Telescope Leadscrew Advice

[Dr_GMJNParticipant @dr_gmjn]

All, I’m making a fine adjuster for altitude on my DIY Dobsonian telescope mount. I have just built an equatorial platform for it for tracking. I’ve 3D printed a 200mm long lever arm, and now need a leadscrew. I had originally designed it for an M8 fine thread, for which I was going to buy some threaded rod, and then get a tap to thread the corresponding nut.

For the highest magnification eyepiece I’ll ever be using, I’ve calculated that to move a point from one side of the field of view to the other, it would take about 1/3 of a turn of the adjuster. For lower magnifications, it would be around 3/4 of a turn. I need to get a compromise between getting a fine enough adjustment for small targets like stars and for panning around the Moon’s surface at high magnification, but then when trying to find an object just out of view at low magnifications, not wanting to have to turn the wheel loads of times (or disengage the lever arm clutch knob and move manually).

I think 1mm pitch is about right, but ideally would be around 0.75mm pitch to give a slightly finer adjustment. Since I’ve already got some 8mm thrust bearings and hand wheels with a 6mm bore, I’d like to stick with something around M8.

Are there any finer “standard” threads to use, with good availability of taps/dies or threaded rod? The overall leadscrew length needs to be 150mm.

Also, if I do stick with M8 fine, is it better to buy threaded rod, or thread plain bar with a die? The leadscrew nut is nylon, and I’m hoping that I can get zero backlash by using that in conjunction with a metal thread. The precise tolerance on the thread itself is unimportant, as is the straightness of the thread (within reason). I don’t really want to get into custom thread turning bar and nuts on the lathe – I want to use the telescope rather than spend more days making parts for it, plus I’ve yet to come up with a design for azimuth fine adjustment, which is far more difficult.

Thanks.

[DC31kParticipant @dc31k]

Use a differential thread. One standard M8 x 1.25; the other M8 x 1.0 . Have the nut on the coarse pitch side rotatable for coarse adjustment and then keep that nut stationary and rotate the differential thread for an effective pitch of 0.25mm.

The beauty of a differential thread, especially if you would 3DP a nut, is the actual pitch does not matter, only the _difference_ in pitches – and for 3DP a coarse pitch is easier to print (so M24 x 3 and M22 x 2.5 ISO coarse would make a 0.5mm pitch differential).

[JasonBModerator @jasonb]

I use M8 x 0.75 quite a lot on my models so easy enough to get hold of taps and dies

[DC31k]

On DC31k Said:

Use a differential thread. One standard M8 x 1.25; the other M8 x 1.0 . Have the nut on the coarse pitch side rotatable for coarse adjustment and then keep that nut stationary and rotate the differential thread for an effective pitch of 0.25mm.

The beauty of a differential thread, especially if you would 3DP a nut, is the actual pitch does not matter, only the _difference_ in pitches – and for 3DP a coarse pitch is easier to print (so M24 x 3 and M22 x 2.5 ISO coarse would make a 0.5mm pitch differential).

Thanks – interesting idea, although I’m not printing the threads, so it might be a bit complicated for what I need. I think a compromise thread pitch would be best.

[JasonB]

On JasonB Said:

I use M8 x 0.75 quite a lot on my models so easy enough to get hold of taps and dies

Thanks Jason. I didn’t know that a 0.75 pitch M8 thread was a standard. If I can find the right hardware I’ll use that.

[Nigel Graham 2Participant @nigelgraham2]

If you use studding buy the stainless-steel version as the finish is generally better than on plated mild-steel.

I am not sure it is available in M8 X 0.75 form though (Metric Fine series).

[Dr_GMJN]

On Dr_GMJN Said:

…it might be a bit complicated for what I need.

I think we have differing definitions of complexity.

A differential thread could be made with off-the-shelf parts and using a drill press (with a couple of wooden jigs).

You would need:

A piece of M8 x 1.25 threaded rod.

A piece of M8 x 1.0 threaded rod. This is available: https://www.gwr-fasteners.co.uk/m8-x-1mm-x-1-metre-fine-pitch-threaded-bar-din-976-1—high-tensile-steel-88-zinc-plated-68869-p.asp

Two M8 x 1.25 studding connectors (long nuts).

