Image of the day

My Account

New to Astromart?

Register an account...

Need Help?

If you have any questions or experiencing an issue please report to


Posted by larry stange   07/21/2004 12:00AM

-Preliminary Investigation-

Using an astronomy program like Skyglobe the normal mouse cursor finds a Star or DSO and identifies it besides giving the RA., DEC., Azimuth & Elevation of that object. This triggered an idea about a year ago to try and modify a basic dob platform to make the dobs actual pointing coincide with the cursor in Skyglobe without any significant expense.

Using a standard mouse on a wood Azimuth wheel marked in degrees (to compare with during the trial development), a mouse was connected to a laptop computer running Skyglobe. A wood degree wheel was constructed as best I could, and made moveable(friction set by hardware tightening & washers), to allow compensation for any compass direction the base platform happens to be generally pointed at. On top of the degree wheel is a second smaller wheel which will allow fine tuning a mouse at any time to correct for pointing errors in the Dobs azimuth from any errors in my degree wheel construction accuracy. It also, is friction tightened enough to not move easily. When rotating the upper dob platform during an Azimuth change, the only thing moving is the upper dob and a pointer for the degree wheel. A standard mouse(unmodified yet) is mounted to ride on top of the fine tuning wheel.

The fine tuning or correction wheel, is simply rotated CW or CCW by fingers. The lower degree wheel is rotated by two hands. Two different levels of friction for the two wheels are so that neither will turn during normal operation. The preliminary test showed the programs cursor follows (roughly) the aiming of the telescope in azimuth and was fairly repeatable, but dob slop in azimuth movement introduced errors.


In order for something to work like this, all play or looseness must be eliminated in the dobs azimuth mounting. A key item here is the central pivot bolt. The central bolt for the dob was made rigid, going thru steel sleeves imbedded into the upper and lower dob platforms, and honed to fit the central bolt snug. Final accuracy of dob is dependant on the continued rigidity of this bolt and the accuracy of the reference degree wheel.. An aluminum plate was fashioned to hold the recessed hex head bolt from turning on the underside of the dob base.

Depending on the bearing type you are using, ( I used a 12" Lazy Susan bearing, and also Teflon on another dob), you may need to file or grind precision skids of Teflon or cutting board plastic at the corners of the dob to eliminate any "rocking", this improves stability in the system.

-Initial Performance Check & Setting-

The dob is pointed East (for example), and the computer program is started. Skyglobe automatically sets the mouse cursor centered in Azimuth, mid screen. (The program allows saving in any direction). The program has an Azimuth scale in degrees along with 15 degree markings in Elevation. The mouse cursor also can be read out in RA., DEC., Azi., & Elev., for its current position in the lower left corner of Skyglobe. In the program set the HORIZON to match the starting position of the screen cursor in screen center and "save" it to make the initial adjustments easier when you want to work on the dob.

My purpose at this point is to see how far off I am when I move the dob 10 degrees in either direction(doesn't matter), in Azimuth with a mounted "basic" mouse running on the upper disk , using the degree wheel pointer for the 10 degree shift, and noting how far the Skyglobe cursor moves for that 10 degee change horizontally in Azimuth.

The standard (unmodified mouse) is turning an internal wheel with many (50 or so) spokes across a light beam sensor when it is in motion. Different brands and models of mice have different numbers of spokes in their wheels. In my case, with a generic mouse, the Skyglobe cursor moved more than 30 degrees in Azimuth. This means too many spokes (too many tic's) by three times. Dis-assembling the mouse was required to fill in with black paint (with needle) some of the spokes. This must be done evenly around the spoked wheel. In my case every 2nd AND 3rd spoke was filled in with black paint. A second test of a 10 degree change on the degree wheel when rotating the dob shows it is fairly accurate now, and continues to be going up to 45 degrees either side of computer screen center for a screen width travel of 90 degrees.

Adjustment of dobs angular degree accuracy (the degree wheel) to the computer screen cursor accuracy in azimuth is done by "where" the mouse ball rides on the upper wooden circular platform in respect to the axis. If the mouse is to far out from the center, it will cause too much deflection in the program, if too close to the center it will not deflect enough. This is a very fine adjustment and must be done carefully while re-checking both degree wheel pointing and computer cursor displacement for that change. It is at THIS point I used the zoom feature of the program to obtain the observation window I wanted, both in azimuth and in elevation, and saved the program to that configuration. All of the adjustments for ths dob are for ONE COMPUTER ONLY!. If you change computers, screen size and resolution will require re-doing the entire calibration over for the new computer. Final adjustment of the azimuth mouse is done at 45 degrees of elevation in the program, makng it most accurate when the screen centered cursor is midway up the sky at that particular zoom. This final positioning of the azimuth mouse is done with a pre-slotted plate or holes (slotted preferred). Also prior to this, a nut with bolt and washers is exactly centered in the mouse shell top by drilling, with enough bolt threads poking thru to allow mtg. to the adjusting plate surface after the calibration of the mouse position is finished.

