Novoflex Castel Micro- stepper controlled

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:
mawyatt wrote:Ray,

I hope to be able to get good speed (>1mm/sec) rail performance and simultaneously achieve sub-micron resolution (~0.3um) with 400 step motors and 1mm pitch screw threads (as on the KR20) with very smooth rail performance with minimum vibration impact and no missed microsteps.

These are my goals for the precision stack & stitch setup I'm creating now. The new control algorithms and controllers should allow this I believe, at least that's what I'm working towards.

Best,
Well, 1mm/sec is what I consider impossibly slow. I move my rail +/-100mm fairly often and waiting nearly 2 min would drive me nuts. Guess I am impatient.
Ray,

If you need much faster than 1mm/sec for long runs and very smooth operation, then a dynamic change of Micro Stepping Code (MSC) would be the method to try. The idea would accelerate say with a MSC of 1/8, then change then MSC to 1/4, then 1/2 and finally to 1/1 which will take you to full velocity. You would do the same on the descending end but backwards; 1/1, then 1/2, then 1/4, and finally 1/8 as you coast into the final position. I think you might be able to achieve speeds approaching 5mm/sec with this approach and still have smooth operation without micro steps missed by carefully crafted algorithms and control.

With my experimental motor & controller setup (nowhere near the final config) I was just able to operate a NEMA 17 2.4 amp 400 step motor under dynamic micro step change mode at speeds that should yield ~ 5mm/sec operation. This was done using a direct hardware Pulse Width Modulated GPIO port write to the Step command input of the controller, while asynchronously controlling the three micro step inputs to dynamically alter the micro stepping "on the fly" with a random keyboard inputs for 1/1, 1/2, 1/4, 1/8 & 1/16 micro steps. I can't verify if any micro steps were missed tho, need additional hardware for this.

I'm actually able to get to 10mm/sec with a different control algorithm that utilizes the micro stepping capability and the direct step input command of the controller.

All these speeds assume a 1mm pitch screw & 400 step motor, with 2mm pitch this speed of course doubles and so on.

So maybe this does look promising.

Best,
Last edited by mawyatt on Sun Sep 30, 2018 11:36 am, edited 1 time in total.
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

Lou Jost
Posts: 5943
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Ray, couldn't you connect your camera to the rails (active and idler) via an Arca clamp, and then just move the camera up and down manually on that?

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote: Ray,

If you need much faster than 1mm/sec for long runs and very smooth operation, then a dynamic change of Micro Stepping Code (MSC) would be the method to try. The idea would accelerate say with a MSC of 1/8, then change then MSC to 1/4, then 1/2 and finally to 1/1 which will take you to full velocity. You would do the same on the descending end but backwards; 1/1, then 1/2, then 1/4, and finally 1/8 as you coast into the final position. I think you might be able to achieve speeds approaching 5mm/sec with this approach and still have smooth operation without micro steps missed by carefully crafted algorithms and control.

With my experimental motor & controller setup (nowhere near the final config) I was just able to operate a NEMA 17 2.4 amp 400 step motor under dynamic micro step change mode at speeds that should yield ~ 5mm/sec operation. This was done using a direct hardware Pulse Width Modulated GPIO port write to the Step command input of the controller, while asynchronously controlling the three micro step inputs to dynamically alter the micro stepping "on the fly" with a random keyboard inputs for 1/1, 1/2, 1/4, 1/8 & 1/16 micro steps. I can't verify if any micro steps were missed tho, need additional hardware for this.

I'm actually able to get to 10mm/sec with a different control algorithm that utilizes the micro stepping capability and the direct step input command of the controller.

So this does look promising.

Best,
Sounds very promising. This could be used with decent algorithmic control to get better speed, lower vibration, and better accuracy during stepping. It also would allow similar "smooth acceleration" from max microstepping up to full steps for larger (manual) moves.

