Local copy, origional at: http://www.sff.net/people/J.Oltion/trackball.htm
The Trackball Telescope
Designed and built by Jerry Oltion
Featured in the August, 2006 issue of Sky & Telescope Magazine
Click this link to go directly to a discussion of the telescope optics
Click this link to go directly to a discussion of the mount
Click this link to go back to Jerry's home page
An overview of the trackball concept
Kathy and I had been enjoying amateur astronomy for about a year when I decided it was time to try my hand at building a telescope. I wasn't happy with equatorial or Dobsonian mounts--both types have difficulty reaching certain parts of the sky, and equatorial mounts make you do some back-breaking contortions to reach the eyepiece once you do find your target--so I decided to see if I could design something that would be easy to point and easy to look into no matter where in the sky it was aimed.
I didn't want to be influenced by what others had done, so I purposefully didn't look for other designs until I had come up with one on my own. I followed a few false leads, but I eventually figured out that a spherical base resting in a socket would let me point the scope anywhere in the sky with equal ease, and it would let me rotate the eyepiece to a comfortable position no matter where I was looking.
That was half the battle, but one of the things I didn't like about Dobsonian mounts is that you have to keep shoving them by hand to track the stars. How could I make my mount track automatically? The answer to that came in a flash of inspiration: The stars move because the Earth--a sphere--rotates. But I was building a spherical telescope, so if I made it rotate in the opposite direction, it would track. Once I realized that, the solution was obvious: rest the sphere against an axle that points at the celestial pole, and rotate the axle.
Confident that I had just reinvented the wheel--almost literally--I went online to see how other people had done it. Surprise! Nobody had. Spherical telescopes were old news (Isaac Newton even mounted his scope on a ball!), but nobody was talking about my kind of mount. I showed a scale model to other astronomers, figuring there must be some fundamental flaw in the design that would explain why nobody was building them, but nobody I talked to could find anything wrong with the concept. And none of them had ever heard of this design, either.
So I searched the U.S. Patent and Trademark Office database, and came up blank there, too. As near as I can tell, I have come up with a new design for a mount that not only eliminates the pointing problems of equatorial mounts and Dobsonian mounts, but also tracks. I call it the "trackball" because there's no better name to describe it. Happily, another search of the USPTO database shows that the term "trackball" has fallen into the public domain, so apparently anyone is free to use it as a generic term for this type of telescope.
Many people have urged me to patent the design, but I would much rather put it into the public domain so anybody can build (and sell) them as they wish. So I have done just that. In July of 2005 I displayed the scope and mount at a public star party, effectively placing it in the public domain. The article in Sky and Telescope reinforced that action, as does this web site. My actions only affect the parts of the design that are new with me, of course, but that much at least is free for everyone to use, and nobody may patent it.
LATE BREAKING NEWS: After the article came out, I heard from several people who remembered seeing similar designs in the past. None of those designs have been quite the same as mine, but the concepts are similar enough to make it clear that the idea has been around for a while. That's okay with me! As with any star party, the more the merrier. I've put more detailed information on the other designs below.
Speaking of other designs, here's a photo of my second trackball, which I made for my wife, Kathy. It's nearly identical to the first in construction and focal length. The biggest difference is that I beefed up the mount a bit. More on that on the mount page.
So how do you build a trackball? The most important thing is not to be afraid to experiment. My design philosophy has been to make it work first, then worry about making it pretty. I'll describe how I built mine, but if you have a better idea for any of aspect of the design, go for it! Look around your shop for whatever will work rather than following my recipe to the letter. So with that in mind, here's a general idea of how to build a trackball. I've divided the instructions into two pages to make them easier to load and view. Click on the part of the above photo you're interested in, or click on the following links. Below these links I've written some instructions for using the completed telescope and mount, and some other ideas and random thoughts about the design.
The trackball is amazingly easy to set up. Just put the mount on a level surface, aim the drive axle at the celestial pole, set the telescope on the mount, and turn on the motor. This simple alignment is all you'll need for visual observing even at relatively high power.
To find your target, just push the scope where you want it to go. If the eyepiece isn't in a comfortable position, rotate the scope until it is. Most times you can sit down while you're observing, which is by far the most comfortable way to view.
Once you've found your target, just let go and the scope will track. You can center it east-west (right ascension) with the fast or slow tracking buttons, but north-south (declination) adjustments are made by nudging the scope by hand.* Also, if there's any play in the gears (and there probably will be), then you'll want to overshoot to the west and then pull the scope back to the east to load the gears so the scope will begin tracking the moment you let go. You'll quickly learn to combine these adjustments into one smooth, quick motion, so you don't even think about it anymore. You'll just point, let go, and observe.
If your target drifts out of the field to the east or west, adjust the tracking speed. If it drifts to the north or south, nudge the mount sideways or raise or lower the axle a bit to bring it back. That's called drift aligning, and it's a piece of cake. If you've never done it, click here for instructions.