When finished and all mouse mtg. hardware is tightened, you COULD stop here!. It was handy to find DSO's after setting the azimuth to line up in Skyglobe on a DSO and then just moving the scope up in Elevation with a common degree dial/pointer assy. This only requires a very small back 'n forth motion to pinpoint the DSO at the correct elevation and my error was less than a degree in most cases. It is easy to find DSO's this way with a wide angle low power eyepiece.

-Second Generation Model-

However, after using it this way for about a year, I decided to approach the Elevation aiming of the dob using the same basic idea. For this I started on a second (and better) 13" dob base from scratch. I wanted to use the dobs upper bearing surface for a mouse pad to avoid making disks. This will require using a TrackBall mouse, and making a spring loaded plate with the guts of a trackball mouse installed on it, or a whole mouse mounted on the plate. I used the guts of a mouse. (Photo 5)

With a repeat of the original azimuth idea redone on the new dob platform.

Having a limited source of mice on hand, (I did not have two identical trackballs), I chose to use the guts of an old burned-up Tandy trackball mouse for the Elevation. It would have been simpler to use a whole mouse but the old Tandy was too ugly!.

Painting of the spokes was not necessary because the Tandy mouse had far fewer spokes. A normal mouse for Elevation will require filling the spokes to calibrate as explained before using the 15 degree marks in Skyglobe and a machinist's level for 0 degrees, 45 degrees, and 90 degrees, at the current and final zoom mode that was saved. The spoke filling procedure will be the same for most mice types. Final calibration is the same, sliding the mouse mounting plate up or down on the dob platform to different points of contact on the bearing radius. Note: - Optical mice probably cannot be modified this way for tic calibration.

Now you have two mice and two connectors..... You must join the two into one connector. I chose to cut the traces on the Azimuth mouse vertical circuit and join the Elevation mouse's vertical traces to it with wire (or 5 wire cable), after cutting the traces free from the original Tandy circuit. Different mice have different circuits too, so I cannot describe how that was done specifically. But it can be done by any technician after he has made certain which traces are to be cut on both mice. Two identical mice are a lot simpler. Mine were not identical. At any point, you can adjust the Elevation mouse simply by moving the ball with fingertips under the spring loaded plate if you need to gain greater accuracy in really high Elevation settings. (Photo 6 )

Skyglobe can present its astronomy program in 90 degree wide azimuth quadrants. You can vary the programs view direction with keyboard pointers but YOU MUST bring the scope back to horizontal either with a bubble level or a marker of some kind showing telescope level in any NEW direction of N., E., S., or W., (or anything else) because remember... Skyglobe always brings the cursor mid screen on program startup and you will have to reset your Skyglobe vertical cursor with fingers turning the vertical mouse ball after leveling the telescope so that it lines up with the horizon shown in the program in order to keep vertical accuracy. This is in effect recalibrating the vertical for the new azimuth change. Of course when changing the compass pointing of the telecope, you will need to also change the programs compass direction to match. All of this sounds hard and complicated.... It's NOT!. it only takes seconds!.

This compass direction change can be done by restarting the program after setting scope horizontal in the new quadrant direction and using the arrow keys or keyboard keys (N,S,E, or W) to center the program at this new direction. I prefer using one of the 4 compass points myself. The azimuth degree wheel will not need changing unless it was not previously accurately set. The azimuth mouse wheel is then adjusted to match the new change in degree on the computer screen if the dob angle is not dead on. The adjustable degree wheel pointer comes into play here for large direction/quadrant changes in Azimuth if it's already has been set at a compass point earlier and you have not moved the dob base. NOTE: -Anytime you run the cursor to the edge of the programs screen, it will require resetting the scope because the horizontal and vertical accuracy is altered when the mouse stops moving from the program limit, but the dob continues to move beyond the screen limit. Avoid doing that.

There are shortcuts available.... Program familiarity allows many shortcuts that save time and effort. (Photo screen)

I will apologize for what is difficult to describe, but is intuitively OBVIOUS after you have done the first Azimuth wheel calibration. The rest of the calibration in the vertical is only a repeat of the method used in the horizontal mouse. It is drop-dead simple in reality except for the circuit merging. More advanced modern computers allow the use of two mice simultaneous, this means you only need to wrap black tape around ONE SENSOR IN EACH MOUSE to disable that part which you do not need, be it the horizontal or vertical wheel.

BEST OF ALL -AT ANY TIME, you can aim at a known star, adjust the programs direction, adjust the mouse lower wheel & upper ball to coincide on that star on the programs screen and you are closely calibrated for that whole 90 degree quadrant of DSO's shown on the computer screen. You need not fool with the degree wheel or anything else.

Realizing this is a working outline and subject to improvement in every way, it is offered only to open up new possibilities in Dobs, Portaballs & Guidance Systems. It is hoped others will continue to develope this concept and hopefully better integrate it with more popular astronomy programs.
Crystal sky to all.

-Larry Stange, Yuba City, California