Harmonic Drives in Mounts

[avandonk]

Some of you may know about the Chronos Mount that uses Harmonic Drives.

Here is a video if you do not know what a Harmonic drive is.

http://www.youtube.com/watch?v=q1v7a0jsp1I


Here is an animation that is self explanatory 6MB

http://d1355990.i49.quadrahosting.com.au/2009_10/gearanimation.wmv

These harmonic drives have ZERO backlash!


Most if not all EQ mounts use gearboxes. My mount does not as it has belt drives to eliminate the gearbox. Image below. This has the effect of eliminating any micro jumps due to the gearbox.

I am seriously thinking of getting a dead or alive G11 or Titan and modifying it to incorporate these drives between the motors be they servo or stepper.

I forgot to say my EQ6 handles 90 lbs+ photographically as long as the wind is weak. See image two



Bert

[DavidU]

That would be a fabulous drive set. Don't you just love fine engineering.
Pezio or magnetic frequency drive?

[Bassnut]

Interesting. Do you find you have "cogging" with the toothed belt drive, or belt flex (or stretch induced oscillation on a guiding correction)?.

Ive seen these harmonic drives are used on (at least one) mounts. Cant tell if PE is eliminated though, what do you think?. A well tuned, slightly out of balance G11 doesnt suffer any backlash effects, would be similar for the EQ6 methinks. Is zero backlash worth the effort?, is that the only benifit for astro?.

[avandonk]

Fred I do not have to worry about any balance offset. There may be stretching but it is no where near 'jumping' I used to have with the standard gearbox.

PE is the biggest furphy as a measure of stability in our autoguided universe. It is random jumps that stuff up images. I got rid of gearboxes years ago on my HEQ5.

Is zero backlash worth it? You have got to be kidding!

I adjust the the backlash to almost zero on both DEC and RA on the EQ6's worms even now.

Bert

[Bassnut]

Bert. I agree, it could be said ppl worry about smooth PE too much, smooth PE can be guided out, yes,its random spikes that are a worry. I didnt bother with PEC after struggling trying to get rid of it, to no bad effect.

mmm, backlash, well, its not a problem with out of balance RA, for DEC, yes, but I cant say I had a problem with it with a standard set up with the worms tweaked a bit. Of course its better with no backlash, im just wondering how far its worth going to eliminate it, (given you have fixed that already).

[avandonk]

Fred I was more thinking about my next mount with more payload and they all have the same problem a gearbox between the motor and worm.

The G11 has a 25:1 gearbox between the servo and worm. You can have the best servo and encoder but it is meaningless if you have backlash in the gearbox and even worse random movements. The best worm will then also fail to perform to it's exquisite parameters.

Bert

[kinetic]

Bert would it rotate at a linear rate or non linear due to
the ellipse at its heart? :shrug:

Being a servo motor with optical feedback I suppose that
could be speed adjusted to make it a linear rate.

My homemade GEM has almost zero backlash now with
the worms meshed properly....press the track on/off
button and it simply starts tracking.

PE is another thing altogether...I have a slight PE curve
due primarily to the non-concentric worm.
PEC zeroes most of this out.
That doesn't mean I can take minutes long exposures
at 1500mm FL.
That will simply never be possible for me.
No amount of autoguiding or PE will prevent the stars
from ending up as blobs over minutes.

Bert, Fred,
Wouldn't belts contribute a similar trait to spur gears?
The only method with almost nil backlash would be a roller
drive method.

Interesting little devices that's for sure.
Steve

[avandonk]

Steve the Chronos mount is basically two harmonic drives and the PE is predictable and small.

Any stepper or servo driving an harmonic drive is far better than any gearbox alternative.

The belt drives leave the gearboxes for dead!

