A Cheap Huge Worm Gear

[Vroobel]

Guys, did you ever think about really big worm gear? The one giving high precision, finally even better than 0.5 arcsec / micro step of your stepper motor?

I just make one of two worm gears for my Dobsonian mount. There is 450mm of diameter and 890 teeth. My stepper motors can work with 1.8 deg step and 8-32 micro steps. Let's assume I will set up the set as 200 steps x 8 micro steps x 890 teeth of worm gear, then I receive 1424000 micro steps per revolution, that means 0.91 deg / micro step. If I set 16 micro steps (that still gives quite good torque), I have 0.45 deg / micro step. I don't consider using 32 micro steps.

Please look at pictures below and also visit my YouTube channel to see the latest video describing how its made.

https://www.youtube.com/watch?v=u7_AuXA8eNs&t=10s

https://www.youtube.com/channel/UC9AMgmFpcYcRZuP65-SghPA

[saac]

Excellent effort Thomas, I like the way you solved problems by making your own tooling, well done .  Is this now installed in your Dob , if so what is your pointing/tracking accuracy like?

Jim 

[Gina]

Brilliant   But i think it's more brilliant than you think - I think your calculation is wrong.  I agree with 1424000 microsteps per revolution, but not the microsteps per degree.  Dividing 1424000 by 360 gives 3955.55 microsteps per degree or in other words 0.000253 degrees per microstep.  HTH.

[Vroobel]

Thanks, Gina! Of course, you are right, I mean arcsec, not degrees. Sorry for the huge mistake, so huge as the wheel...

 

[Star101]

Very impressive.

Would love to see this in action. It could be exceedingly more accurate than most

The seeing would be your biggest worry

 

[Vroobel]

4 hours ago, saac said:

Excellent effort Thomas, I like the way you solved problems by making your own tooling, well done .  Is this now installed in your Dob , if so what is your pointing/tracking accuracy like?

Jim 

I'm just working with the drive, so I can't tell you anything more than I expect precision better then mentioned above 0.91.or 0.45.arcsec / micro step, depending on number of micro steps. 

[skybadger]

Hi Thomas, 

What's your experience in use ? I would be looking to measure expansion/contraction with humidity and worry about flatness and reproduce ability for PEC or auto guiding. But then you're using a dob, so maybe you're not driving the scope an d just using them to hold position ?

[Vroobel]

That's valuable notice! As every wooden (plywood, MDF) part of my addons, this wheel will be protected by a kind of fluid against the moisture. My first wheel for Alt axis was made of the MDF. The actual wheel is made of the plywood, which is much better because of plenty of crossing layers. 

I have the Dob, so both the PEC and the auto guiding are not a problem for me. But the mount is motorised and the drive is just amending. I use RPi for a transformation of coordinates, the telescope was able to follow the objects even with previous drive using a GT2 timing belt, but the precision was not satisfactory.

[Merlin]

It's good to know that in this age of  off-the-shelf products, there are still skilled ATMers ( Amateur Telescope-Makers ) out there.

[Vroobel]

[Vroobel]

11 hours ago, Star101 said:

Very impressive.

Would love to see this in action. It could be exceedingly more accurate than most

The seeing would be your biggest worry

 

There is easy way to see it in action, just enough to meet somewhere under a dark sky. I hope to finish both Az and Alt axes before a mid spring, then I'll be ready to my first astro-trip. But where to go?

[JamesF]

Nice video   I loved the press for bending the bar.

James

[Vroobel]

My first experience of bending the alu bar was with the 2mm thick bar, I used hand made bender as well, but that one was simple, based on MDF board. It was very surprising and even funny  to see the bender bending itself! That's why I have had to make the stronger one. Of course, it has took some weeks to finish everything. 

[Peter Drew]

3 hours ago, JamesF said:

Nice video   I loved the press for bending the bar.

James

Me too!     ?

[Stub Mandrel]

Nicely done, I like the way you used various ball races to control the hobbing process. That should help reduce any errors, as will the huge diameter.

My own experience of hobbing is that it isn't always as accurate as I would like and the number of teeth is rarely what you calculate, unless you gash the blink first. The problem is the circumference at the top of the teeth (when you start cutting) is typically enough to fit an extra tooth in so you end up with more than you calculate based on the pitch circle diameter half way down the tooth form.

Did you aim for a specific number of teeth, or just set it running and accept what you got?

