Ideas for a Home Made Mount for my WF Triple Imaging Rig

[Gina]

In rethinking the mini observatory for my widefield triple imaging rig, the size of the NEQ6 mount in my current design looks very much OTT compared with the imaging rig.  Furthermore, it is the biggest part by far.  A smaller mount would mean a much reduced mini observatory size, particularly the opening roof.

A couple of years ago I was seriously considering making my own lightweight mount for just this sort of application and posted a thread on it which I now can't find.  That was before I had a 3D printer and I had designed it mainly in aluminium and wood.  It used MXL timing belts and pulleys rather than worm and worm gear.

Here are a couple of the designs in SketchUp - aluminium and wood.

However, I'm thinking now that instead of a GEM I might go for a fork mount.

[nmoushon]

If you are thinking of a fork mount I think you could buy a used one cheaper than you could build one. With it only being a widefield rig you dont need the extreme accuracy; plus guiding will take care of most of it. Also a used one will have all you guide ports and GoTo already. Its not a bad idea and would keep you obsy foot print smaller. This would really work well with a clam shell style roof, imo. I think if you go with a clam shell roof  you might even be able to mount it straight to the pier itself and not need an actual building...unless you want a building around it.

If you are thinking of a GEM then again I think its still cheaper to buy a used one instead of building one. Again because of it being a wide field rig you dont need a super accurate mount and thus cutting down on the price of  the needed mount. The new AVX mount has been getting some really good reviews and I've seen several up for second hand already even though they are still pretty new. The GEM style mount does need a bigger foot print than the fork but if you go with a smaller mount is there a consiterable size difference? I'm not sure. Plus if this mount is only going to be dedicated to this wide field rig you could build a short counter weight bar and cut down the foot print even more. You'll need a bit more weight bc of the shorter bar but I'm sure that would be cheaper than building a biger building.

[nmoushon]

Also if you are wanting to build one just because you can then thats different and throws everything out the window. If the price isnt that big of an issue then I would go with the fork. Smaller foot print and with not as big of a need for super accurate tracking I think it would work out great, especially with no meridian flip to worry about!

[Tinker1947]

Don't know how many rigs/dslr you would get on one of these, i will be buying one soonish, just making a power pack to run it and dew heater.....will mount it on my NEQ6 tripod.....

http://www.firstlightoptics.com/skywatcher-star-adventurer/skywatcher-star-adventurer-astronomy-bundle.html

[Gina]

I like the idea of a fork both to reduce size and the lack of meridian flip.  It would certainly save some work to buy one, if I can find one second hand   But I still will be considering making one sometime in the future I think - if I live long enough   I have a long list of projects to do first.  I can always play with the design in the meantime - and dream

[Gina]

Don't know how many rigs/dslr you would get on one of these, i will be buying one soonish, just making a power pack to run it and dew heater.....will mount it on my NEQ6 tripod.....

http://www.firstlightoptics.com/skywatcher-star-adventurer/skywatcher-star-adventurer-astronomy-bundle.html

That looks very interesting - thank you

[Dave_D]

Do a horseshoe just for giggles

[Horwig]

Gina, don't discount a GEM on the grounds of meridian flips. With the short overhang of your rig you probably won't need to flip. My widefield rig is a pentax 300mm lens and an sx h35, I've lifted the assembly off the dovetail by a 5 cm block, now I can reach virtually the whole sky with no flips.

If you fancy another project I have an eq5 which I'm no longer using, it has dc motor drives with encoders, but no control electronics, any use?

Huw

[Gina]

Do a horseshoe just for giggles

Will this do?

[Gina]

Gina, don't discount a GEM on the grounds of meridian flips. With the short overhang of your rig you probably won't need to flip. My widefield rig is a pentax 300mm lens and an sx h35, I've lifted the assembly off the dovetail by a 5 cm block, now I can reach virtually the whole sky with no flips.

If you fancy another project I have an eq5 which I'm no longer using, it has dc motor drives with encoders, but no control electronics, any use?

