Jump to content

Narrowband

DIY Fork Mount for Widefield Imaging Rig


Gina

Recommended Posts

This project is back on having virtually completed my widefield imaging rig and wanting to reserve the EQ8 mount for telescopes.  The plan is to make a fork mount and matching dome based mini observatory for automatic remote controlled operation.  This follows on from a previous thread relating to a mount for my triple imaging widefield rig which is now defunct.

Edited by Gina
Link to comment
Share on other sites

Basic fork mount.  The box represents a telescope or other imaging system.  The main (large) axle is RA and is aligned with the polar axis with the fork pointing towards the pole - the small axle at the end of the fork is the DEC axis.

WF Fork Mount 01.JPG

Link to comment
Share on other sites

Both axes will have drive gears or pulleys for pointing the mount and traversing the sky in RA during imaging.  The main axle will be mounted in bearings on platform (pier) with azimuth and altitude adjustment for aligning to the polar axis (Polar Alignment).  This is critical, particularly if guiding is to be avoided.  This means two more adjustments/variables apart from the RA and DEC controls for the pointing.  Since the imaging rig will be enclosed in a fairly close fitting container (observatory) they will also want to be remote controlled.  However, these should only need setting rarely and do not need absolute accuracy (they are in a manual control loop).

I plan to use NEMA17 stepper motors for the RA and DEC controls for accuracy but the AZ and ALT PA controls could use the ubiquitous 28BYJ-48 mini stepper motor with built-in gearbox.  These are not suitable for the RA and DEC controls not because they're too lightweight but because the gearbox has a peculiar ratio (a little under 64:1)

Link to comment
Share on other sites

I'll look into the gearing and drive systems next. 

  1. Longest FL lens I shall use with this mount will be 200mm and the camera resolution is 3.8microns per pixel.  Pixel resolution should be fine.
  2. Angular resolution is given by tan(θ) = 3.8 / 200x1000 = 0.000019 giving θ as about 0.001°. 
  3. NEMA17 steppers have 1.8° per step but using 16x micro-stepping = 1.8 / 16 = 0.1125°.  I think we can call that 0.1°
  4. So to get a resolution of 0.001° we need a "gear" ratio of 100:1.  A slightly larger ratio to give a workable steps per second of arc would be good.

This will be suitable for the RA axis where the drive is moving to follow the target.  With good PA the DEC axis should not need changing during the exposure and less accuracy will be adequate.  Of course this does depend on good PA but I hopeful of achieving this.  The main requirement of the DEC control will be framing the target withing the imaging frame and I think something like 10% of the frame should be adequate.

Link to comment
Share on other sites

So is the plan direct drive to the axis via belts ?

Guess some precision engineering will be required, as it will be unguided will you have any way of controlling the drive speed or are you "stuck" with the final drive ratio once built ?

Dave

Link to comment
Share on other sites

Continuing with the DEC control, the camera resolution is 4656×3520 so allowing 10% of the height gives about 350 pixels.  This would mean even a 1:1 gear ratio would work but we can do much better than that.  Something like a 10:1 ratio would give a 10 pixel resolution and be easy to implement.  From my experience with 3D printers fishing cord and drums would be quite adequate.  A micro-stepping mode of less than 16 would be adequate too.  I might stick to timing belt drive though - I'll think about it.

Link to comment
Share on other sites

Just now, Davey-T said:

So is the plan direct drive to the axis via belts ?

Guess some precision engineering will be required, as it will be unguided will you have any way of controlling the drive speed or are you "stuck" with the final drive ratio once built ?

Dave

Yes - direct drive via belts.  The drive speed is controlled by the electronics and software that drives the stepper motors.  I think I'll be able to use an INDI driver that was devised for one of the proprietary mounts by altering the drive ratio setting in the software.  Then the mount will work just like any other under control of KStars/Ekos on the client computer.

The hardware of the mount will need fairly precise engineering but nothing like that needed for a mount carrying a long focal length telescope and long exposure imaging.

  • Like 1
Link to comment
Share on other sites

The RA drive will need a double step down to get the 100:1 plus rotation ratio between motor and axis. 

I think MXL size timing belts and pulleys will easily handle this application and I have used these extensively for various projects in the past.  I have a 100t MXL timing pulley with 5mm bore plus some others - I'll have to see what I've got.  The smallest pulley with a 5mm bore has 15 teeth so the first step down belt drive will be with 15t motor pulley and 100t on an intermediate shaft giving a ratio of 3:20 (1:6.67).  The final drive can have a 15t pulley on the intermediate shaft and a large pulley sized to give an overall ratio of 1:100.  This would have 100 x 3 / 20 x 15 = 225 teeth.  The pitch diameter works out at about 145mm which seems quite reasonable.