*It turns out there's an easy way to make a declination adjustment. Thanks go to Pierre Lemay (see below) for the idea. He figured out that you can make a declination adjustment by mounting the idler bearings on a pivot that lets you move them toward and away from the drive axle. That will force the ball to roll up or down the axle, which translates to motion in declination. Very neat! I've got a couple other thoughts about this concept on the mount page.
A few other random thoughts
There's not a lot of air flow inside the ball. My scope takes about half an hour to cool on an average night, but then again, my mirror is pretty thin. If yours is thicker, it might take longer, and you might need to install a fan. I've experimented with a 1.5" muffin fan mounted at an angle inside the ball where it's out of the light path and can swirl air down under the mirror and back out the other side of the ball, and that seems to work okay. You might be able to drill holes in the ball and mount a fan to blow air straight out (or in), but if you do, use a small drill so the holes don't mess up your tracking. They have to be small enough that the grid of them will ride across the bearings without bumping.
You might try putting insulation over the counterweight to keep it from radiating heat into the ball, but that might just slow down the cooling process. It depends on how fast its heat will escape through the outer surface of the ball. Experiment with removable foam before you spray permanent stuff in there!
You could cut a big enough hole in the ball for the secondary cage to nest in there during transport, but that would probably cut into your ability to look close to the horizon. I thought collapsibility would be a neat thing when I was designing my scope, but I quickly gave it up when I saw how difficult it would be, and in truth I've never needed to take the telescope apart except to tinker with it anyway. Maybe if it was longer, but at 38 inches it fits into the back seat (or upright in the passenger seat) of my Volkswagen just fine.
I haven't put a handle on mine yet, but the part of the ball directly below the eyepiece never rides on the bearings, so a handle there shouldn't get in the way of anything. On the other hand, it's pretty easy to carry the scope by the lip of the ball, so I may just keep doing that. One other thought on the subject: I'm not sure how strong an acrylic sphere is. My fiberglass sphere is plenty strong enough to support the entire scope's weight from a couple of handle attachment points, but you might want to err on the side of caution with an acrylic sphere.
Ideas to experiment with
(Any of these that are not already patented I also place in the public domain)
- The ball could be made of an open grid of wires or hoops or whatever. You would need larger bearing surfaces--maybe slides--so the gaps wouldn't affect the tracking, but it would really help with cooling (and maybe with weight, too).
- The axle could be a conveyor belt or a sling or anything else that pushes the ball. It could be spring-loaded to press into the ball while tracking and moved away while slewing. (This would allow the use of a very grippy substance for the axle, which could help eliminate fussiness over balance.)
- The mount could be any kind of a cradle, even a toilet-seat-style hole, with the axle pressed against the scope (maybe spring loaded as mentioned above).
- Rotation could be constrained to the correct direction with multiple bearing surfaces oriented in the proper direction, or with a suction cup on a bearing that stuck to the bottom of the ball. (That would probably have to be on a lever so it could be moved away while slewing.)
- The axle could be cone-shaped, so variable speed could be accomplished by moving the axle longitudinally rather than changing the motor speed.
- You could paint a star map on the ball and use a pointer on the mount to help you aim the scope. Of course when you rotate the scope for a comfortable eyepiece position, your map will move...
- You could use the telescope itself for precise latitude alignment. Drift align the scope once, and after you've got it tracking perfectly, put Polaris (or Sigma Octantis if you're in the southern hemisphere) in the center of the view of a medium-power eyepiece. Mark the point where the ball and the axle come together, then next time you set up just sight on Polaris (or Sigma Octantis) and move the axle up and down until the marks line up. You could probably use a similar method for left-right alignment using the idler bearings.
- You could use magnets for fine-tuning the counterweight system. Overdo the weight inside the ball by a pound or so, and put a metal strip up near the focuser. Stick magnets on the metal strip, and remove them when you put a heavy eyepiece in. Or you could embed a metal strip in the ball (easiest if you're making a fiberglass ball) on the side opposite the focuser and stick a few magnets to that when you use a heavy eyepiece.
- Credit for this idea goes to Pierre Lemay (see below). You can make a declination adjustment by mounting the idler bearings on a pivot that lets you move them toward and away from the drive axle. That will force the ball to roll up or down the axle, which translates to motion in declination. Very neat! You could get the same effect by moving the drive axle in and out (being careful not to change your polar alignment in the process--and being careful not to push the ball off the other bearings!)
I put about $500 into my trackball. Some of that was blind alleys (making a fiberglass sphere instead of buying an acrylic one, experimenting with Teflon instead of bearings, etc), so I'm guessing a person could shave $50-100 off that figure if they were careful. If you buy a finished primary mirror rather than make your own, it will probably run you more. Or you could rob the mirrors and other hardware out of your Newtonian and build the whole thing for practically nothing. Once you use a trackball, you'll never go back to that old scope anyway. :-)
Here's my scale model: a Christmas tree ornament with a film can taped to it to represent the top end of the scope. I smooshed modeling clay inside the ball to counterbalance the film can, and set it on a triangle of sticks. When I spun the rubber-hose axle that I placed over one of the sticks, the film can would trace an arc across the sky. Near the pole, it would just spin in place. It worked!
