Can you sell me a complete kit? An assembled
unit?
Where do I get the parts?
What encoders should I use?
Can I use encoders from a mouse?
How do I attach the encoders to my telescope?
How do I figure out my encoder resolution?
What kind of serial cable do I need?
My computer doesn't have a serial port. How can I
connect the interface?
What software works with your interface?
Does your interface emulate other types of
interfaces?
Can I use your interface with a PDA instead of a
PC?
How can I bench-test the interface? Must I have the
encoders connected?
Why do both the RA and declination change in my
software even if I only turn one encoder?
I've built the interface but it doesn't work. What
do I do now?
Can you add an LCD display to your interface and
eliminate the need for a PC?
Nope. You gotta collect the parts and build it for yourself. Relax--it's not hard, and you might learn something. If you get stuck, send me an email and I'll do my best to help.
FAR Circuits sells a printed circuit board and a programmed PIC chip for this project. Go here for ordering details. The rest of the parts can be purchased from nearly any electronics supply house. I use Jameco and Mouser. A parts list with part numbers for Jameco and Mouser is on my how-to-build page.
I recommend the S1 or S2 encoders from U.S. Digital. You don't need any of the options like index, ball bearings, etc. Encoders are rated in counts per revolution (CPR), but because we use both channels in quadrature mode, the actual resolution (what I call tics/rev) is four times the CPR. So, encoders rated at 1024 CPR will give us 4096 tics/rev, or about 0.09 degrees per tic (assuming a gear ratio of 1:1 when attached to the telescope). I recommend choosing encoders that give somewhere between 4000 and 10,000 tics per revolution of the telescope on its axis, taking into account any gear ratio due to the way you attached the encoders to your telescope. The maximum allowed tics/rev is 65535, but 4000 to 10,000 tics/rev is good enough.
Some people have done that successfully, but I can't tell you how to do it--you'll have to figure it out for yourself. The interface is expecting each encoder to have two channels that output 0 V and 5 V as the encoder is turned. Mouse encoders are fairly low resolution, too, so you'll need some gearing.
If you have a dob, I recommend attaching the encoders directly to the two telescope axes. For other types of mounts, you're on your own. There are too many different types of mounts for me to be able to tell you how to do it. Just remember two things: the encoders must turn whenever the telescope moves, regardless of whether it's being motor driven or pushed around by hand, and there must not be any encoder slippage or backlash.
Multiply the encoder's CPR rating by four, and then by the gear ratio (if any) by which it's attached to the telescope.
You need a straight-through cable, not a null-modem cable (unless you're using a PDA instead of a PC--see below for details).
You'll need a USB-to-serial converter. I've been told that not all of them work very well, so you might want to do a little research. If you find one that works, let me know about it and I'll add it to the FAQ. I'm researching the possibility of creating a USB version of this project.
The packages listed on my software page are known to work.
Yes, it emulates the various Tangent Instruments interfaces (BBox, NGC-MAX, etc.). It also emulates David Lane's Microguider III.
Yes, if you have a serial cable for your PDA. A USB cable will not work, even with a USB-to-serial adapter. You'll also need a null-modem adapter in order to be able to plug in the PDA's serial cable. See my project description page for details and a special note about Handspring Visors.
You can bench-test the interface without the encoders connected to verify that the communication with the PC is working. See the Testing section of my how-to-build page for details.
This happens because your telescope is not perfectly polar-aligned. Don't worry about it. As long as you're finding objects, everything's working fine. This often throws people when they are bench testing using my Windows software. It's natural to assume that only one number should change when only one encoder is turned, but there's a lot of math going on to convert the encoder counts to celestial coordinates.
Follow the procedures listed in the Testing section of my how-to-build page.
No. I get asked this a lot. The PIC chip isn't powerful enough to handle the math needed for something like this. I'd need to replace it with a full-fledged microprocessor, and I don't know how to do that. Sorry.
Last update: 8 Aug 2004