as minimal as an embroidery hoop with a black card sticking out of it for a backdrop. The only real requirement is that it be light, because every ounce up there is going to require 5-10 times as much weight inside the ball to balance it. I used the top of a 5-gallon bucket. The fins around the top make it surprisingly stiff, and the lip where the handle went makes a great place to mount the trusses. The bucket tapers, so you'll have to shim the focuser a bit to make it aim directly across, but that's easy.

I used 1/2" oak dowels for trusses. The top of the trusses can be filed flat to fit inside the bucket rim and attached with bolts and wingnuts for easy assembly. Note that they don't meet on top. When I tried that, they forced the bottom of the bucket into a cloverleaf shape. A longer focal length would probably avoid that problem, but there's no reason why the trusses have to come together at the top (or at the bottom, either, for that matter). As long as they've got some angle to them, the scope will be rigid.

While you're deciding where they'll go, look down from the top and make sure they don't cut through your light path. If they're angled too steeply, they will.

The plastic bucket will probably take paint without scuffing it first, but I hit mine with some fine grit sandpaper just to make sure. Use several very light coats of spray paint rather than one heavy one. When the inevitable scrapes and bangs expose white plastic, I touch it up with a black Sharpie.

I learned too late about a special paint for plastic made by Krylon. It's their "Fusion" brand paint, and it sticks to plastic way better than regular paint. I also learned too late that they make black buckets, too. D'oh! Of course they're too shiny to use as-is, but you can paint them flat black just as easily as you can a white bucket (easier, actually) and they don't show white when you scratch them.


Spider and focuserThe spider and focuser

I bought my spider from Murnaghan Instruments. This is their 11.5-12.5" model, and it fit in the bucket without any modification whatsoever. It's lightweight and easy to collimate. People who are used to 3 adjustment screws may wonder if four will work as well, but I've used both kinds for a while now and I actually prefer 4. You can adjust one axis without budging the other, and it's much more intuitive which axis you're adjusting!

The heavier the spider, focuser and finder, the more counterweight you're going to need down inside the ball. I decided it was worth the trouble to use a good focuser, but that did lead to some balance trouble. With my original counterweights centered in the ball, the weight of the focuser and finders would pull the scope to the side, so I had to add more weight to the secondary cage on the side opposite the focuser, which meant I had to add more weight inside the sphere, and so on. I was actually glad when my first (inadequately glued) counterweights inside the sphere broke loose, because that meant I could put the replacement weight off to the side a bit so the scope would balance better. Now I don't need extra counterweights on top at all.

Scopestuff sells a great two-speed Crayford focuser for half the price of most two-speeds. (It's not pictured here -- these photos were taken before I bought it).


FindersThe finders

I use two finders. The main one is a really lightweight finder made by a friend of mine named Chuck Lott. Chuck is a genius with finders. This one is made from a 35mm film can, a lens, and some clear plastic sheeting, and it weighs all of 3 ounces. If you're interested in the design, follow this link to a separate page about it.

3 ounces is such an insignificant weight that I went ahead and mounted my green laser pointer beside it for a "lazy man's finder." Turns out the laser is great for rough pointing, but the 35mm finder is much more precise, so I may remove the laser pointer.

The stars are stickers that Kathy and I put on whenever we both see something cool through the scope.


Dust coverThe dust cover

I use a plastic bag over the secondary, secured with a rubber band taped to the top of the bag. I'm sure there are more elegant ways to do this, but as the old saying goes, "Temporary arrangements tend to become permanent."

I used a piece of masonite for a primary cover. I cut notches in it so the truss mounts would help hold it in place, and I made it stick out over the rim of the ball a ways for two reasons: it gave me a place to put some half-round closed-cell foam insulation to seal it tight against the ball, and it made the disk wide enough that there's no way it will fit through the hole and bang into the mirror if I accidentally drop it. It fits tight enough that the bolt heads on the truss clamps hold it in place.

One problem with masonite: it has to be sealed or it's dusty, but it sucks up varnish like a sponge. I must have used half a pint on that tiny disk alone, and then it outgassed so bad I was afraid to get it near the mirror until I'd let it cure for a couple of weeks in the sun. On my subsequent trackballs I used a 1/4"-thick sheet of black plastic. Much better!


Paint and wax

I used automotive touch-up paint for the outer surface of the ball, figuring it would be more durable than standard spray paint. I don't know if it is; but it does seem to hold up pretty well. I tried Turtle wax to let it glide over the PVC bearings better than the paint alone did, but it creaked and groaned and had some "stiction" when I'd try to nudge the scope. I tried parafin, but that was too slippery and made balance supercritical, so I tried ski wax. The hard blue stuff for really cold temperatures works great! I had to balance the scope much more carefully than when I used Turtle Wax alone, but not so much as I did when I tried parafin. Now the scope moves smoothly even when you want to just a little nudge, and it still tracks like a dream.

Clamp WeightTo make balancing it simple and painless, I attached lead weights to spring clips that I can clip on the trusses wherever they're needed, so when I change eyepieces I can just clip on or unclip one of the weights.

They're easy to make: take a 4-oz. spherical weight and pound it into a rectangle that's narrow enough to fit between the grips when the spring is open wide enough to clamp onto a truss. Drill holes in the clamp and screw the weight onto it. Cover the screw heads with black vinyl tape, and wrap tape around the clamp so it won't scratch your trusses. Make three or four of these, and you can trim your scope to just about any configuration of eyepieces, finders, etc. you want. (This assumes your counterweight in the base is heavy enough to compensate for extra weight up top.) If you're worried about lead poisoning, cover the lead with paint.


That's pretty much the scope. You can come up with all sorts of variations that would still work on a trackball mount. The key is to make something that has its center of mass (and thus its center of balance) in the middle of the sphere, so there's no tendency for it to rise or fall or twist no matter what position it's in. A little bit of imbalance is okay, but if it's too much the scope will slip against the bearings and won't track properly.

Click here to go to the discussion of the mount itself.

Click here to go back to the trackball front page.

Click here to go back to Jerry's home page.