Completely new idea for telescope mount propulsion

[vlaiv]

I'm working on a 3d printed star tracker, and have been exploring different ways to get speed reduction.

In doing so, I came across very interesting video:

https://www.youtube.com/watch?v=2SUiwQVWe8w

While drive itself proved to be fail for the purpose it was built - it gave me very interesting idea.

One of major problems with motorized telescope mounts is how to get required reduction in output speed. Speed reduction goes from several hundred up to few thousands to one - in order to get smooth tracking. Another very big issue is that mechanical means of doing so, due to manufacturing errors introduce two kind of unwanted behavior.

First is periodic error and second is backlash. While first is not something we want - it is not as bad as second - at least for imaging as smooth periodic error can be guided out - provided that the mount mechanics is responsive enough. That means minimizing second error - or backlash.

Well, above idea solves two issues - large reduction ratios / step precision and removes backlash.

It works a bit like differential in a car - which allows for drive wheels to turn at different rate while still being coupled to engine.

In above setup - two step motors will be used to drive one mount axis instead of one stepper motor - but they will be used in such configuration that their speeds subtract in order to get resulting speed.

If we for example want to have 15"/s sidereal speed - we don't need to drive our motors at that speed - we can simply drive one at 1015"/s and other at 1000"/s in other direction and their difference will give us wanted speed.

This also means that we don't need to have very large reduction of our motors RPMs in order to get very slow output speed.

Both motors will be constantly running and turning gears, gears will always be engaged - so there won't be backlash as no gear will change direction (not entirely sure on this one - will need to test it) - they will simply slow down or speed up - depending on what is wanted on the output.

[andrew s]

Hi @vlaiv, interesting idea. Have you worked out the effective torque?  That might be an issue for slewing. Regards Andrew 

Edited March 12, 2023 by andrew s

[vlaiv]

11 minutes ago, andrew s said:

Hi @vlaiv, interesting idea. Have you worked out the effective torque?  That might be an issue for slewing. Regards Andrew 

I haven't but from that video it seems that it will give lower of the two in "addition" mode.

I think that it actually might have issue with holding torque rather than with slewing. Steppers have lower torque with higher speeds - so holding torque is greater than torque at higher RPM.

If both steppers are running at same but high RPM - we will have effectively "holding torque" as mount will be at rest - this is interesting in case of DEC and resistance to wind for example, it will be as high as either of two motors at that RPM - so can be lower than holding torque of either motor.

[AstroKeith]

Interesting concept.

But as a drive designer I'm thinking about the speed accuracy. I can drive a stepper (via gears) at 15.00 quite easily (to say 0.01% accuracy), but if I drive both the 1015 and 1000 speeds at 0.01% and subtract. the 15 could be as much as 1.5% in error.

I'm also trying to think though the torque issues. We're almost in direct drive territory!

[vlaiv]

1 hour ago, AstroKeith said:

Interesting concept.

But as a drive designer I'm thinking about the speed accuracy. I can drive a stepper (via gears) at 15.00 quite easily (to say 0.01% accuracy), but if I drive both the 1015 and 1000 speeds at 0.01% and subtract. the 15 could be as much as 1.5% in error.

I'm also trying to think though the torque issues. We're almost in direct drive territory!

When you say accuracy - what exactly do you mean?

Timing issues or step / micro step accuracy?

[Sunshine]

Looking at the CAD design of his drive and watching all the parts involved I can't help but marvel at its complexity. It seems very intricate and I wonder how many points of failure such a complex design introduces and how it would handle wear. Regardless of the unsuccessful result, I think this kid should be working for NASA.

Edited March 12, 2023 by Sunshine

[globular]

2 hours ago, AstroKeith said:

Interesting concept.

But as a drive designer I'm thinking about the speed accuracy. I can drive a stepper (via gears) at 15.00 quite easily (to say 0.01% accuracy), but if I drive both the 1015 and 1000 speeds at 0.01% and subtract. the 15 could be as much as 1.5% in error.

I'm also trying to think though the torque issues. We're almost in direct drive territory!