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Lou Jost wrote:Ray, couldn't you connect your camera to the rails (active and idler) via an Arca clamp, and then just move the camera up and down manually on that?
I already do something like that now, but once magnification is set by extension, I prefer not moving either camera or lens standards. I'm using a bellows now, but have eliminated the focus rail to improve stability. The system is much more stable, but I must use the rail to move the system up/down once mag is set. I do have a small Z-stage integrated but it only has 10mm or so of travel.

But as I said I don't need super small step sizes, so a 200-step motor works fine. I'm using the mjkzz controller, which has an easy adjust for fast/medium/slow speeds. If I use the new XYZ software it also allows choice of microstepping as well as speed, so fairly flexible. That said, even with fastest speed and smallest (1/4) microstepping, a 1mm pitch and 400-step motor would be tediously slow.

Edited to add: when I modified the WeMacro vertical stand to add the idler and arca rail, I put together an interesting contraption that I may eventually decide to use elsewhere...ie I did not sell it with the stand. On that stand, I had installed a very long arca rail (400mm I believe) which spanned between mounts on the moving stepper rail and the idler rail. For bellows, I took advantage of the Vivitar bellows' individual 1/4" mounts on each standard, and installed a small arca clamp on each. I left the bellows' twin-rail clamps in place, and this allows the extension to be fixed while still allowing the entire assembly to be moved along the rail. Interestingly, either of the two clamps can be loosened, and that standard moved while the other standard stays put. I thought it was quite ingenious so kept it for future use. I suppose I should have told the buyer about it, but if he's reading, now he knows. Indeed it was a bit tricky to get the standards mounted such that the bellows' twin rails and the arca clamps were properly aligned, but once set up it worked brilliantly.

The problem with this arrangement, which is not easily solved with the Vivitar bellows, is limited max extension. For my current system, I can add a second bellows to double the available extension, making it possible to use lenses like the Multiphot 65mm and 120mm, the 105mm Inspec.x L, etc which require more extension than standard bellows allow.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

mawyatt wrote:
ray_parkhurst wrote:
mawyatt wrote:Ray,

I hope to be able to get good speed (>1mm/sec) rail performance and simultaneously achieve sub-micron resolution (~0.3um) with 400 step motors and 1mm pitch screw threads (as on the KR20) with very smooth rail performance with minimum vibration impact and no missed microsteps.

These are my goals for the precision stack & stitch setup I'm creating now. The new control algorithms and controllers should allow this I believe, at least that's what I'm working towards.

Best,
Well, 1mm/sec is what I consider impossibly slow. I move my rail +/-100mm fairly often and waiting nearly 2 min would drive me nuts. Guess I am impatient.
Ray,

If you need much faster than 1mm/sec for long runs and very smooth operation, then a dynamic change of Micro Stepping Code (MSC) would be the method to try. The idea would accelerate say with a MSC of 1/8, then change then MSC to 1/4, then 1/2 and finally to 1/1 which will take you to full velocity. You would do the same on the descending end but backwards; 1/1, then 1/2, then 1/4, and finally 1/8 as you coast into the final position. I think you might be able to achieve speeds approaching 5mm/sec with this approach and still have smooth operation without micro steps missed by carefully crafted algorithms and control.

With my experimental motor & controller setup (nowhere near the final config) I was just able to operate a NEMA 17 2.4 amp 400 step motor under dynamic micro step change mode at speeds that should yield ~ 5mm/sec operation. This was done using a direct hardware Pulse Width Modulated GPIO port write to the Step command input of the controller, while asynchronously controlling the three micro step inputs to dynamically alter the micro stepping "on the fly" with a random keyboard inputs for 1/1, 1/2, 1/4, 1/8 & 1/16 micro steps. I can't verify if any micro steps were missed tho, need additional hardware for this.

I'm actually able to get to 10mm/sec with a different control algorithm that utilizes the micro stepping capability and the direct step input command of the controller.

All these speeds assume a 1mm pitch screw & 400 step motor, with 2mm pitch this speed of course doubles and so on.

So maybe this does look promising.