Huw

You're probably right about meridian flip not being needed    I'll let you know if I fancy an EQ5 project - thank you

[Gina]

I shall be staying with the NEQ6 mount for my widefield triple imaging rig for the time being but will look into the design of a fork mount with timing belt drive for a possible future construction project.

My earlier design used 10mm thick acrylic discs for the large timing pulleys but with my projects for bigger 3D printers I might print these instead.  With a size of 200mm or more I can achieve quite a large reduction ratio for the final drive.  A 200mm pulley would give just over 300 teeth at MXL pitch of 2.03mm so with a 10t small pulley I can get 30:1 reduction.  I'll post some calculations later but I have two possible ideas for the drives.  One is to use Nema 17 stepper motors with microstepping and the type of control used in 3D printers.  The other is the small stepper motors with built-in gearbox as I use for remote focussing if these will provide enough power.

[Gina]

There are two extremes to be covered by the drive systems - a resolution of the order of one second of arc and the ability to move quickly from one position to another.

Taking the resolution issue first, the 1" needs to be translated to one microstep of the stepper motor.  In the case of the Nema 17 the step angle is 1.8 degrees and we can achieve 16x microstepping using the A4988 stepper controller chip in a RAMPS controller as in RepRap style 3D printers.  So the microstepping angle is 1.8/16 degrees = 0.1125 degrees = 6.75 mins = 405 secs.  Therefore we require an overall reduction ratio of around 400x.  So in addition to the 30x achieved in the final drive we require a further reduction of 40:3 or about 13:1.  This is too much for a single set of standard timing pulleys with 10t smallest and 100t largest.

Three solutions - do it in two steps, use a worm drive or increase the final drive large timing pulley.  The Avalon M-Uno mount uses an extra two step timing pulley reduction.  Having escaped from the deficiencies of a worm drive it would be silly to add it back in so I'm rejecting that.  To use a single drive we could increase the final drive pulley to 300mm diameter and then use 10:1 ie 10t to 100t first reduction drive.

OTOH with widefield do we really need 1" resolution?  I think not.  1" corresponds to about a pixel with a 1000mm FL and 5micron pixel size.  The longest FL I shall use will be 200mm so a resolution of 5" corresponss to a pixel.  OK so the resolution wants to be better than a pixel to get the best images without oval stars but a guiding resolution of 2" gives better than half a pixel resolution and will be quite adequate.  So we're back to an overall two stage timing belt reduction drive.

[Gina]

Turning now to the other type of stepper motor, the little 28BYJ-48.  Step angle is 5.625°/64 = 0.087890625° = 5.2734375' = 316.4"  Call it 300" so 2" resolution of pointing means a reduction ratio of 150:1 whch may be achieved by 30:1 and 5:1 reduction ratios.  In fact in my focuser projects I use half stepping so that would require a reduction ratio of just 75:1 allowing an increase from the tiny 10t timing pulleys to something a bit bigger while still needing only a 2 stage reduction.  So the 28BYJ-48 stepper motor can readily achieve the precision required with ease.

Now to the fast slewing mode.  With the 28BYJ-48 the maximum step rate is 100 steps per second.  So the final drive rate will be 100 x 5.625°/64 divided by the reduction ratio of 75 giving 0.1171875° per second (if my calculattions are right).  That is very slow compared with about 3 degrees per second that standard mounts achieve.  eg. 90° slew would take 768 secs or nearly 13 minutes.  That is NOT good   I could probably go to 16x microstepping for tracking/guiding giving an 8x reduction in drive ratio and 8x faster slewing of around 1 degree per second.  Not so bad (if the power proves sufficient).

[Chris]

Do you have a spare Peltier cooler for you Brain Gina?

I know you're putting this one on the back burner but for the record I think its a fantastic idea to build your own mount, plus its about the only thing you haven't done yet!

P.s. The triple rigs looking ace!

