A good idea for a tracking mount....?

[adman]

Having sold my G11 and scopes earlier in the year, I have been toying with the idea of getting a star tracker mount for some time to do some widefield stuff - think astro-trac or lasomandy star-lapse, but have been feeling a bit tight-fisted so decided to have another crack at making a barn door mount (Loosely based on this one (http://www.garyseronik.com/?q=node/52)). In the process of doing this I thought that it would be nice to be able to balance the mount so you could use a smaller motor to drive it.

This got me thinking about variations on the barn door, and this is what I came up with. I am not sure whether it has been done before, so I am flying blind and would like some input as to whether it is worth heading down this path any further.

Basically my idea, initially, was to have a 90 degree arc pivoting around the RA axis, camera on top, balance weights below driven using the same threaded rod arrangement as the barn door at the same radius from the axis etc. A 90 degree arc will give up to 6 hours of picture-taking fun before having to reset it. The pictures show how far I have got building this so far. I have used a threaded rod also (with nuts glued into the pivot point) for the RA axis as it will turn quite smoothly and only needs to go through a quarter turn. It also holds everything in place more simply than smooth rod / bearings, which are a bit overkill for this mount.

Anyway, what became apparent after having got this far was why the barn door has one end of the threaded rod floating free. It it because it is difficult to get the rod into a perfect arc so that where it passes through the drive nut and gear, the rod can move to accomodate errors in the arc. If I want to use this method of driving it, the drive nut/gear and motor/pinion gear will need to be able to float to accomodate errors in the arc. Hmmm...

My next thought was to replace the threaded rod with an arc of plywood of the same radius and attach a flexible rack gear to the outer edge which I could drive directly with a stepper motor. The flexible rack gear I have seen (http://www.servocity.com/html/gear_racks.html) has 18 teeth in a 2" (50.8mm) length - so 1 tooth per 2.82mm If I have a 300mm radius of arc, which needs to be moved at approx 15 degrees per hour, this equates to 78.54mm of arc per hour. The pinion gear to match has 24 teeth, so one rotation would drive the arc 24 x 2.82mm = 67.68mm, giving a required rotation rate of the pinion of 78.54/67.68 = 1.16 per hour. Or I could use an arc length of 258mm to get one rotation required per hour to make things simpler...(can some smart person verify my maths here...?)

This is where I run into trouble. Not being an engineer, there are several things that I need to know, but don't have the smarts:

1. would I need to gear down the stepper motor with a gearbox to get either enough torque or a slow/smooth enough rotation?

2. How much torque would be required given that the setup will be balanced? The arc plus camera gear will weigh 4-5 kgs or so...

3. does anyone have a simple circuit to drive the stepper motor at the required rate. I would need to be able to vary the rate to get it just right. I was thinking of venturing into Arduino territory to accomplish this (I have been meaning to check them out for a while anyway), but if their is a better system - either standalone or PC controlled, I will happily receive suggestions.

Cheers!
Adam

[adman]

one thought I had was if I thread the RA rod through far enough, I can clamp my laser pointer to it and use it with binos to polar align.

Adam

[ZeroID]

One of the problems with using a straight threaded drive rod is that at over 90 degrees the movement will not be linear and constant due to the angle of change of the rod. Some systems have used curved rods and driven the nut so it follows a proper arc and there has been multi platform systems that tend to better emulate the correct movement..
By reducing the angle travelled before a reset is required this is reduced with straight drive rods and makes the need for a variable speed control unecessary. Hence I haven't seen any tracking sytems that will cover 90 degrees accurately.
I found a 2 rpm 240 vac motor at a surplus store I used on my old EQ2 mount. It had plenty of torque due to the low gearing and I have spotted DC versions as well. pays to hunt around.
Keep up the DIY, it's fun just to do.

[2stroke]

You are better off using a double arm, i've even seen a tripple arm which is better again. I am thinking hand guiding using a guide scope with a variable speed controller, use a 2rpm dc geared motor ebay and a pwm controller :)

[bojan]

It will work.. I have seen similar designs even on this forum (Leon's design, for example. Try to search in older threads for details, or simply ask him directly).
On DC motors for tracking, it's not a good idea (it may work for short exposures and wide fields though) - because it is VERY hard to adjust the right tracking speed without feedback (the motor rotational speed depends on variety of factors, like load, supply voltage.. type of lubricant used, temperature and so on).
It is much better to use stepper motor with appropriate circuit (I designed one such circuit with small processor, others did similar things, as well - Alistaism, for example).

You can see some pictures of my design here:
http://www.iceinspace.com.au/forum/attachment_browse.php?a=57884

Good option is (if you have power point available, like in your backyard), to use geared sychro motor as suggested by Brent, if you find one. They are more commonly available on US market, but since the rate of such motor is dependent on line frequency, you have to be aware that in Oz the rate will be ~17% slower (0.8333x exactly - because we use 50Hz, and US uses 60Hz) - which is easily compensated by adequate (smaller) radius of your curved threaded rod). Also you may need a transformer (or resistor in series with motor coil, as Yanks use 110VAC. And, you have to be careful of high voltages in the dark and dew environment.
I am not sure if inverters have line frequency stable enough for tracking.