Best,
Just ran a quick test of a NEMA 17 Wandai 400 Step motor model # 42BYGHM810. This is a popular motor with 0.9 degree steps, 2.4V @ 2.4A with 1.8mH inductance and 4800 gcm holding torque. I'm using a 12V supply and a Pololu A4988 Driver and connected to a Raspberry Pi 3B. I've written my own code for testing and this uses a Raspberry PWM Hardware Pin to connect to the A4988 Driver. I was able to get up to 26 revolutions per second (1560rpm) in a 1/1 micro step mode before the motor locked up and quit turning. I'm sure this will be much slower when attached to a focus rail, but that's much faster than I thought the motor/controller could support.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote:
mawyatt wrote:
ray_parkhurst wrote:
mawyatt wrote:Ray,

I hope to be able to get good speed (>1mm/sec) rail performance and simultaneously achieve sub-micron resolution (~0.3um) with 400 step motors and 1mm pitch screw threads (as on the KR20) with very smooth rail performance with minimum vibration impact and no missed microsteps.

These are my goals for the precision stack & stitch setup I'm creating now. The new control algorithms and controllers should allow this I believe, at least that's what I'm working towards.

Best,
Well, 1mm/sec is what I consider impossibly slow. I move my rail +/-100mm fairly often and waiting nearly 2 min would drive me nuts. Guess I am impatient.
Ray,

If you need much faster than 1mm/sec for long runs and very smooth operation, then a dynamic change of Micro Stepping Code (MSC) would be the method to try. The idea would accelerate say with a MSC of 1/8, then change then MSC to 1/4, then 1/2 and finally to 1/1 which will take you to full velocity. You would do the same on the descending end but backwards; 1/1, then 1/2, then 1/4, and finally 1/8 as you coast into the final position. I think you might be able to achieve speeds approaching 5mm/sec with this approach and still have smooth operation without micro steps missed by carefully crafted algorithms and control.

With my experimental motor & controller setup (nowhere near the final config) I was just able to operate a NEMA 17 2.4 amp 400 step motor under dynamic micro step change mode at speeds that should yield ~ 5mm/sec operation. This was done using a direct hardware Pulse Width Modulated GPIO port write to the Step command input of the controller, while asynchronously controlling the three micro step inputs to dynamically alter the micro stepping "on the fly" with a random keyboard inputs for 1/1, 1/2, 1/4, 1/8 & 1/16 micro steps. I can't verify if any micro steps were missed tho, need additional hardware for this.

I'm actually able to get to 10mm/sec with a different control algorithm that utilizes the micro stepping capability and the direct step input command of the controller.

All these speeds assume a 1mm pitch screw & 400 step motor, with 2mm pitch this speed of course doubles and so on.

So maybe this does look promising.

Best,
Just ran a quick test of a NEMA 17 Wandai 400 Step motor model # 42BYGHM810. This is a popular motor with 0.9 degree steps, 2.4V @ 2.4A with 1.8mH inductance and 4800 gcm holding torque. I'm using a 12V supply and a Pololu A4988 Driver and connected to a Raspberry Pi 3B. I've written my own code for testing and this uses a Raspberry PWM Hardware Pin to connect to the A4988 Driver. I was able to get up to 26 revolutions per second (1560rpm) in a 1/1 micro step mode before the motor locked up and quit turning. I'm sure this will be much slower when attached to a focus rail, but that's much faster than I thought the motor/controller could support.

Best,
Wow, that's not too shabby. I think that's about 10kHz, correct? Assuming my calculator is giving me the right numbers today.

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

mawyatt wrote:without micro steps missed
Mike, can you clarify your concern with "micro steps missed"?

I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.

I certainly see how things could get messed up if say you tell the controller to change from 1/16 to 1/4 mode when it's not in a state that can be reached in 1/4 mode, or if you somehow managed to change its mode at the very same time you were commanding it to step. Either of those could result in the controller being internally in a state different from what the rest of the system imagines.

But I can't tell whether that's your concern, or something else that I haven't thought of.

Can you clarify?