An M8 x 1.0 standard nut.

Join the two pieces of studding end-to-end by drilling, tapping M4 and using M4 threaded rod and loctite.

Drill out half the threads in one of the connectors, slap some JB Weld over the join in the two rods and put the nut central over the joint. It reinforces the joint and gives something to grip when turning the rods. You could 3DP a fancy knob with a 13mm internal hexagon to go over the connector.

The design challenge is making the long nut on the far end of the coarser threaded rod rotatable, for the coarse adjustment. The setup would give a 5:1 ratio between coarse and fine adjustment.

[DC31kParticipant @dc31k]

Too late to edit above post, but an even simpler solution, provided one of the rods could increase to M10, is one of these:

M8 – M10 Rod Reducer

This is a very easy way to make a 0.25mm pitch screw – two bits of standard ISO coarse rod, a reducer nut and a dab of threadlock. No messing around trying to source metric fine hardware, no drills, taps or dies required.

[Neil WyattModerator @neilwyatt]

One of the advantages of a Dobsonian is being able to rapidly swing it around to a new target.

A 0.75mm pitch thread should be fine, and if you want finer adjustment, just 3D print a larger handle.

3D print a simple clamp with a matching thread for the far end of the rod. This can be a block with a hole for the rod and another hole for a piece that matches the rod can be. You can figure out one of dozen ways to hold that part in place.

[DC31k]

On DC31k Said:

Too late to edit above post, but an even simpler solution, provided one of the rods could increase to M10, is one of these:

M8 – M10 Rod Reducer

This is a very easy way to make a 0.25mm pitch screw – two bits of standard ISO coarse rod, a reducer nut and a dab of threadlock. No messing around trying to source metric fine hardware, no drills, taps or dies required.

 

I’d say that tapping a hole in some nylon is less complex than making a differential thread…but of course that also gives less functionality than what I think you’re suggesting.

For some context, this is the torque arm I’ve designed. Self-explanatory, apart from perhaps the bushes in the leadscrew bracket, and the thrust bearings behind each hand wheel:

And this is it printed, with the leadscrew balanced in place. There will be a handwheel either side of the thread. The large printed knob in the centre is a clutch that can be released to allow the telescope to be easily moved by hand (as it is by default). It also allows fast assembly / disassembly for storage. I’ve also added a smaller knob on the leadscrew nut, which allows the arm and nut to be locked together – just in case I end up with backlash or wear in the lower end of the mechanism:

I want to adjust the lever arm while looking through the eyepiece, to centralise an object. It doesn’t need to be within microns of central, the main thing is the thread is the best compromise between how much I have to rotate the wheel vs how fast the object crosses the field of view (which depends on magnification and the field of view of the eyepiece). At present with a 0.75 pitch, and a higher magnification eyepiece, it’s about 1 turn, which seems reasonable.

Happy to try a differential thread if it could be designed for example that turning one handwheel gave coarse adjustment, and the other fine. What I wouldn’t want is the need to have a hand on each wheel at the same time – that would be awkward to operate while looking therough the eyepiece.

I think at this point, a diagram would be helpful to show what you’re suggesting.

I did order an M8 x 0.75 tap and die, and I have some 8mm brass rod that I was going to thread for this. I also have standard M8 taps and a die, so I do have the means to make the two threads.

Thanks.

[JasonBModerator @jasonb]

Another option for finer adjustment would be to print a couple of gears and then have a second handwheel with a small pinion drive the handwheel on the 0.75pitch screw.

Turn one handwheel for fast adjustment and the other for fine.

[JasonB]

On JasonB Said:

Another option for finer adjustment would be to print a couple of gears and then have a second handwheel with a small pinion drive the handwheel on the 0.75pitch screw.

Turn one handwheel for fast adjustment and the other for fine.

Yes, but the problem is backlash; if I’m hunting to centralise something (which is inevitable), and there’s any bit of backlash, it will become a pain, especally in conjunction with the stiction in the bearings on the altitude axis (teflon on textured laminate). Even though it’s hardly anything, it can cause issues.

