Jump to content

NLCbanner2024.jpg.2478be509670e60c2d6efd04834b8b47.jpg

Why do we not use Belt Drives?

Earl

Recommended Posts

brianb

brianb

Posted
Posted

Stretch & slippage. Simply not accurate enough for the job. A worm drive can be pretty accurate, it's a really simple thing to make and adjustment is by no means difficult.

The only obvious advance on worm drive is direct drive - there are mounts using that principle but the engineering issues involved in making the technique work properly seem not to be that easy to solve.

Link to comment
Share on other sites
Totnesdave

Totnesdave

Posted
Posted

Hmmmm, well, I'm not an expert (although I did spend a year or two maintaining and restoring historic clock movements) ...there's going to be a lot of gearing down necessary in a small space to achieve the slow speeds for 1x tracking, and I would imagine a clock mechanism to be the most compact, plus with belts there's slippage, perishing, expansion/contraction issues, basically a less solid mechanical coupling, operating at slow speeds I would expect more 'give' in the mechanism for belts compared to a well made solid brass movement. The heavy brass gears will also have their own inertia for a nice smooth drive, I would imagine belts having a 'rebound' effect at any hiccup in the power. I know my DK-3 is silky smooth, I've had to resist the temptation to take it apart and peek inside to see how it's made :)

Just my two pennorth :icon_scratch:

Dave

Link to comment
Share on other sites
wayne weedon

wayne weedon

Posted
Posted

I have some experience with modern toothed belt drives as used in my cnc machinery. Very high accuracy stuff too.

You can compensate for just about anything if the control loop is closed. From the short time I have owned a telescope and been reading on the subject it's been interesting to note that generally the vast majority of motorised mounts are totally open loop using steppers.

Now a stepper motor whether you microstep or not can only accurately resolve to the nearest 1/2 step (generally 0.9 degrees on the motor shaft as most stepper motors are 200 step/rev) Any microsteps may or may not be linearly spaced between those 1/2 steps, all depends on the quality of the motor/drive combination.

Servo motors on the other hand as used in all but one of my cnc machines generally make use of incremental or sometimes absolute encoders to track following error. My cnc lathe for example uses 5000 count quadrature encoders so thats 20,000 divisions of each motor rev. and thats on the other end of the belt not on the motor, so any pitch error in the belt/pulleys is compensated for.

Belt stretch? Most quality belts are steel or kevlar reinforced it's a non issue even with the high comparative loads my machines deal with.

Just my 2p's worth.

Wayne....

Link to comment
Share on other sites
brianb

brianb

Posted
Posted
I have some experience with modern toothed belt drives as used in my cnc machinery. Very high accuracy stuff too.

But you're not trying to move kilograms of weight to an accuracy of a small fraction of a second of arc.

Link to comment
Share on other sites
callump

callump

Posted
Posted

There is quite a lot of discussion about belt drives and direct drive motors in the Alt Az Initiative being led by Russ Genet - interesting stuff.

I suspect that direct drive motors will be the way of the future, if the costs could come down.

/callump

Link to comment
Share on other sites
wayne weedon

wayne weedon

Posted
Posted

I should dig out some old gear from an old project and give a servo and toothed belt a try just as a RA drive on our mount. I could fairly easily knock up a bit of microcontroller code to devise a basic RA drive. Anything further than that is beyond my capabilities I think.

I need to do something as I bought a Celestron RA drive (Chinese manufacture) It broke within an hour and had to be fixed (Slipping Brass Gear) and once fixed found it just caused too much vibration for video use and also was just short on max speed. We have a NES Mount? and maybe is has a lower worm ratio than the Celestron mount the drive was designed for.

Wayne....

Link to comment
Share on other sites
wayne weedon

wayne weedon

Posted
Posted
But you're not trying to move kilograms of weight to an accuracy of a small fraction of a second of arc.

No I'm not. I'm moving several hundred kilos's of cast iron at 10 metres/min and ending up within 10 microns of desired position when rapiding or moving much slower and holding tighter tolerances when cutting.

Wayne...

Edit : I should add of course we are not just using toothed belts but rather driving through very high quality Ball screws. BUT I think these systems have some merit over what appears to me coming in from the outside so to speak as very crude existing systems. It would be interesting to see what the larger class telescopes are using. I would imagine a fully closed loop system.

Link to comment
Share on other sites
ollypenrice

ollypenrice

Posted
Posted

Gears do have backlash but an early sixties Zeiss coude bought by a former club member had lovely gears split into two halves and spring loaded in opposite directions to reduce or eliminate backlash in the mesh. This does seem a good idea.

Intuitively I agree with Brian on belt drives. A little shock absorbing elasticity seems to be their big attraction where they are used, which is not at all what's wanted here. But I'm not an expert.

Direct drive is so attractive but like many things in engineering its day has to come. That means that the ancilliary requirements have to be met in a real world context. (Ducati could make desmodromic valve trains that could rev to 17,500 in the late fifties but no one could make points that would would exceed 15,000. Examples are legion).

Olly

Link to comment
Share on other sites
wayne weedon

wayne weedon

Posted
Posted
Gears do have backlash but an early sixties Zeiss coude bought by a former club member had lovely gears split into two halves and spring loaded in opposite directions to reduce or eliminate backlash in the mesh. This does seem a good idea.

Olly Antibacklash gears as you describe are off the shelf items, you can also get the same in worm wheels.

Toothed belts of the short lengths we would be talking about are really not that stretchy believe me! But in my case I would propose still using the existing worm drive in my mount initially for my RA drive requirement. It's the easiest way to to get more reduction. You can't influence the input of a worm drive that much via knocks.

Ah the desmo's! Love the ducati bevels, but I'm more a moto morini man at heart.

Wayne....

Link to comment
Share on other sites
brantuk

brantuk

Posted
Posted

Keele University have a 24" reflector in their observatory. It's a really old one - huge weight (steel) and stands about 20+ft high. I was amazed that it's driven in altitude by a small motor from an old line printer with a belt drive. I guess it must be nicely balanced - they used another similar motor to move the dome around lol :)

Link to comment
Share on other sites
Astro_Baby

Astro_Baby

Posted
Posted

I think I have seen an upgrade for Meade LXD75 mounts that replaces the gears with belt drives.

Years ago I was involved with a specialised 'screen' for blind people and the system there used a belt based drive system which was VERY accurate - it had to be. The belt was made out of a special acetate type material and was very hard wearing and under control of a servo system could produce very fine movement.

Link to comment
Share on other sites
Starflyer

Starflyer

Posted
Posted

There's several people in the EQ6 Yahoo group that are using belt drives and reporting improved tracking accuracy. EQMOD will allow you to use different gearing ratios and this can have a dramatic effect in reducing certain periodic errors.

As far as I know none are using a closed loop system, just a stepper, original or upgraded, with accurate pulleys and good quality belts.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • 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.