--Rik

elf
Posts: 1416
Joined: Sun Nov 18, 2007 12:10 pm

Post by elf »

mawyatt wrote: I hope to be able to get good speed (>1mm/sec) rail performance and simultaneously achieve sub-micron resolution (~0.3um) with 400 step motors and 1mm pitch screw threads (as on the KR20) with very smooth rail performance with minimum vibration impact and no missed microsteps.
I don't think that should be too difficult to achieve. I was able to get <1um steps with 6mmX1mm stainless allthread. The key was spending several tedious hours lapping it with a 75mm long lap. The zero backlash nut was made of Acetal, heat and pressure formed around the screw.

p.s. A Teensy 3.5 (Arduino compatible) is capable of driving steppers at 300,000 (micro)steps per second.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

rjlittlefield wrote:
mawyatt wrote:without micro steps missed
Mike, can you clarify your concern with "micro steps missed"?

I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.

I certainly see how things could get messed up if say you tell the controller to change from 1/16 to 1/4 mode when it's not in a state that can be reached in 1/4 mode, or if you somehow managed to change its mode at the very same time you were commanding it to step. Either of those could result in the controller being internally in a state different from what the rest of the system imagines.

But I can't tell whether that's your concern, or something else that I haven't thought of.

Can you clarify?

--Rik
Rik,

Sure. Some of this is for others since I know you have a good knowledge of stepper motors and controllers having written the useful Zerene interface for the nice Stackshot rail system.

First almost all stepper based systems are open loop without feedback, so position is "assumed" not verified. Missing full steps is common with some controller & motors, especially if the acceleration and inertia is high. This can be prevented by proper motor size and details and carefully controlled acceleration and velocity profiles. Motor internal details become important here with static magnetic fields and the field created by the winding coil current.

The motor has an internal electrical time constant, which is the inductance times the resistance (L*R), the exciting current follows a classic exponential profile (exp^t/tau) relationship when under voltage control, and generally needs to have at least 4 constants to reach final value (exp^-4 ~ 2%). Since the coil induced magnetic field is directly proportional to the coil current, short pulses of current where the current can't get close to final value will also reduce the induced magnetic field and eventually not allow the rotor to step to the new position, thus a missed step. What happens on the outside of the motor also influences the ability to make a step, high friction and/or inertial being examples.

A simple example of some of these principles can be experienced by just running a motor without anything attached to the rotor. As you increase the step rate the motor speeds up and eventually you reach an electrical step rate where the coil current can't create enough magnetic field to "pull" the rotor to the next step, so the rotor moves forward some then flips back to the original position and the process repeats and the rotor doesn't actually take another step, so appears to be "locked". Now if you reduce the step rate slightly the motor will still remain "locked", even though you can get the motor to operate at this reduced step rate if you slowly approach from a much lower step rate. Why, because the rotor has inertia and this allows the rotor to "jump" to the new step as it's already heading in that direction, but without this inertia the rotor must accelerate and make the step from standstill which is obviously more difficult than when already "running" in the right direction. This explains some of the weird behavior of stepper motors in various setups and control schemes we've experienced.

Some of the newer controllers allow the motor operation from a much higher voltage, as example the motor I mentioned is a 2.4V motor with 1 ohm resistance, 2.4 Amp coil current. I'm operating the motor with 12V which would normally burn up the motor, controller or power supply because of excessive current (5amps!!). What's happening here is the controller is pulsing the motor coils using a Pulse Width Modulation (PWM) concept that is faster than the motor time constant, so the coil current doesn't achieve final value. However, the average coil current can still achieve the desired level the motor requires because a higher voltage is used to "force" the coil current, abet during a shorter interval, which in effect creates a current controlled concept rather than voltage controlled (remember the discussions long ago on the advantages of current control over voltage control!). Since motors are fundamentally a current (magnetic field is current induced, not voltage) device this control scheme is more desirable. The beauty of this technique is twofold, first the higher supply voltage allows a higher initial torque vs. time and potentially better motor performance, and second the clever use of the H bridge to allow this coil current to return to the supply during the "discharge" time period (rather than waste it) dramatically reduces the overall system power consumption.