[Vroobel]

Thanks for this notice, that is what I concidered before I began anything. I was corresponding with lads from Polish astro-forum, I had hundreds question annoying them very much. I found out, that even running without calculations I will have quite good results and the difference will be spread between some teeth. But I am stubborn and set the size of wheel (taking into account the alu bar) exactly for 450mm. I checked firstly how many teeth can fit into 450mm, because it is size of my Lazy Suzan bearing. I bought 3/8" x 16 BSW tap, it's pitch is 1.588mm wide, so that was about 890 teeth. Of course, I have had to calculate back size of the wheel having 890 teeth, so the wheel is bit smaller. That's all.

[Stub Mandrel]

Not sure if I shoudl open this can of worms, but have you counted the actual number of teeth?

I ask as I have carefully calculated for 60 teeth and on two occasions got 61, which is not helpful when you are trying to make a rotary table...

[Davey-T]

Here's a guy with a big drive wheel 

Dave

 

[Saganite]

On 25/10/2018 at 16:14, Vroobel said:

Guys, did you ever think about really big worm gear? The one giving high precision, finally even better than 0.5 arcsec / micro step of your stepper motor?

I just make one of two worm gears for my Dobsonian mount. There is 450mm of diameter and 890 teeth. My stepper motors can work with 1.8 deg step and 8-32 micro steps. Let's assume I will set up the set as 200 steps x 8 micro steps x 890 teeth of worm gear, then I receive 1424000 micro steps per revolution, that means 0.91 deg / micro step. If I set 16 micro steps (that still gives quite good torque), I have 0.45 deg / micro step. I don't consider using 32 micro steps.

Please look at pictures below and also visit my YouTube channel to see the latest video describing how its made.

https://www.youtube.com/watch?v=u7_AuXA8eNs&t=10s

https://www.youtube.com/channel/UC9AMgmFpcYcRZuP65-SghPA

Amazing and brilliant, well done !

[tooth_dr]

Great video, enjoyed watching it, lots of talent 

[Vroobel]

1 hour ago, Stub Mandrel said:

Not sure if I shoudl open this can of worms, but have you counted the actual number of teeth?

I ask as I have carefully calculated for 60 teeth and on two occasions got 61, which is not helpful when you are trying to make a rotary table...

Interesting... I just counted, I have 896 teeth. That's surprising, I expected +/- 1, but not 6...

OK, let's count. 896 teeth x exactly 1.5877mm of pitch / Pi = 452.764mm, whilst average diameter of my wheel is 449mm. Where is the difference? 

Is possible, that I have wrong tap, not 3/8" BSW? I chect even this way: 449mm x Pi / 25.4mm x 16 = 888.55, so can't be 896!

Moreover, I've bought 1m of proper threaded rod that is matching perfectly to the grove in my wheel. 

Could you help to solve the problem, please?

[JamesF]

Might be useful to measure the circumference of the wheel, perhaps?

James

[Peter Drew]

I used to make gear sets professionally using similar methods. Although you can calculate the diameter needed for a tap to cut a given number of teeth,  minute variations in the pitch of the tap teeth can result in + or- a tooth or so, the larger the diameter, the greater the potential error. One way to guarantee correct numbers of teeth is to gash the gear blank mounted on a rotary table, the tap will then pick up the cuts accurately. The driving worm is the critical part of the set, although stock threaded rod is adequate for visual performance, it is not good enough for astrophotography of DSO's. This item is best turned from scratch on a lathe with the threads and the bearing journals all produced in the same setting, this ensures concentricity and improves periodic error.    ?

[Vroobel]

It's hard to dispute with mats. The circumstance is 1414mm, that gives 450.09mm of average diameter and should give even 890.71 teeth. So why gives 896 teeth? ?

[Vroobel]

14 minutes ago, Peter Drew said:

I used to make gear sets professionally using similar methods. Although you can calculate the diameter needed for a tap to cut a given number of teeth,  minute variations in the pitch of the tap teeth can result in + or- a tooth or so, the larger the diameter, the greater the potential error. One way to guarantee correct numbers of teeth is to gash the gear blank mounted on a rotary table, the tap will then pick up the cuts accurately. The driving worm is the critical part of the set, although stock threaded rod is adequate for visual performance, it is not good enough for astrophotography of DSO's. This item is best turned from scratch on a lathe with the threads and the bearing journals all produced in the same setting, this ensures concentricity and improves periodic error.    ?

That makes sense. So, it's quite possible to buy the tap with opposite error and have 6 teeth less then the expected number. 

OK, as I wrote, that's not a problem for my set, even with rod from stock. My RPi can deal with that, I just have to provide 896 instead of 890, that is even better. But I'm really worried about the second wheel, same tap, same machine - should give same number of teeth.