DIY telescope control system
December 29, 2013 8:53 PM   Subscribe

How can I learn more about DIY electronic controls for DIY telescopes? Surely there's a reason I'm not finding my particular idea represented online.

I've been reading about making my own Dobsonian telescope and I'd want to build electronic control into it from the beginning. I've searched online and have found lots of resources that are very old covering how to use rotary encoders and a known starting position, but I'm wondering if there's either a simpler way, or a reason why I'm not finding anything else.

It seems like getting a sensitive solid state accelerometer and compass then using those references directly to drive some smooth motors with a bit of hysterisis would be an option that wouldn't rely on getting a starting reference and carefully tracking movements from there. Are these things just not sensitive enough? Too much jitter? Something else?

Are there resources I can use that are written recently to take advantage of cheap new electronic components?
posted by odinsdream to Science & Nature (7 answers total) 2 users marked this as a favorite
One problem with the Dobson alt-azimuth mount is that the two movements that are required to track an stellar target are not isolated from each other. For example, in a coelostat, all you need is a 1 revolution-per-day motor, and a daily adjustment of an angle, and you have a good tracker.

An alt-azimuth gun-carriage mount has two simultaneously varying transcendental functions that interact. This is not conducive to a non-software solution, and software is "hard", so that's probably why you don't see them that often.

Deduced reckoning based on integration of accelerations is prone to more drift than a nice shaft encoder, and shaft encoders are pretty cheap.
posted by the Real Dan at 9:38 PM on December 29, 2013

Oddly, a friend who's into astronomy recently asked me to write some code to test out accelerometers for just this purpose. Turns out that solid state accelerometers stable enough to use for astronomy purposes are expensive and temperature sensitive and need to be carefully measured with a fairly high reject rate.

If you're looking for ways that newfangled electronic components make guiding cheap: Given that imaging is getting way way cheaper, you're best bet at putting together a guiding system these days is tying a wider field spotting scope to your main scope, putting a camera on that (where it can see brighter objects than your main scope), and guiding based on motion in your secondary scope.
posted by straw at 9:49 PM on December 29, 2013

There's no reason to use a complex solution when a simple one is available. In this case the simple solution is optical encoders, which are cheap and reliable. That is the standard way to measure the position of a rotary control.
posted by Chocolate Pickle at 10:15 PM on December 29, 2013

Double-integration errors on your accelerometers will cause it to have quite poor tracking. It will rapidly drift off-target, when actually you need to maintain a lock within a very very tiny tolerance over many minutes at least order to capture good images.
posted by polyglot at 10:25 PM on December 29, 2013

This is not conducive to a non-software solution, and software is "hard", so that's probably why you don't see them that often.

Sorry I didn't specifically mention software. I figured of course a microcontroller would take the sensor inputs and continuously control the two axis movements.
posted by odinsdream at 5:08 AM on December 30, 2013

To answer your resource question; Cloudy Nights is a good online hangout for DIY astronomy.

I'm not sure any magnetic compass is accurate enough to be useful. GPS is a big help in many mounts though. Still need a star fix or two for alignment.
posted by Nelson at 7:53 AM on December 30, 2013

Aren't most fine tracking control systems purposed towards astro-photography? In long exposures, the tube of a Dobsonian would need to rotate around it's axis as well. This can be addressed by an equatorial platform but that needs to be built for a certain latitude and is a pretty huge, bulky thing for a large scope. Perhaps for most long exposure photo work, a small equatorially mounted scope is so much easier to handle that the benefit of a huge aperture isn't worth the trouble. I'm not saying that you shouldn't try to work it out -just a reason why you're not getting many google results.

Then again, film was still king and CCD cameras were water cooled contraptions when I was moving away from the hobby, so rotation might not matter as much with the new cameras/software.
posted by bonobothegreat at 8:17 AM on December 30, 2013

« Older Looking for experiences of job...   |  Looking at trying to twist up ... Newer »

You are not logged in, either login or create an account to post comments