The PWM technique along with the H bridge also allows the coil currents to be controlled in such a way that fractional amounts of average coil current and direction can be controlled. By controlling the relative coil current levels and direction partial steps can be achieved but at reduced holding torque. Much info is available online, here's couple sites for more details.

https://www.orientalmotor.com/stepper-m ... rview.html

https://www.youtube.com/watch?v=Ew6eVGnj7r0

Since the motor position is only inferred by the number of, direction and type of pulses to the coil windings it's possible for the system to misrepresent the motor position. Counters are generally employed to count the number of pulses, these can be hardware or software. The popular IC stepper motor driver chips don't employ any internal counting capability (although some recent types do), so the motor position algorithms must rely on other means. Depending on how the counter is configured and how the counts are acquired and used it's possible for the system to compute a position that is based upon the total number of counts which could include missed steps, either full or micro steps. For example the step controller directs the driver to make a step, but the driver doesn't recognize the step command until it's finished with some other task (changing direction, or micro-step mode, or motor current control and so on), and when it's ready to execute the step another step comes in, so only one step is performed but two were issued. If the counter didn't recognize this missed step the count would be off by 1 step. There are many reasons why a step could be missed or not achieved by the motor, from mechanical hardware (motor/system resonances, intermittent torque, load dynamics, changing speed, acceleration, direction), to the controller computer interface, glitches, and even design bugs in the controller/driver system.

The folks in digital communications deal with these bit errors all the time and use various techniques from simple CRC to complex Viterbi/R-S encoding, but the stepper motors have no way to verify the right code has been performed without some sort of feedback since Forward Error Correction doesn't seem possible in this application.

Hope this helps,

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

rjlittlefield wrote:
mawyatt wrote:without micro steps missed
Mike, can you clarify your concern with "micro steps missed"?

I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

Yes as mentioned many things can cause the "missed step".

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.

However most popular driver chips simply represent these micro-steps to the stepper controller as steps, and the effect is just a step being executed by the motor, either a full step or micro-step. Micro-steps are more prone to being missed simply because the motor torque is lower, much lower in higher micro-steps (follows the sine(90degrees/microstep number) I recall).

I certainly see how things could get messed up if say you tell the controller to change from 1/16 to 1/4 mode when it's not in a state that can be reached in 1/4 mode, or if you somehow managed to change its mode at the very same time you were commanding it to step. Either of those could result in the controller being internally in a state different from what the rest of the system imagines.

Yes, highly possible I believe. We probably don't generally know this because we don't generally require a return to a precise position. If we miss a step now and then it's only a couple microns off and probably not going to affect a stack with 20 micron steps. If we do notice this it will likely be blamed on the rail!!

But I can't tell whether that's your concern, or something else that I haven't thought of.

I'm concerned about being able to have somewhat precise & repeatable positioning for a stack and stitch system I'm developing. Must admit I didn't realize how sophisticated these stepper motors, controllers and drivers have become. The current mode use, micro-stepping capability and the ingenious current recycling.....clever stuff indeed.

Can you clarify?

--Rik
Rik,

More specific comments to your questions above.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

elf wrote:
mawyatt wrote: I hope to be able to get good speed (>1mm/sec) rail performance and simultaneously achieve sub-micron resolution (~0.3um) with 400 step motors and 1mm pitch screw threads (as on the KR20) with very smooth rail performance with minimum vibration impact and no missed microsteps.
I don't think that should be too difficult to achieve. I was able to get <1um steps with 6mmX1mm stainless allthread. The key was spending several tedious hours lapping it with a 75mm long lap. The zero backlash nut was made of Acetal, heat and pressure formed around the screw.

p.s. A Teensy 3.5 (Arduino compatible) is capable of driving steppers at 300,000 (micro)steps per second.
Elf,

The motor I mentioned can support 1mm/sec and sub-micron positioning with a 1mm pitch screw thread I believe. I'm more worried about the ability to return to the same position, or any position with micron level precision. Assuming the temperature is constant, then the motor, rail and controller should be the limiting factor. The ability to not miss any steps, full or micro, over an entire stack and stitch session (thousands of images) is the desired result, assuming the rail (THK KR20 or KR26 with 1mm pitch screw threads) is stable and good enough.