Don't you just run them both at about 1000 so that the output doesn't move - it doesn't matter what the actual speeds are, as long as the output is stationary.
You then control the output by speeding up or slowing down one of the motors (which you can do to .01% accuracy) - and the output will move by the same amount (to the same accuracy).

[vlaiv]

1 hour ago, Sunshine said:

Looking at the CAD design of his drive and watching all the parts involved I can't help but marvel at its complexity. It seems very intricate and I wonder how many points of failure such a complex design introduces and how it would handle wear. Regardless of the unsuccessful result, I think this kid should be working for NASA.

He has more videos worth watching. Has some other interesting designs too.

There is a community of makes on youtube that mostly design reduction boxes for robotic purposes - but one can learn a lot from their videos on this topic.

[markse68]

Does it actually get rid of backlash or double it?

Mark

[vlaiv]

3 minutes ago, markse68 said:

Does it actually get rid of backlash or double it?

Mark

I think it acts a bit like "east heavy" setup.

Motors and gears always move in same direction, so there is always load on them and backlash, although present because of gears - is always "eaten up" - or gears are always in contact on one side - in the direction of motion.

I'm just not sure if that is true on connection of annulus ring and planets though.

It's a bit like relative velocity thing. If I'm moving in one direction and you are moving in the same direction but faster than me - it will look from my point of view as if you were approaching me and I'm standing still (sort of). Similarly - if I'm faster than you  - it will look like you are moving away from me although we are moving in the same direction.

Now, I'm not really sure what it means when gears are involved - if some of the gears actually start to spin in opposite direction with respect to some other gears - and if they mesh - there could still be some backlash.

[vlaiv]

10 minutes ago, markse68 said:

Does it actually get rid of backlash or double it?

Mark

I don't think there will be backlash.

I'll try to make a diagram to explain things.

this system depends on planetary gears - it can't work with ordinary gears.

In planetary gear system - there is sun gear (central small one), planet gears (3 same gears) and outer gear. We usually use sun as input and planets as outputs (their centers are "rigid" and rotate so when connected they form output "body" that we attach shaft to). During this time outer gear / annulus gear is stationary.

What this system does is to rotate that annulus gear instead of keeping it stationary. We have 3 choices:

1. Move it in same direction as rotation of planets - or up arrow

2. leave it stationary

3. move it in opposite direction

What happens here is a bit like using escalator or those person conveyor belts / moving sidewalks - we will walk faster if "ground" is moving in the same direction - then we move at speed that is our speed + "ground", but if we walk in opposite direction and we match the speed of moving sidewalk - to someone looking from a side we would appear not to move.

Ok, so as we now have analogy with person and a sidewalk, let's introduce backlash analogy. Effects of backlash happen when we walk and we want to change direction. At some point we need to stop and turn - that is our backlash. If we need to turn in order to change direction of motion - then we will have backlash.

Let's do our escalator / moving sidewalk example again.

Mount is standing still - in our case that is me going to the right at some speed and moving sidewalk going the opposite direction at the same speed. For the rest of the world I appear to stand still.

If I want to move to the right for the rest of the world - all I need to do is to keep walking as I did and we reduce moving sidewalk speed in the opposite direction - net speed will be to the right. Did I turn to do that? No.

We want to reverse the mount, what do we need to do? We need to increase moving sidewalk speed to above my walking speed - and I will start moving in opposite direction to the rest of the world (but I'll keep constant speed to myself - I'll just pace at the same pace all this time).

Notice that - neither me nor sidewalk changed direction of motion whole time - we never stood still and turned around - no backlash.

But this analysis pointed something very important and that is resolution.

I don't think that I understand fully impact on resolution if one is using stepper motors. My gut feeling says that we won't have improvement in resolution.

Here is what I'm thinking - if we have minimum step size - then our position can only change in these steps - regardless of how many steps we do per unit time. It's a bit like that old saying - one step forward, two step back. No matter how many steps we perform forward and backward - total movement will be integer number of base steps as we subtract the two.

You can't move half a step back if you move integer steps forward and then another integer steps backward.

[Mr H in Yorkshire]

All very interesting but the gizmo has gears and where there are gears there will be backlash. If I follow your reasoning above, you won't reverse but merely change relative rotational speed, as a change occurs is that not where the backlash will appear, as the pressure on the gears changes even if in the same overall direction?

