Topics

0.9A vs 1.7A stepper for EQ5


John Wagner
 

Hello Guys! I need some help because cannot understand one thing.

I went through a showcase of onstep upgrades for EQ5 type mount and did some search in the google and it seems that this stepper is a popular and recommended option to use:
Nema 17 Bipolar 0.9deg 36Ncm (51oz.in) 0.9A 5.4V

But I also noticed that some people use this option, which seems more reasonable:
Nema 17 Bipolar 0.9deg 44Ncm (62.3oz.in) 1.68A 2.8V

It has more torque at full A but if you would like you can run it at 40% power to be more efficient.

So the question is why 0.9A is more recommended, are there any benefits of using it over 1.7A variant?

Thank you!


However


Howard Dutton
 
Edited

The 0.9A motor (or similar) is tried and true.  Back in the "old days" it was a much better match for a DRV8825 vs. the 1.7A motor ("step stuck" problem.)  Hence that is what we used.

The 1.7A also suffers from a lack of availability of (step stick) drivers which can fully power it.  But I also came to the realization some time ago that about 40% power (1A) was often more optimal (slewing and tracking performance) than 71% (1.7A) or 100% (2.4A) so largely removing that concern even before the S109 and TMC5160 came along.

So now we have "tried and true" 0.9A that definitely works in a variety of examples vs. being hesitant to recommend a motor that's been used by few. 

Take your pick.


Dave Schwartz
 

The .9A one is recommended for the EQ5 because any of the supported StepStick-type drivers will support that current. If you get the 1.68A one you take a few options off the table and also have to deal with some serious waste heat (at full current).

The .9A has enough power for the EQ5. Most people think they need a lot of power but they don't. My favorite example is that the original motor for the 100 ton Hale telescope was 1/12hp (the size motor in a cheap utility pump). As long as the mount is balanced and in reasonable shape (open the locks and you can move the axes easily by hand) you don't need much power due to the high reduction ratios used on a telescope (power multiplies in the same ratio as reduction reduces speed). You don't need much "holding torque" either because its impossible to drive a worm gear set backwards.

With a Vmot of 12 to 24V that OnStep uses, you can (and should) run most steppers (the .9A included) at a current lower than rated (50% or less) with no reduction in performance, avoids heat dissipation issues, and lets you use smaller batteries/power supplies or have longer battery life. I run mine at .5A and it moves just fine.

Also, manufacturers rate the voltage of their motors with full duty cycle (i.e. non current-limited PWM or chopped) waveforms. All our recommended StepStick drivers implement current limiting which allows higher voltages to be used (which gives certain benefits) with the same overall power (and its related heat). Thus if you buy the 1.68A motor and then current-limit the driver under .9A to keep the heat under control, you've wasted your money.

On 2020-01-23 12:26 a.m., John Wagner wrote:
Hello Guys! I need some help because cannot understand one thing.

I went through a showcase of onstep upgrades for EQ5 type mount and did some search in the google and it seems that this stepper is a popular and recommended option to use:
Nema 17 Bipolar 0.9deg 36Ncm (51oz.in) 0.9A 5.4V

But I also noticed that some people use this option, which seems more reasonable:
Nema 17 Bipolar 0.9deg 44Ncm (62.3oz.in) 1.68A 2.8V

It has more torque at full A but if you would like you can run it at 40% power to be more efficient.

So the question is why 0.9A is more recommended, are there any benefits of using it over 1.7A variant?

Thank you!


However


John Wagner
 

Thank you so much guys for your clarification!

So basically Amperage of the motor doesn't affect tracking accuracy much but rather improve holding torque, which is already sufficient enough with a smaller rated motor?

Also, quick other question if taking this result into reference - https://onstep.groups.io/g/main/wiki/Stepper-Motor-Accuracy

What is the difference in results between Tracking Due to Load and Tracking Overall?


Khalid Baheyeldin
 

As Howard and Dave said: Use the 0.9A motor.
I add: Couple it with either the TMC2130 SPI version (silent), or the LV8729 (less heat dissipation, and allows faster slewing with a different microsteps slewing settings).


John Wagner
 
Edited

Yes, I think I will follow the advice.

Why I started to think about this in general, based on https://onstep.groups.io/g/main/wiki/Stepper-Motor-Accuracy  first option TMC2130 + 1.5A, 400 step, NEMA17 vs second option TMC2130 + 0.85A, 400 step, NEMA17 has tracking due to load value 0.05 arc-sec vs 0.13 arc-sec. So I assumed that tracking performance is 3 times better, giving the cost almost equal between two motors, option with a more powerful motor seemed more reasonable. 

Thank you guys!