Bert

[kinetic]

Bert, from my limited knowledge wouldn't it be comparable to a
planetary gear reduction, only with a non linear eccentricity
built in?:shrug:
To my way of thinking, that's still a gear reduction, teeth meshing
with teeth.
And any such gearbox is only as accurate as the cut of the teeth.
Not taking anything away from the principle but if it was say a
100:1 harmonic reduction, then it should exhibit a similar PEC to
a 100:1 planetary or 100:1 worm and worm gear.
I would love to see some figures comparing these to spur gear trains.

Steve

[avandonk]

Here is a data sheet for the gearheads that would be suitable

http://www.harmonicdrive.de/cms/upload/pdf/GK2007/english/3_Gearboxes/PMG/PMG_complete_en.pdf

Transmission accuracy is less than 2 arcmin variation and repeatability <+ or - 1 arcmin.

If you then have a 180 tooth worm these tiny variations are then even smaller. That is about one arc second at most variation in DEC or RA.

The other advantage is that about 30% of the teeth are always meshing. There is no non linear effect due to the elliptical wave generator.


Bert

[Spanrz]

When I was researching some of this stuff, I found that it cost $$$$$$ and it wasn't stocked.
I found a company in Melbourne that could supply zero backlash drives (planetary even down to 1000:1), but you had to buy the drive with the motor, as the manufacturer set some sort of preload on the drive set.
And the cost wasn't cheap. I was told it would be weeks, maybe months before I even seen it in my hands since the date of purchase.
The harmonic system looks a good alternative to the standard gearing system.

The less moving parts, is the less chance for backlash.

I just wish I could use some parts (gear systems) from work, but a 20kg worm gear set wouldn't go to well, on the scope.

[marki]

Bert, just out of interest, have you tried kevlar belts on your pulley system?

Mark

[multiweb]

In my experience backlash has only been a problem for me in the past when calibrating with PHD. If there is too much of it and PHD is calibrating within the backlash then there can be troubles in guiding. Having said that if the mount is "loaded" both in RA and DEC, backlash has never been an issue while guiding. It also never impacted on guiding accuracy. That's what I have observed with my G11 over the past few years. As long as you're pushing the load and not reversing the servos direction while guiding you're good to go. Worm backlash can be eliminated by spring loading the meshing. The backlash inside the gearbox is very minimal also and wouldn't be a major problem I'd say compared to the worm backlash.

[Ian Robinson]

Bert I really think you need to tidy up that birds' nest of cables that are dangling everywhere.

[erosnicolau]

Hi, guys,

Actually, harmonic drives DO have backlash. Even the guys at HDS admit it in certain documents - should you look deep enough. "Zero backlash" is nothing more than a marketing gimmick. Furthermore, they tend to have more backlash in one direction than in the other (for instance: 0.67 arc-minutes CCW and 0.76 arc-minutes CW) - and this, coming from "the guys" themselves.

Now, to complete with my own 2 cents:
Passionate about day and night time-lapse, and trying to achieve a decent "zero" backlash without having to sell a kidney, I built a spin-off strain-wave generator reduction (similar principle, but eliminating the flex spline altogether) and achieved as far as 0.55 arc-minutes of backlash (tested with a laser beam mounted on the output shaft and pointing at a target 8 meters away and getting only 1mm of error). But that "miracle" only lasted for 2-3 hours of continuous running. After that, the gears (currently made of brass) ground down and widened enough to induce as far as 20 arc-minutes of backlash. To give you a better picture: I've done the calculations and every 3 microns of radial error induce - in my case - one arc-minute of backlash. That's good enough reason to have a new pair of gears currently made, out of 55 HRC hard steel, this time http://forums.reprap.org/mods/smileys/images/smilie3.gif Keeping fingers crossed here! http://forums.reprap.org/mods/smileys/images/smilie3.gif
(our current CNC only works with hundredths of millimeters, not microns - so for now I need to accept lower resolutions and hence larger backlash in the prototypes - but I'm hoping to find better machining solutions some time soon)

Anyway, here's my current prototype in action, grinding away on a lathe:

https://www.youtube.com/watch?v=1Clmg_CT6Dg