[Gina]

Thank you Chris

I have always wanted to build my own mount   I've also desired to make my own scope but that is not on my agenda at present.  I reckon I will build a mount eventually

The triple widefield imaging rig will depend on turning a couple more adapters which in turn needs either warmer weather, or enough energy on my part to clear some room and get help to move my lathe into the store room.  The latter to be my workshop when I get more organised   All this needs my health to improve from this ongoing dreaded virus

The medium triple imaging rig basically just needs me well enough to go out at night plus clear night skies, of course!

The virus is not affecting my brain too much fortunately though that does mean I get a bit bored and frustrated at times.

[Chris]

That virus is lasting ages, is it slowly getting better at least? If not have you thought about bringing your GP up to speed?

[Gina]

Yes, thank you Chris, it is slowly getting better   I've heard there is a lot of it about and some people have had it even longer   It seems to be very virulent

[Chris]

Well I'm pleased to hear its getting better at least Hopefully your body will be back on pace with your brain soon

[skybadger]

Hi Gina

I think you are underselling the smaller stepper. 100 steps per sec seems too low to me for a small motor.

If you accept that you can vary the microstepping to ramp up the speed of the slew, then you can move in whole steps quite quickly .

At 100 steps/sec, 30 steps/rev and gearing of 1:150 you now achieve 1/50th rev per second or 50 secs for a complete circle.

The ramp up can also be driven by overdriving in voltage with ballast resistors and current chopping to control the current.

I am also not sure you need the dec axis. Think of a calibrated wedge as the dec mount for the rig using a motor driving a very stiff altitude screw. It depends on the degree of control you are aiming for.

Mike

[Dave Lloyd]

I think if Gina builds a scope it will be after she's knocked up an arduino controlled mirror grinding machine.

[Gina]

Hi Gina

I think you are underselling the smaller stepper. 100 steps per sec seems too low to me for a small motor.

If you accept that you can vary the microstepping to ramp up the speed of the slew, then you can move in whole steps quite quickly .

At 100 steps/sec, 30 steps/rev and gearing of 1:150 you now achieve 1/50th rev per second or 50 secs for a complete circle.

The ramp up can also be driven by overdriving in voltage with ballast resistors and current chopping to control the current.

I am also not sure you need the dec axis. Think of a calibrated wedge as the dec mount for the rig using a motor driving a very stiff altitude screw. It depends on the degree of control you are aiming for.

Mike

Yes, I agree that it should be possible to get more than 100 setps per second and I think I'll run a test.  This stepper is internally wired as a uni-polar motor but a small change will allow bi-polar use and an H bridge drive.  That should give twice the power as all of each winding will be used rather than half of it.  I think I should be able to use the A4988 chip which gives 16x microstepping and H bridge output. 

[Gina]

Well I'm pleased to hear its getting better at least Hopefully your body will be back on pace with your brain soon

I'm feeling a lot better today and have been out to the local farm shop and stocked up on groceries including plenty of fresh vegetables.  I have also been out to the observatory and checked my imaging rig.  I have a few problems to sort out   Though I'm feeling better I won't be overdoing it and decided against going out to WI this evening.

[Gina]

OK then returning to calculations for the 28BYJ-48. 

Step angle is 5.625°/64 = 0.087890625° = 5.2734375' = 316.4" and using 16x microstepping gives 19.775"  Call it 20" so 2" resolution of pointing means a reduction ratio of a mere 10:1 whch may be achieved in one stage with a 30t motor pulley with the 300t large pulley (or maybe 20mm motor pulley and a correspondingly smaller large pulley).

In the fast slewing mode I feel pretty confident that I could get 200 steps per second.  So the final drive rate will be 200 x 5.625°/64 divided by the reduction ratio of 10 giving 1.7578125° per second (if my calculattions are right).  That is a bit over half the 3 degrees per second that standard mounts achieve and would be acceptable.

In conclusion, yes, assuming sufficient power, the little 28BYJ-48 might well do the job

[Gina]

I've come up with some info from testing a while back.  The output shaft of the 28BYJ-48 will do one revolution in 10s that's 36°/s.  So with a step down ratio of 10:1 that's 3.6°/s - well up to standard mount slewing speed   RESULT

[Tinker1947]

These seem cheap, are they the same thing........