Howard Dutton
 

On Thu, Jan 23, 2020 at 08:00 AM, John Wagner wrote:
Why I started to think about this in general, based on https://onstep.groups.io/g/main/wiki/Stepper-Motor-Accuracy  first option TMC2130 + 1.5A, 400 step, NEMA17 vs second option TMC2130 + 0.85A, 400 step, NEMA17 has tracking due to load value 0.05 arc-sec vs 0.13 arc-sec. So I assumed that tracking performance is 3 times better, giving the cost almost equal between two motors, option with a more powerful motor seemed more reasonable.
The figures quoted are for a 360:1 overall reduction ratio and are an estimate for my G11 mount.  Where a higher overall reduction is used the accuracy requirement is rapidly reduced (even 1.5x more is a big help, for example.)

Indeed the first motor 1.5A PKP246MD15 (I'm putting these on my G11) tested significantly better than the 1.7A OMC motor. The PKP246MD15 is larger (longer) and more expensive $$ vs. the OMC motors.  And in most cases (driver/motor/current) the 1.7A motor tested better than the 0.85A "OMC 0.9A like" motor.

The 1.7A OMC motor should provide the smoothest/fastest slew speeds followed by the 1.5A and last the 0.85A/0.9A (due to inductance.)

So don't get me wrong, if I were converting an EQ5 mount to goto I'd give the 1.7A OMC motors a shot given what I know at this point.  I don't feel the extra cost of the PKP246MD15 is justified given the EQ5's so-so quality though.


Howard Dutton
 

On Thu, Jan 23, 2020 at 09:12 AM, Howard Dutton wrote:
So don't get me wrong, if I were converting an EQ5 mount to goto I'd give the 1.7A OMC motors a shot given what I know at this point. 
I should add... vs. the 0.9A motors... not so much for tracking accuracy (doubt that matters) but for the smoother/faster slews and the possibility to run from a 12V supply instead of 24V.


John Wagner
 

Ok, Howard, thank you for your advice! I will give a try to 1.7A version, I am thinking of getting started with 3:1 ration for pulley (60T:20T) as a starting point as you recommended in other posts. It doesn't matter number of teeth for pulley as long as ration is the same?


Howard Dutton
 

On Thu, Jan 23, 2020 at 09:27 AM, John Wagner wrote:
Ok, Howard, thank you for your advice!
As I said, for me that's what I'd do, I don't mind taking a little chance to see if it works better.  And I know exactly how to source/setup a variety of stepper drivers.

I will give a try to 1.7A version, I am thinking of getting started with 3:1 ration for pulley (60T:20T) as a starting point as you recommended in other posts. It doesn't matter number of teeth for pulley as long as ration is the same?
60/20 = 3.


John Wagner
 

I meant 60/20 vs 48/16, the same gives 3:1 ration but was not sure if there is any difference in performance or just pick whenever easier to find.

Sorry, one more question, I tried to play with xlsx configurator and for all other being equal 5:1 ration gives 0.141 arc-sec resolution when 3:1 ration gives 0.234 arc-sec. Is there any downside going to higher ration like 5:1 versus 3:1? 

Thank you again! 


Khalid Baheyeldin
 

On Thu, Jan 23, 2020 at 01:27 PM, John Wagner wrote:
I meant 60/20 vs 48/16, the same gives 3:1 ration but was not sure if there is any difference in performance or just pick whenever easier to find.
No functional difference between them.
Pick whichever will fit your mount without obstruction or interference (e.g. the larger pulley can hit something in certain positions). That applies more to the 60T or 48T. The latter will have a smaller diameter and less likely to interfere.

In theory, the 20T pulley has more teeth 'engaged' than the 16T, but no one on this group has ever reported a slippage problem on the smaller pulley and having to remedy that by using the larger one.

Sorry, one more question, I tried to play with xlsx configurator and for all other being equal 5:1 ration gives 0.141 arc-sec resolution when 3:1 ration gives 0.234 arc-sec. Is there any downside going to higher ration like 5:1 versus 3:1? 
A higher ratio means finer resolution (as you found out).

But there are two potential issues:

1. More reduction also means larger pulleys on the worm gear, which may not fit.

2. It can also reduce the slewing rate. To get over this, use a different microsteps goto (e.g. 1/2 or 1/4) on a board that supports this feature (MaxPCB, MiniPCB or STM32) with a driver that supports it (LV8729 for 0.9A motor, or S109 for 1.7A).

In my case I has similar resolution (0.13"/step) and it works well, being able to take some images with 180 seconds unguided (but there are other factors at play in addition to the resolution). I can do close to 3 degrees per second slews with the STM32 and LV8729.

My parameters are:

200 step motor, 1/16 microstep, 18:1 reduction (geared motor), 180:1 worm wheel (Vixen SXD).