Since mount axes only have to rotate 180°, the belt could simply be attached by its ends to points on the pulley, saving the need for teeth and meaning the pulley could simply be turned from a 150mm diameter blank on the lathe.  I've done this size pulley before on my little Chinese heap of a lathe quite successfully.  OTOH I could 3D print a timing pulley and see how well it works - I can achieve a concentricity of around 0.2mm in 100mm with my latest 3D printer setup.  This just might be enough with an exposure of around a minute or less.

Edited by Gina
Link to comment
Share on other sites

I guess I should be able to calculate the accuracy required given the exposure time and angular resolution.  Let's see... 

  1. Earth rotates at 15° per hour.
  2. One minute is is 1/60th of an hour so the angle to traverse in one minute is 15/60 = 0.25°.
  3. One pixel is 0.001 degrees of arc.
  4. Hence 250 pixels are traversed in a minute.
  5. Accuracy required is 1 in 250.

This is about one tooth of the MXL belt - I think I should be able to manage that.

Link to comment
Share on other sites

I don't think this mount needs to be as accurate as I thought :)  I reckon I might get away with 3D printed gears.  That would simplify things a lot.  I could try it and see how well it works.

Link to comment
Share on other sites

I've had some "ordinary" EQ  mounts that were pretty rubbish at unguided tracking, much / most of which was down to poor execution of the basically OK design.

Misaligned gears and wobbly worms causing huge PE, it doesn't take much to lose the necessary accuracy even for short f/l wide field especially for longer exposure NB subs.

Not necessarily the overall tracking more sudden jumps.

Not sure how accurate printed gears are but I'd be inclined towards belt drive.

Dave

Edited by Davey-T
Link to comment
Share on other sites

Yes, I think I'll stick to belt drive particularly for the final drive to the RA axis.  I already have some MXL belting and several timing pulleys so no problem.  The drive from motor to intermediate axle could be 3D printed gears though (as long as I take care of backlash).  In fact I've found that fishing line cord and drums were more accurate than timing belt and pulleys on my 3D printers.  The limitation of this system is that it can only handle limited motion - not round and round.

 

Link to comment
Share on other sites

The fork part is pretty straightforward but I need to work out the mounting for it.  I don't think plastic will be suitable as it is liable to move, particularly in altitude (reducing it due to sag) so I'm thinking two aluminium plates mounted on a swivel base.  The azimuth adjustment is straightforward - a large spur gear or maybe a lever with a gear rack on the end but the altitude needs more thinking about.  Think I'll sleep on it :D

Link to comment
Share on other sites

17 hours ago, Gina said:

The fork part is pretty straightforward but I need to work out the mounting for it.  I don't think plastic will be suitable as it is liable to move, particularly in altitude (reducing it due to sag)

Would a 3D printed box structure with internal bracing be strong enough for the job, or is this not an option? I've been thinking about doing something similar for an alt-az mount with coarse and fine controls, but so far the design is only in my head until I get my printer

  • Like 1
Link to comment
Share on other sites

Here's a cross-section diagram of the mount, imaging rig and dome enclosure/observatory.  Just the basic geometry drawn to scale without drive motors and belts etc.  The dome shown is 600mm diameter and more than big enough.

590719d33095d_DomeMount01.png.effabfece0c53a1e8ef63509e1423d56.png

Link to comment
Share on other sites

Been trying to work out all the drives and their motors as well as the mounting hardware and most seems to indicate an aluminium structure.  Maybe with 3D printed parts as well but...

This diagram shows the RA axis drive, plus the altitude and azimuth geometry.  The DEC drive would be on the fork.

590736a203890_DomeMount02.png.a1032adf187043ab0d757b8dc357937e.png

Link to comment
Share on other sites

Hmm...  Taking drives and supports together doesn't seem to give the answers.

I think I'm coming back to 3D printing with extra support.  Anyway, 3D printing makes for easy design and could act as a test setup to prove the geometry. 

I'm thinking of ball bearings for the RA axis and ball bearings would be easiest for the DEC axis as well.  The widefield imaging rig weighs just under 3Kg including an 11" ADM dovetail bar which would probably not be used with a fork mount.  This 3Kg will be overhung on the main RA bearings by 200mm from the top bearing.  If the RA bearings were spaced by 200mm the top bearing would be supporting 6Kg due to the leverage plus the weight of the forks etc.  The spacing shown in the diagram above is 140mm but can be increased fairly readily.  This is making me think of quite large bearings.

Link to comment
Share on other sites

I have a pair of very large pillow block ball bearings which I bought for the fold-down flap on my observatory but changed to some lighter weight stainless steel cased bearings.  I get the impression that to use these for the fork mount would be like "using a sledgehammer to crack a nut" but I wonder.  I have these and no current use for them and they fit some aluminium pipe I have.

590782bfdd234_BigBallBearing01.thumb.JPG.a4752f885eee7179927ffa5f43aaabff.JPG

Edited by Gina
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.