[Nico13]

This sounds exactly like a drive that a friend made for a fork mounted Newt a lot of years back and he drove it with a synchronous motor and a small circuit to vary the frequency for speed control.
His arc segment was about 90 deg and he went to the trouble of moulding the thread around the arc, the rod was held in a straight line mounted at either end with a guide to hold in the arc thread.
This has been running for thirty years and is still in operation today.

[Nico13]

This sounds exactly like a drive that a friend made for a fork mounted Newt a lot of years back and he drove it with a synchronous motor and a small circuit to vary the frequency for speed control.
His arc segment was about 90 deg and he went to the trouble of moulding the thread around the arc, the rod was held in a straight line mounted at either end with a guide to hold in the arc thread.
This has been running for thirty years and is still in operation today.

[adman]

I think I am going to change the drive mechanism from the curved threaded rod to something like the picture below - a linear friction drive?

[bojan]

Perhaps, have a look at how it was done to move RW head on old, 5.25" floppy drives (belt/track fixed on wheel, no slippage possible).

But, all this is an overkill for wide fields.

[adman]

absolutely - fun though.

I was at bunnings at lunchtime and found a very solid drawer slide which will be perfect for the moving carriage. Will be able to do some more construction tomorrow, and will post update pics.

Adam

[adman]

Hi Brent - hopefully I shouldn't have any issues with tangent error as the drive rod in the picture is curved (see my labelled picture). I have since changed the drive mechanism as it was going to be hard to get a decent curve. I am now going to use the rack and pinion-type friction drive shown in the pictures above. I will post some more pictures of myt progress tomorrow.

Cheers
Adam

[ZeroID]

Aha !! that picture makes it a lot clearer. Quite feasible that way if a little on the large size compared to other builds. Accurate speed control could make that a very effective system. Probably hold quite bit of weight as well if balanced correctly.
Results will be scrutinised and judged according to standard ATM rules ... you know, originality, parts scrounging\conversion % etc etc :thumbsup: so far, so good.

Just love DIY stuff ;) Thats where the fun is ...

[2stroke]

Love that friction drive, gives me an something to play with on a eq platform :)

[adman]

Yep it does look cool, but I don't think it is quite in its finished state yet in those pictures. First of all, the drive nut is not connected to the moving carriage, the threaded rod has no visible means of support in the bearings against axial movement, and the moving carriage itself would be free to move at right angles to the drive movement as there is nothing holding it in place other than the friction if the arc.

[ZeroID]

Details, mere details ...... :P


:lol:

[OzEclipse]

Hi Adam,

In answer to your question about stepper motors, I have used the Jaycar stepper kit in a couple of projects. It comes with a controller card and all components and the motor. I get the kits at trade price but I think the retail is around $50.

http://search.jaycar.com.au/search?w=stepper motor
Cat no = KV 3594

The Jaycar reduction gears are poor quality plastic so I remove the cog from the end and then adapt a precision RS gearbox ($52) to it.
http://australia.rs-online.com/web/p/gearboxes/0336416/

Adapting requires a lathe to machine two simple parts - a spacer and an adaptor. A machine shop could do this if you don't have lathe access.

I mount the controller in a box. The circuit is based around a 555 timer so instead of installing the timing potentiometer, I install wires and take them out to switches that switch different resistors to give me fast slow and normal speeds. Old school but effective. I use it to run a custom lightweight mount I built for eclipse chasing.

GEARBOX
Your calcs are correct however you will need a very high reduction gear box. The rack & pinion drive equates to operate at a very course 27.85:1 reduction. If you direct drive the pinion with a 7.5 deg step motor. each step will be around 900 arc sec or 0.25 degrees. You need at least 1000:1 but 4000:1 would be better.

Motor torque
A 20mNm motor will be plenty if you gear it down > 1000:1
___________________________________ ______________

Barn door type drive
10 years ago I designed a bi-symmetric variant of the Trott type 4 double arm barn door to drive the RA axis of my homemade GEM. It was theoretically capable of driving 4 hours at the +/- 5 arc second error rate providing it could be made accurately enough however it was mechanically complex and going to require a lot more machining than I wanted to do. Around that time I acquired a good quality worm & wheel for much less than what the machining would have cost so I decided that I'd had my fun conceiving and designing it and shelved the idea and went with the worm drive. Good luck with your project!

Joe

[adman]

Hi Joe - thanks for the info - very helpful - and welcome to IceinSpace!

I am struggling a bit with the drive mechanism. As you pointed out the rack and pinion is a bit coarse, but for widefields, probably OK. I have started down the road to a friction drive (in a similar vein to the one above) driven by a threaded rod - essentially rack and pinion, but much lower gearing. I initially thought of using sandpaper as the friction material (my surfaces are plywood...) but I am not sure this will be any better than a firm wood/wood contact. I need some way of adjusting the moving slider up and down to provide a clutch and also to adjust the friction. It is starting to get a little more involved than I wanted, so I am thinking of going back to the drawing board and starting again. I can't shake the feeling that there is a much simpler way...

Those McLennan gearboxes are very nice by the way - I bought some for my G11, but didn't end up using them - I was able to get the PE way down by using their high precision worm blocks.

Cheers
Adam