Joe
 

I’m using a Vixen GP with .9A 400 step motors with the 8729 drivers, and a 2:1 ratio (20+40) teeth). With close to 19 LBS on the mount I have no problems losing steps or overheating. The slew rate is reasonable at 8X. However any extra weight (tried using a DSLR)  at certain angles the motors drag and lose steps. I’d gather that at some angles pointing up the tube becomes off balanced because of the offset weight of the camera. I’m going to change to 3:1.
5:1 may be too slow with not much benefit for you.



On Thursday, January 23, 2020, 1:38 PM, Khalid Baheyeldin <kbahey@...> wrote:

On Thu, Jan 23, 2020 at 01:27 PM, John Wagner wrote:
I meant 60/20 vs 48/16, the same gives 3:1 ration but was not sure if there is any difference in performance or just pick whenever easier to find.
No functional difference between them.
Pick whichever will fit your mount without obstruction or interference (e.g. the larger pulley can hit something in certain positions). That applies more to the 60T or 48T. The latter will have a smaller diameter and less likely to interfere.

In theory, the 20T pulley has more teeth 'engaged' than the 16T, but no one on this group has ever reported a slippage problem on the smaller pulley and having to remedy that by using the larger one.

Sorry, one more question, I tried to play with xlsx configurator and for all other being equal 5:1 ration gives 0.141 arc-sec resolution when 3:1 ration gives 0.234 arc-sec. Is there any downside going to higher ration like 5:1 versus 3:1? 
A higher ratio means finer resolution (as you found out).

But there are two potential issues:

1. More reduction also means larger pulleys on the worm gear, which may not fit.

2. It can also reduce the slewing rate. To get over this, use a different microsteps goto (e.g. 1/2 or 1/4) on a board that supports this feature (MaxPCB, MiniPCB or STM32) with a driver that supports it (LV8729 for 0.9A motor, or S109 for 1.7A).

In my case I has similar resolution (0.13"/step) and it works well, being able to take some images with 180 seconds unguided (but there are other factors at play in addition to the resolution). I can do close to 3 degrees per second slews with the STM32 and LV8729.

My parameters are:

200 step motor, 1/16 microstep, 18:1 reduction (geared motor), 180:1 worm wheel (Vixen SXD).


Dave Schwartz
 

In my case, with my generic EQ-5, the 60T pulley would not have fit while having sufficient depth of the worm shaft in the pulley bore for stability.

I also have a 60T pulley for another project and, in the attached picture, I hold it against the 48T for comparison. The gap between the 48T pulley's flange and the RA casting (highlighted by the ellipse) is no more than 3mm and the radius of the 60T pulley's flange would have required more than that. Same thing on the DEC side.

As said, the 16T pulley on the motor has fewer teeth engaged with the belt than a 20T would but that's not an issue. Reinforced GT2 belts have no stretch so when you adjust the pulley separation properly something would have to go desperately wrong for the belt to skip.

On 2020-01-23 1:27 p.m., John Wagner wrote:
I meant 60/20 vs 48/16, the same gives 3:1 ration but was not sure if there is any difference in performance or just pick whenever easier to find.

Sorry, one more question, I tried to play with xlsx configurator and for all other being equal 5:1 ration gives 0.141 arc-sec resolution when 3:1 ration gives 0.234 arc-sec. Is there any downside going to higher ration like 5:1 versus 3:1?

Thank you again!


John Wagner
 

Thank you for the info! I think I will have the same issue with bigger pulley you had with not enough space, I probably will try to measure max available space for the bigger pulley and will have smaller pulley based from it. 

I appreciate your help guys!


John Wagner
 

Also, Dave, what is that yellow thing between stepper and mount? Is it a spacer or some kind of anti-vibration thing?


Dave Schwartz
 

Yes, it is an anti-vibration pad and washers. You can find the design on Thingiverse and I printed them using TPU.

On 2020-01-23 3:24 p.m., John Wagner wrote:
Also, Dave, what is that yellow thing between stepper and mount? Is it a spacer or some kind of anti-vibration thing?


John Wagner
 

Great! I will search for the mount and vibration pads on Thingiverse should fit on my mount perfectly.


Chris Vaughan
 

Dave what are you using for plugs and sockets for the stepper?


Mike Ahner
 

Hi John, if you find you really need/want a 60T pulley, you might try these:
http://shop.sdp-si.com/catalog/product/?id=A%206A51M060NF0606

They are "flangeless" and so have a little smaller overall radius. Of course, you still have to have 60T but it may fit your mounting. I tried them and they do fit on my Vixen GP2, but I want to reinstall the blue plastic cover over the RA and this had just a little too much width. So I ended up with 16T/48T which work really well. If you have access to a metal lathe, you could chuck up the flanged pulleys and remove the flange without much effort.

I also use a 1.7a stepper motor and without any real issues. I haven't turned mine down to 40% yet as Howard suggests, but even at higher current, it's smooth and quiet running on 12v.

-Mike