This is why I went for a nylon leadscrew nut – I’m hoping initially it will be slightly tight on the screw to eliminate backlash. If it’s not, I’ll have to figure out a way of splitting the nut and pre-loading it to the thread. I don’t think it will come to that though.

All this is assumed – It’s a work in progress. It might all be fine as-designed even with a bit of stiction and backlash, but I’d rather at least consider it before making anything. I’m improving the telescope incrementally:

I made the Dobsonian mount to get rid of the terrible standard equatrial mount that vibrated badly through lack of stiffness (it didn’t track properly either).

With the vibration solved with the Dobsonian mount, then the images drifed out of view becasue there’s still no tracking.

So then I’ve made an equatorial platform, which freezes the images in the eyepiece.

But…now the images are stationary, it becomes more important to centralise them intitially, and to not accidentally knock them out of alignment. Hence the current work.

It would be easier to just buy a modern telescope with all this functionality built-in, but where’s the fun in that? Luckily the mirrors in this telescope are known to be high quality, despite them being fitted into rather budget hardware. It’a a Meade 10″ Starfiner from the early 1990s.

[Michael GilliganParticipant @michaelgilligan61133]

To my mind … it’s crying-out for a ball-screw

MichaelG.

[JasonBModerator @jasonb]

can you get them at such a fine pitch?

[Michael GilliganParticipant @michaelgilligan61133]

1mm pitch is easy [albeit potentially spendy] from Gloucestershire
https://www.matara.com/videos/in-house-ballscrew-machining/

MichaelG.

.

Alternatively: https://www.steinmeyer.com/fileadmin/media/downloads/en/Steinmeyer_Catalogue_Miniature_Ball_Screws_3-16mm_EN.pdf

[Neil Wyatt]

On Neil Wyatt Said:

One of the advantages of a Dobsonian is being able to rapidly swing it around to a new target.

A 0.75mm pitch thread should be fine, and if you want finer adjustment, just 3D print a larger handle.

3D print a simple clamp with a matching thread for the far end of the rod. This can be a block with a hole for the rod and another hole for a piece that matches the rod can be. You can figure out one of dozen ways to hold that part in place.

Thanks Neil.

I’m not sure what you’re getting at with the last paragraph about the clamp? What’s all that for?

[Robert Atkinson 2Participant @robertatkinson2]

Have you thought of putting a stepper motor on it? a small driver board with step/direction and a humain interface. This could be a single speed setting knob and Left-Right push buttons, sone kind of joystick or my prime suggestion, 6 push buttons in a row. 3 left, 3 right with the center two being single step, next two slow and outer pair fast. That can all be done with a 555 timers or if code is your thing your favorite microprocessor (I’d go for a pic and build the stepper driver into it for a bipolar motor. That just needs the pic, four transistors and a couple of resistors and capacitors.
You can even get steppers with a leadscrew through them e.g. https://www.mclennan.co.uk/product/35dbm-l-permanent-magnet-stepper-digital-linear-actuator

Robert.

[Robert Atkinson 2]

On Robert Atkinson 2 Said:

Have you thought of putting a stepper motor on it? a small driver board with step/direction and a humain interface. This could be a single speed setting knob and Left-Right push buttons, sone kind of joystick or my prime suggestion, 6 push buttons in a row. 3 left, 3 right with the center two being single step, next two slow and outer pair fast. That can all be done with a 555 timers or if code is your thing your favorite microprocessor (I’d go for a pic and build the stepper driver into it for a bipolar motor. That just needs the pic, four transistors and a couple of resistors and capacitors.
You can even get steppers with a leadscrew through them e.g. https://www.mclennan.co.uk/product/35dbm-l-permanent-magnet-stepper-digital-linear-actuator

Robert.

Thanks Robert. Yes, I did think of that, and also going a step further and putting one on the azimuth axis so I could track targets using a guide star. Then we get into field rotation when using altitude and azimuth to track a target that’s moving in an arc. It’s all possible, and has been done before, but then I’d be complicating things a step too far than I want to.

My driven platform eliminates rotation (at least for an hour or so), and once I’ve sorted the initial fine adjustment and locking out, I should be able to have a fairly relaxed method of visually observing planets and the Moon.