I'm not very good at coding, but OK with the hardware side of things. Think I've got the hardware figured out but now struggling with the coding, but learning as I go so that's part of the fun!! I really envy you folks that can code all these wonderful micros available today, limitless possibilities!!

Lapping the screw threads sounds like a really good idea if you have the equipment. I don't know what type of screw threads & bearings that THK uses, but their rails are very smooth and repeatable.

I don't think 300,000 steps/sec would be very useful with motors that have millisecond time constants, and (assuming I did this correctly) 300,000 micro steps would yield a micro step torque of 0.00000524 times the full step torque...certainly not useful!!

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

Adalbert
Posts: 2426
Joined: Mon Nov 30, 2015 1:09 pm

Post by Adalbert »

Hello everyone,
The technical parameters are common to my rail but the box is much better and I like this fantastic blue knob :-)
My rail doesn’t have any box :-(
http://www.photomacrography.net/forum/v ... highlight=
BR, ADi

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

rjlittlefield wrote: I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.
...
It's true that microsteps are not mechanically-quantized like full steps are, but indeed folks use them to attempt sub-step resolutions. In general I don't rely on anything smaller than quarter-steps, since the torque curve for even smaller-n steps drops rapidly. If loading and detent torque comes anywhere near the drive torque, those smaller steps can potentially be "missed". One trick to ensuring a bit more accuracy is to increase the base drive current, but this of course generates more heat and can become a problem.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:
rjlittlefield wrote: I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.
...
It's true that microsteps are not mechanically-quantized like full steps are, but indeed folks use them to attempt sub-step resolutions. In general I don't rely on anything smaller than quarter-steps, since the torque curve for even smaller-n steps drops rapidly. If loading and detent torque comes anywhere near the drive torque, those smaller steps can potentially be "missed". One trick to ensuring a bit more accuracy is to increase the base drive current, but this of course generates more heat and can become a problem.
Ray,

1/4 or 1/8 steps are about my limit as well, since torque falls off to 38% and 20% respectively.

With these drivers that utilize the supply recirculating current scheme, they don't get as hot as expected. Would have never though a small chip package without heat sink could deliver 1.5~2A coil currents from 12V to a 2.4V, 1 ohm motor without overheating. The multilayered PCB uses a integrated heat spreader in the PCB, but no external heat sink.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3413
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote:
ray_parkhurst wrote:
rjlittlefield wrote: I think it's clear what missing a full step means. That's when you think the motor is on tooth N when it's really on N-1, for example because it stalled out with not enough torque or trying to step it too fast.

On the other hand, microsteps are really just part of the command protocol of the controller. They don't have a digital presence on the mechanical side.
...
It's true that microsteps are not mechanically-quantized like full steps are, but indeed folks use them to attempt sub-step resolutions. In general I don't rely on anything smaller than quarter-steps, since the torque curve for even smaller-n steps drops rapidly. If loading and detent torque comes anywhere near the drive torque, those smaller steps can potentially be "missed". One trick to ensuring a bit more accuracy is to increase the base drive current, but this of course generates more heat and can become a problem.
Ray,

1/4 or 1/8 steps are about my limit as well, since torque falls off to 38% and 20% respectively.

With these drivers that utilize the supply recirculating current scheme, they don't get as hot as expected. Would have never though a small chip package without heat sink could deliver 1.5~2A coil currents from 12V to a 2.4V, 1 ohm motor without overheating. The multilayered PCB uses a integrated heat spreader in the PCB, but no external heat sink.

Best,
Mike...I was more concerned with the motor overheating. To get enough torque for 1/16 or especially 1/32 you may need to use more than rated current.

Post Reply Previous topicNext topic