DAN'S "CANNED MOUSE"
By Dan Stixrud
No this isn,t about the kind with fur, you'll have to find another page for that, maybe under survival under extreme circumstances, eh? This page is for info on converting the low resolution optical encoders, typically found in serial mice, into high resolution encoders needed for DSC's (Digital Setting Circles), for your telescope, etc.
The serial mouse encoders I found had the dpi stamped right in the side of the slotted disk1, in my case 200 was stamped in the side. The mouse must be a serial type (like a logitech 3 button mouse), the type that plugs into the serial port. Do not use the common2 mouse that plugs into the mouse port, they use continuous turn pots, not optical encoders.
The serial mice have a serial interface for the optical encoders already and there isn't any reason why you couldn't reuse that portion of the circuit board as well if you needed to interface directly to your computer's serial port. In my case I discarded that portion when I dissected mine with a jewelers saw as I didn't need it.
Here is a neat program, with instructions, compliments of Jonathan Rock, for using a serial mouse for just this function.
The program you use should somewhere ask for the resolution of the encoders and you can calculate an initial setting based on the dpi times the circumference of your bearing surface, if you attach the encoders like I did. Once you have them working, you test it by rotating the axis through a known number of degrees, preferably a full 360 and watch the readout on your program or digital setting circles to make sure it tracks properly. If there is a difference, simply adjust your final resolution to compensate and then retest. In case the dpi isn't stamped in the side simply count the number of slots in the disk and multiply by four, then take a micrometer and measure the diameter of the little drive wheel on the encoder shaft and divide the previous calculation by (pi times the dia in inches) and it should give you the dpi. The result should come very close to a standard dpi.
If I were to describe my mounting system, (you can see it is actually very simple), I would call it an "isolated pressure roller transmission" in order that I can use a delicate pressure on the encoder side and a heavier pressure on the axis bearing of the telescope. This way I end up with a zero slippage, zero backlash encoder. It is possible that there is a bit of periodic error, in fact there has to be some, but with the way I use my telescope I do not notice it. I even have some visable eccentricty in my azimuth bearing and the system still works fine. If you were using the encoders to track with you could factor in any periodic error in an appropriate algorithim anyway, that of course complicates things a little more. The resolution I am getting on my elevation bearing is around 4500 counts as I have a 7 1/4 inch diameter bearing. My azimuth bearing resolution ends up being over 9000 counts.
AXIS BEARING SIZE
The bottom line is that serial mouse encoders can work fine if you are careful with your design and use large axis bearings. If you have have either too small of axis bearings or require direct drive encoders this method isn't going to work too well for you.
EXAMPLE - For a 200dpi mouse encoder you will need at least a 6 1/2 inch diameter bearing to give the recommended 4000 counts of resolution, ie: for a 400 dpi mouse encoder you only need a 3 1/4 inch diameter bearing.
DIGITAL SETTING CIRCLES
By the way the DSC I am using is a homemade one, the Micro-Guider I, plans availible through David Lane of Nova Astronomics (available on-line only).
MOUSE CIRCUIT BOARD
Above are three pictures of a LOGITECH 3 Button mouse Printed Circuit Board very similar to the one I used. The optical chips and slotted wheels are identical to what I used, however the serial interface IC is different. The best way to dissect the board in this case is to first remove that large rectangular IC, don't damage the board at this point. You can clip the leads of the IC if you don't have specialized tools, and then unsolder the pins individually. Remove all other components except, do not remove the the eight optical chips, the housing that carries the encoder shafts and slotted wheels, and the 120 ohm resistors that are in series with the LED Xmtr chips. Carefully mark your cutting lines with a fine tipped felt marker, making sure that you will not be cutting through either of the encoder shafts. A hand jewelers saw does a very nice job.
The LED Xmtr's are almost universally found in a two lead package4 like the above picture. The lead connected to the larger plate in the LED interior is the negative side. Or you can verify the Forward Bias direction with a multimeter set for the diode check mode. Two of these are connected in series with a 120 ohm resistor across the 5 volt supply.
Detectors or Rcvr's are found in, both, 2 & 3 lead packages, the above picture shows the 3 lead package3. In this case that middle lead is the Channel Output! The two outer leads connect directly to the 5 volt supply. The outside lead that is connected to the larger plate inside the chip itself, should be the negative side. You should be able to verify the polarity easily enough by tracing the leads on the Printed Circuit Board.
The above diagram shows how each optical encoder should be wired . After dissecting the Printed Circuit Board you may have to restore some connections that were severed by your jewelers saw in order to make the circuit work. In my case I had to make one connection directly to one of the optical chip leads, be very careful not to over heat the chips with your soldering iron. Please note that the schematic symbol that I used to represent the optical recievers is a guess on my part, as I don't have exact data on the chip itself.
If your detectors / Rcvr's are the two lead kind4, the hook up should be similar to the above diagram.
Feel free to e-mail
me with questions
Clear Skies, Dan Stixrud
Editor's Notes: Press the Back button to return.
|2||A PS/2 mouse or bus mouse.|