Idea for new gadget

[acey]

Make sure you get a patent on it! But what about the appearance of the spot - I'm intrigued to know what it looks like (brightness, sharpness, colour). I can imagine some people happily staring at that instead of tiny faint DSOs...

[themos]

Well, it's not a spot, it's a broad beam ending in a kind of round profile. It's green! As for brightness, it's hard to estimate but perhaps like a bright comet?

[Psychobilly]

Make sure you get a patent on it! ...

Horse would have bolted...seeing as the "idea" has been discussed in the public domian ...

Billy...

[acey]

Well, it's not a spot, it's a broad beam ending in a kind of round profile. It's green! As for brightness, it's hard to estimate but perhaps like a bright comet?

How about a photo?

Only lasers I have are red ones whose beam isn't round, more like a little line. I suppose you'd want a nice round disc when you put this thing on the market - preferably with cross-hairs and circles on it!

[themos]

[jonl]

The two options for motor drive and control are either stepper motors, or, simple DC motors and quadrature encoders. Both would need gearing to get acceptable position resolution (21600 = 1 arcminute, so double this so we know we're within 1 arcminute = 43200 steps for azimuth).

The simplest option would be DC motors and quadrature encoders (just look for an off-the-shelf motor & gearbox with the right gearing and torque, the encoders you can make) - but the slewing rate would need to checked against the ability to count the encoder output (haven't checked, but would imagine that an Arduino should cope with 1KHz counting, so 44 seconds to drive azimuth full-cirlce). Stepper motors don't have this limit as you count each time you tell it to step, but, you'd need to source a suitable gearbox, and check that torque was sufficient to move the mount.

Motor drive circuit. Nothing special here as this would be low power, so a couple of home-brew H-bridges would suffice. Pulse Width Modulation on DC motors would provide for smooth acceleration. Drive rate to the steppers could be altered to provide the same and I doubt, at this gearing, you'd particularly notice the steps.

Mount. Find a suitable bearing for azimuth (you could probably get away with two PTFE rings) and remember that cabling needs to run through the middle (and have a method of ensuring they don't wrap round and round). A yoke design is probably the easiest way to mount elevation (using a nylon bush for the bearings) but would need to be reasonably rigid (don't just bend some thin plate). It will be difficult to make this look 'nice' without going to castings/injection molding.

Tripod. A camera tripod would probably be the easiest and cheapest to source.

Next up is user interface. You could use bluetooth but then you'll need to write an application for a device to control it - probably easier, quicker and no more expensive to add a small LCD and 6 buttons.

And to use it. You'd need to tell the unit current location, ensure it was level, line up on a known star, enter RA & Dec of this star, and then enter RA & Dec of required position. For greater accuracy you'd want to line up on multiple stars (to calculate pitch and roll of the base).

Cost wise:

Arduino £15

DC Motors + Gears £20 * 2

Quadrature Encoders £5 * 2

H-Bridge £10 * 2

LCD £20

Buttons £2 * 6

Camera tripod £25 ?

Laser pointer ??

Mount £50? (buying metal in small quantities is expensive)

Above is rough estimate of buying stuff new from the high street (e.g. Maplins). Depending what you've got laying around and going to trade sources you might be able to get down towards the £100 mark.

However, it will be a fair amount of work and you'll need to know your spherical trig :-)

[rfdesigner]

(haven't checked, but would imagine that an Arduino should cope with 1KHz counting, so 44 seconds to drive azimuth full-cirlce).

The problem is not the counting or timing, it's the spherical geometry.

try typing sin(x) into a microprocessor code compiler.. I've tried it for my homebrew GOTO mount using a PIC 16F690, a somewhat similar device to the atmel device used in the traditional arduino board, and one 32bit sin(X) equation took 70% of the chips memory. You need a fair bit of trig to correct for alignment error. In my drive system I will migrate to a DSP PIC once it's good enough to need to correct for pointing errors/atmostpheric distortion etc. The advantage of the dspPIC is that it can do 16bit multiplies in a couple of cycles, so doing a 32bit sin(x) would take only a small fraction of the memory and it wouldn't take long. The best bonus is that the dspPICs are relatively cheap only ~£10 or so sold singly and the lower efficiency compiler is free.

Hope this helps

Derek

EDIT: looking deeper into Arduino boards it looks like you can get ones with sufficient maths... still best to check it will be up to it.

[themos]

I took apart an old HP printer so now I have 2 dc motors and one stepper. It used optical encoders for positioning the head and the paper. Thanks for your input, jonl anf rfdesigner!

[omneferuss]

The best, nay only thing useful for this concept that I can conceive of (besides your own satisfaction in the DIY challenge) is where a small "orchestra" of Dob users could be "conducted" on a sky tour, following the 'green baton'..... a club/outreach/starparty type of thing, where a single small GoTo unit eg EQ3-GoTo, could serve a number of users simultaneously.

cheers

[themos]

Quite. I've already tried the green laser pointer as a way of communicating to other observers an object to look at.

[jonl]

The Arduino is a very nice environment for a non-expert. The entry price (~£15) is low and you just plug it in to a USB port and start playing. There is plenty of hardware and software support available and all is open source (both hardware designs and software). The programming environment is very simple to understand and get things happening (simple IDE compiles and will upload to the target board). Technically you are writing in C, but there is sufficient pre-processing, wrappers, libraries etc that hides this from a novice.

If I/O, ROM space etc isn't sufficient there are several processors readily available in the range - although the price does jump (~£50 for a board with the largest device).

Would certainly recommend for any little project in need of a simple processor and I/O.