[vlaiv]

15 minutes ago, Mr H in Yorkshire said:

If I follow your reasoning above, you won't reverse but merely change relative rotational speed, as a change occurs is that not where the backlash will appear, as the pressure on the gears changes even if in the same overall direction?

You might be right - I'm having tough time visualizing it.

I don't think it is as simple as regular gears - it probably depends on relative motion of components prior to change and the way you change them in order to achieve wanted result.

Here is an example

Say that two gears - annulus and planetary have same speed and that mount is still - it does not move.

You can initiate mount movement in two different ways - you can either increase sun (and hence planet) speed - or you can decrease annulus speed. Difference will change in the same way and mount will move the same. Question is - will backlash work the same in both cases?

[markse68]

It’s hard to visualise isn’t it. I think you have to build one Vlaiv  

Mark

[vlaiv]

@Mr H in Yorkshire and @markse68

I think that you are right.

I've spent some time visualizing what is going on ("riding the gear" in order to change perspective) - and yep, you are right - it won't deal with backlash.

Here is how we can see that.

If we have situation that two motors are rotating in opposite direction and output shaft is standing still - that can happen even if there is zero engagement between planet and annulus. We can imagine that backlash is equally spread in forward and backward direction.

look at above gif - this animation shows situation when the mount is standing still (we don't have planetary carrier show in the image - but it is output shaft).

Now look at annulus gear and any of planets.

Planets are driven by sun and annulus is driven by other motor (or gear assembly - irrelevant - we can just use motor with different speed setting) - so each is driven independently - they don't even need to be in contact and can still have same tangential velocity.

In any case - backlash can be "split" between forward and backward direction like this:

whatever we do next slow or speed up outer gear or slow down or speed up planets (via sun) - it will need to clear some of that backlash - either side of it.

Ok, so we came to conclusion that this drive is useless for mounts - it won't deal with backlash and it wont increase stepper resolution

 

 

[globular]

I was drafting a post making this same argument.  I'm so glad you beat me to it - your post is better than mine was.

One point I was going to make that you didn't - backlash is actually worse in this drive because it can occur with a change in speed, not just a change in direction.

 

[Elp]

Don't know if this will give you some ideas:

https://www.machinedesign.com/mechanical-motion-systems/article/21831618/methods-to-minimize-gear-backlash

 

[AstroKeith]

14 hours ago, vlaiv said:

When you say accuracy - what exactly do you mean?

Timing issues or step / micro step accuracy?

I was thinking of how to generate an accurate tracking speed of 15 arcsec/sec. As long as one counts steps driven on both motors then positional accuracy can always be calculated. Thus at the end of a slew, you will know where you are, but that is 'after the fact'.

In generating a tracking velocity it needs to be accurate in 'real time'. I think now this can be done by checking every second or so how far the compound movement has gone, and applying any small correction needed.

14 hours ago, globular said:

Don't you just run them both at about 1000 so that the output doesn't move - it doesn't matter what the actual speeds are, as long as the output is stationary.
You then control the output by speeding up or slowing down one of the motors (which you can do to .01% accuracy) - and the output will move by the same amount (to the same accuracy).

I havent thought of an away of actually doing this in practice. Ye two motors could be run at 1000Hz from the same frequency generator, but the moment you change one of the speeds, you have broken the link between them. You now need two frequency generators. If I just use a clock and drive the steppers at say 1000 and 1015 steps in the same clock period, it would seem to work. But to do that the stepper drivers need to generate the frequencies, so we are back to the same problem.

As in my reply above to vlaiv, it will work by driving the motors as best you can, but use a single clock to check actual compound movement over a second or so. Then apply the correction. The deviation between corrections has only a second or so to build up and so will be in the milli-arc sec range.

I use a similar technique in my 'normal' dobsonian stepper drive. At any one time the true  Az & Alt tracking speeds require fractions of a micro step. This cant be done, so I drive at the nearest whole number of micro steps. But I keep a tally of how far ahead or behind this puts me with respect to perfect speed. After a few seconds, I can add or subtract a whole microstep and get closer to perfect. I keep the tally going until the scope is moved.