[Dr_GMJNParticipant @dr_gmjn]

<p style=”text-align: left;”>The next thing I need to do is make a similar fine adjustment for the azimuth axis. For the mount I used a lazy Susan bearing. The inner part is fixed to the upper mount (the part that rotates). The outer part sits directly on the equatorial platform, just located by three sectors, so the mount can be lifted on and off the platform by hand.</p>

Whatever adjuster method I end up with needs to be very low geared, no backlash, lockable (or no back-drive), and be capable of being disengaged easily for free rotation by hand.

The adjuster wheel also needs to be accessible while viewing through the eyepiece (so on an extension to raise it up from the base to hand height).

I thought of using a small diameter friction roller against the outside of the bearing (or an upper extension of it to avoid the location sectors). Possibly using an over-centre type latch to engage or disengage it.

It’s not as easy as designing the altitude adjuster. Any ideas? A job for Robert’s stepper motor and remote control idea? Thanks.

[duncan webster 1Participant @duncanwebster1]

Unless it’s well balanced I would have thought gravity would take out backlash in the altitude adjuster

[duncan webster 1]

On duncan webster 1 Said:

Unless it’s well balanced I would have thought gravity would take out backlash in the altitude adjuster

<p style=”text-align: left;”>True, but I made a set of adjustable tube clamps so I can balance it pretty well. It’ll not be perfect of course, so that might work in my favour.</p>

[Dr_GMJNParticipant @dr_gmjn]

In terms of stepper motors – can anyone give advice on types/sizes/voltages, and how to control them? I’ve heard an arduino would be a good option, but don’t really know where to start.

For the azimuth system I’ve got in mind it would be a direct friction drive roller against the edge of the slew bearing. It would need a brake, and a rotary speed control, the type where you can keep turning it (either CW or CCW) and the motor will move in sync, then lock when you stop turning.

I could also add one to the altitude control (as previously suggested) if it works well, and also another to drive the equatorial platform.

The loads involved are quite low – a finger can move both axes quite easily when they’re balanced.

The mechanics I can do, but the control of steppers I need to learn fast!

Thanks.

[duncan webster 1Participant @duncanwebster1]

Depending on what resolution you need,a stepper motor can act as its own brake,just leave it energised,possibly at a lower current, but not essential.

An arguing based rotary table controller might be a good place to start for software

 

[duncan webster 1]

On duncan webster 1 Said:

Depending on what resolution you need,a stepper motor can act as its own brake,just leave it energised,possibly at a lower current, but not essential.

An arguing based rotary table controller might be a good place to start for software

 

Thanks Duncan.
<p style=”text-align: left;”>I found an Arduino YouTube video of exactly this kind of functionality, but figuring the programming out will be interesting. Control could be a rotary encoder which would take care of CW/CWW.</p>
<p style=”text-align: left;”>I’m working on the following assumptions at the moment. Might be wrong, but even if they’re not, I seem to have some work to do…</p>
If I assume a 0.3 degree field of view, a 20mm diameter drive roller and my slew bearing is 390mm diameter, that gives a 1:19.5 ratio. So one rev of the drive would be 360/19.5=18.5 degrees of the telescope. According to Google, a Nema17 motor has 200 steps per rev, so if one click of the rotary control could be 18.5/200=0.09 degrees of the telescope.

So to traverse a star from one side of the field of view to the other would be 0.3/0.09= 3 clicks…which isn’t enough. I’d think it needs to be at least 5 clicks. Need to reduce drive diameter, or find another way of reducing degrees per click I guess.

 

 

 

 

 

[duncan webster 1Participant @duncanwebster1]

You can set a stepper controller to do microstepping,which as its name suggests means it does a lot less then 1/200 of a rev every time it gets an input  pulse from the arduino. I’m not sure how well it would stand still doing that, you can also get gearboxes to bolt straight onto stepper motors.ivd gog one somewhere,ill dig it out and take a picture. For something this simple,where you at a minimum only need clockwise or anticlockwise, continuous or jog you could actually do it with something like a 555 and some push buttons, although I’d use a processor as you are bound to want to add functionality. I always do.

[Author]

Posts