Astro sector drive - tilt sensors - accelerometers - GPS - compass

I am thinking about building a sector drive to track and photograph the night sky - long exposures. The unit will sit on a rotating level platform and align itself to N/S and set the latitude of the right ascension axis automatically. Leveling could also be automatic, but that's not essential.

The plan is to achieve North - South alignment and latitude setting by reference to an electronic compass (GPS?). The compass will operate on a level platform - no tilt issues. Latitude setting, derived from a GPS module, will be from 0 - 90 (North/South is irrelevant). The tracking drive will be static - that is, no electronic control, just relying on predetermined radii of the sector arm and drive pully/gear/friction drive.

I have looked at tilt sensors and accelerometers, GPS modules and compasses. I am a little confused about whether I should use a tilt sensor to measure the latitude setting of the unit - related to the GPS position, or an accelerometer. Furthermore, how accurate are electronic compasses - is a few arc seconds asking too much?

This will be my 5th Arduino project.

is a few arc seconds asking too much?

Yes.
You can't get direction from a GPS if the platform is static.
I think accelerometers are also not accurat enough.
Why not just use an equatorial mount and just one motor?

geoland:
I have looked at tilt sensors and accelerometers, GPS modules and compasses. I am a little confused about whether I should use a tilt sensor to measure the latitude setting of the unit - related to the GPS position, or an accelerometer.

The only purpose I can see for the the accelerometer is for leveling the device, and I don't see what latitude or GPS have to do with that.

All this problem seems to call for is an equatorial plane set to the current latitude and aligned with true North, and a spirit level. If you want it to be completely self-leveling, self-aligning and self adjusting then it would be possible to automate all that by using GPS to obtain latitude, dual axis accelerometer to measure horizontal alignment and compass to measure alignment with North. And then some sort of servo mechanism to adjust the level, alignment and elevation.

Like all the best geek projects, I expect you would spend more time creating an automated setup than you would ever have spent doing it manually. That's not a reason not to do it, of course. :slight_smile:

Why not just use an equatorial mount and just one motor?

To avoid the need for an optical alignment scope and much quicker, theoretically.

Like all the best geek projects, I expect you would spend more time creating an automated setup than you would ever have spent doing it manually. That's not a reason not to do it, of course.

:slight_smile: That's what I'm afraid of and my experience to date... Fortunately this project can be built in stages - functional from the start.

Will an encoder/GPS work in this configuration - a quadrant? I think the '0' end point would have to be calibrated at each set up, after leveling the device. Which might be simpler and more accurate than a graduated scale.

The stepper motor is one of those geared 1/64, 64 step 28Yxx 5v steppers of which there are thousands on ebay.

sectordrivelatitudeencoder.jpg

You should consider using gears to get more precice accuracy and to increase torque.

What is the main optics you are going to use, telescope or something short like camera objectives?

Cheers,
Kari

It will be a camera and lens.

Ok, then you should be fine, wide optics are easy, and if you are not going for more 1-2 minutes, it will work.
Polar aligment plays big role, when that is done carefully, you have easily 5-10 minute exposures.

Cheers,
Kari