Several people have emailed me since the article came out in Sky & Telescope magazine, telling me about other designs for making sperically mounted scopes track.
The earliest I've heard of so far was built in the early 1970s by Norman James. His scope floated in a pan of water and was driven with an axle attached to the low end of the ball by a suction cup. Here are links to a couple of pictures of that:
The person who emailed me that said there's a publication about this same inventor's ideas. I haven't looked it up yet, but here's the citation:
James, Norman. Fiberglass Applications for the Telescope Maker. pp29-52. Proceedings of the 1970 Annual Convention of the Western Amateur Astronomers. (8 inch Newtonian, spherical mounting)
Then there's a Belgian man named Alphonse Pouplier who motorized an Astroscan and had an article about it in the August 1993 Sky & Telescope. He didn't use a polar-aligned axle, but he did come up with a clever tracking system--and his even has go-to capability. He has posted his article online; the URL is http://users.skynet.be/alphonse/skytel.htm
The third, and so far closest design to mine, was built by Pierre Lemay of Quebec. He tried a single drive axle with two omnidirectional idler bearings and had trouble with the axle slipping against the sphere, so he switched to two drive axles and a single idler. That gave him better control of the ball, and also allowed him to devise a very clever way to make fine adjustments in declination. He mounted his idler bearing on a pivot so it can move toward or away from the drive axles. That forces the ball to roll up or down the axles, which provides declination control.
Pierre also arrived at the same conclusion I did regarding patents. He decided to put his idea in the public domain, so he showed it at the 1995 Stellafane convention, and it was featured in the "Ten Top Telescope Ideas of 1995" article in the January 1996 issue of Sky & Telescope.
Pierre would be happy to discuss his design with anyone who is interested. His email address is to the right. (Sorry you can't click on it or copy and paste it; it's a graphic file to thwart spambots that search the internet for addresses to send junk mail to.)
Drift alignment can seem pretty involved if you get a complicated set of instructions, but it's actually very simple.
In the northern hemisphere, do this:
Aim the telescope at a star to the south, and somewhere near the celestial equator. (That's 90 degrees away from the pole, along a line that runs straight overhead.) Look for north-south motion of the star. We're only concerned with north-south motion. If it drifts east or west, your drive is set too fast or slow. And I'm talking about actual motion, i.e. the direction you have to push the telescope to re-center the star. If the star drifts south, then rotate the entire mount so the axle points farther west. If it drifts north, rotate the mount so the axle points farther east.
Now look at a star near the horizon to the east, and see if it drifts north or south. This time if your target drifts south, then angle your axle farther upward (toward a higher latitude on your scale). If it drifts north, lower the axle.
If you were a long way off in either direction, do the whole procedure again to fine-tune it.
In the southern hemisphere, do this:
Aim the telescope at a star to the north, and somewhere near the celestial equator. (That's 90 degrees away from the pole, along a line that runs straight overhead.) Look for north-south motion of the star. We're only concerned with north-south motion. If it drifts east or west, your drive is set too fast or slow. And I'm talking about actual motion, i.e. the direction you have to push the telescope to re-center the star. If the star drifts south, then rotate the entire mount so the axle points farther east. If it drifts north, rotate the mount so the axle points farther west.
Now look at a star near the horizon to the east, and see if it drifts north or south. This time if your target drifts south, then angle your axle farther downward (toward a lower latitude on your scale). If it drifts north, raise the axle.
If you were a long way off in either direction, do the whole procedure again to fine-tune it.
I've had a lot of help in developing the trackball idea. My wife, Kathy, has been incredibly supportive and has helped brainstorm many a gadget at unlikely hours of the day. David Davis and Mel Bartels, thin mirror gurus and telescope builders extraordinaire, have also been most generous with their time and enthusiasm. Thanks also to Tom Conlin, Ted Touw, Craig Daniels, Chuck Lott (of the Lott 35mm finder fame), Arnie Wittstein, Jim Ruzicka, Dave Cole, Bill Murray, and the entire Eugene Astronomical Society for help and support. Thanks also to Gary Seronik, my editor at Sky & Telescope, who knows how to make a writer feel appreciated. Thanks to you all, I've had a ton of fun on this project.
After the article came out, I heard from many people who remembered seeing similar designs, and from one person who had actually built a similar design over a decade ago. My thanks to all of you who helped flesh out the history of this idea, and my thanks especially to Pierre Lemay, who not only arrived at the same design before I did, but who placed it in the public domain just as I did. Great minds think alike!
I hope anyone else who builds a trackball will get as much enjoyment from it as I have had with mine.
How to contact me
I'd love to hear from people who are interested in the trackball. Please feel free to email me at the address on the right. (Sorry you can't click on it or copy and paste it; it's a graphic file to thwart spambots that search the internet for addresses to send junk mail to.) I have no idea how much mail this idea will generate, so I can't guarantee a response, but I'll do my best to answer everyone who writes with a genuine question or comment about the trackball.