How to calibrate Pololu A4983 drivers

I am having trouble with my motors skipping steps and am trying to dial in the correct current on my Pololu A4983 drivers.... im basically using trial and error right now, but is there a good step by step way to correctly set the current adjusting pot on the drivers?
Thanks,
-Drew

Edited 1 time(s). Last edit at 08/17/2011 03:00PM by daufhammer.

If your electronic powers your motors when idle, could try the heat method as described here.
Otherwise a multimeter plugged in serial to one motor wire - also described on that wiki - can help : turn down the pot until the motor won't have enough current to move then back up a little to give 1.25 to 1.5 times the current you measured.
Also the sound helps - motors cry high or low when current is not appropriate.

Measure the voltage on the wiper of the pot you are adjusting and set it to 0.4V. That will give 1A, which is the max the Polulu will do without a heatsink, It seems to be enough to run all the axes of a Prusa, it might be a bit low for an extruder.

---

[www.hydraraptor.blogspot.com]

Thanks Nophead and DeuxVis i will give all those methods a try when I get a chance and let you know how it goes.

Nophead, how did you come to 0.4V being a good value... is that just what you have found to work well in your experience or did you calculate that somehow?

I adjusted the pot until the chips where just too hot to touch and then I measured it and found it was about 0.4V, so 1A. On my Mendel I also used about 1A on the X & Y motors, so it seems like a good starting point to try to avoid a heatsink or fan.

---

[www.hydraraptor.blogspot.com]

Thanks Nophead, im still having trouble because even at about 0.35V (measured between the little open probe point by the Pololu Pot and Ground) my X motor is too hot to touch.... I turned it down a bit and it seemed to help no more missed steps yet.

I should mention that my printer is a RepStrap built almost entirely (aside from the Adruino mega & Pololu drivers) from junk parts i already had lying around.... So the stepper motors aren't standard....

I did just swap in a nice new stepper motor on the X axis that i bought for the Prusa Mendel i'm planning to build.... that is the motor that got too hot too hot to touch even 0.35V

Im trying to figure out what the best underlying principles are to dial in the correct amount of current for each motor... because I would think that 0.4V may be good on some motors (like yours), but may be too high or low for someone else's motors. Is that right? or would 0.4v and 1A have the same effect no matter what your motor specs are?

No 1A is about right for the standard high torque NEMA17s we use. They are rated at 1.65A, two coils on, 80C rise. With the current is set to 1A, when both coils are on the driver gives 0.7A each. The temperature rise in the motor will be 80C * (0.7/1.65)^2 = 14C, so the motors only get warm in ambient conditions. When I run them inside my chamber at 40C, they get to 65C, too hot to touch!

The torque will be only 0.44 * 0.7/1.65 = 0.19 N.m.

Stepper motors run at their maximum spec will get to 100C, which is way too hot for plastic brackets, so we use high torque motors and under run them. The particular motor we use also has the correct voltage / current ratio to hit the sweet spot of the Pololu driver.

With the motor you are using it seems motor temperature is the limiting factor rather than the Pololu. In which case simply reduce the current until the motor does not get too hot to melt the plastic brackets. If you don't have enough torque to move the axes then the motor is not suitable. If you have a data sheet for the motor you can do the same calculations above to see what torque you get at a particular temperature.

---

[www.hydraraptor.blogspot.com]

For reference, my drivers give out 0.4 to 0.5 A per phase (measured) and everything works fine.

The temperature stays quite low on the motors, never above 50°c - I didn't have a look at chips temperature yet, will do.

Thanks guys, that makes sense... I can't find the datasheet, but the motors for my Prusa that i just swapped onto the x axis of my RepStrap are....

Japan Servo Co
KH42JM2B140E
DC 5.16V 1.2A 1.8degree/step

The guy i bought them from said they were about 33.5 oz/in when wired bi-polar

Hopefully they will work.... one cool thing is that they have a really strong rubber ring and a seperate mounting plate, so that should insulate the hot motors from the plastic and keep it from melting. It also quiets the motors and reduces vibrations.


Edit: I got them locally for $5 each, so it was a good deal and not that much if they don't end up working.

P.S. I think i finally got things dialed in, steppers are pretty warm to the touch, but not scalding hot, not missing anymore steps (so far anyways)

Edited 2 time(s). Last edit at 08/18/2011 02:53PM by daufhammer.

Did the KH42JM2B140E motors work afterall?

Yeah they did end up working. They may be a tad under powered, but they got the job done on my RepStrap. I successfully printed all the Prusa parts with them. Now i only have one of them on my Prusa, it's on the geared extruder and seems to have plenty of power for that. I swapped up to some higher torque steppers for the rest of my Prusa though. The KH42JM2B140E's worked fine, but i just wanted a little extra assurance of not missing steps so i went with some steppers with a higher torque.

Thanks nophead my "X" motor works it's little butt off and get's a bit hot on long prints it's set at 0.44 as i shake to much to set it exact
but i notice that the little bitty pointer on the potentiometer is pointing the same way as all the others they are also close on 0.4
the e motor gets hot too... hell its working hard also but gets free cooling moving back and forth thru the air so i glued heat sinks to it
and the x motor too
it's not hot enough to melt any thing but cooler is better no? (Kysan 1124090) s in my case
going cheep on motors will bite one in the butt in the long run .You get what you pay for most times.

As long as they don't melt the brackets it is not a problem running motors hot. They are rated for about 100C. I.e. the current rating is for 80C rise. We run them well below that to prevent the brackets melting. My extruder motors run too hot to touch when enclosed in a chamber at 45C, they are 60-70C, about the limit for ABS.

The resistance increases when they got hot so the performance drops with constant voltage drive,. but with constant current and a supply voltage much bigger than the motor voltage it makes little difference.

---

[www.hydraraptor.blogspot.com]

With Pololus, the sensing resistors are Rs=0.05 ohm, and per the A4988 datasheet, the calculation for the maximum trip current I_TripMax= Vref/(8*Rs), so Vref of 0.4 should produce a maximum current of 0.4/(8*0.05)=1A

I'm aiming for max 0.7A on my 1A rated steppers, so I rearrange it as:

Vref = I_TripMax * 8 * Rs or

Vref = 0.7 * 8 * 0.05 = 0.280V

With a measured Vref of 0.273V, I should expect 0.6825A, and I measured the current through one coil as 0.486A in full step mode, which should be 0.7071 of the full trip current, or I_TripMax= 0.486/0.7071= 0.687A, which seems close enough.

For stepsticks, the Rs=0.2 ohm, so you'd get 4 times the Vref at the same current.

A related question to calibrating Pololu drivers (A4988).

I have assembled a Sanguinololu board. Should I use a separate breadboard to calibrate each of my stepper drivers? Or can I check each one placed in its position on the existing board which should be able to provide the right grounding and power to the pins.

No you don't need a separate bread board. You calibrate them in situ on the Sanguinololu by measuring the voltage on the Vref test point.

---

[www.hydraraptor.blogspot.com]

I don't see a VREF test pad on my A4983 drivers. Does anyone know how to find the proper place to measure the voltage?

---

3D Printer: Reprap Original Mendel with RAMPS 1.4 electronics, j-head MK V-BV 0.4MM extruder, Wade's geared extruder, setup for 1.75MM filament

EmailGuy look here at Jose Manual Ortiz's blog .
Go down about half way and you will see a labeled close-up of a stepper driver board. It will have the locations labeled for you.

---

__________________________________________________________________________
"I like to be, what I like to see, in you and me.
I am the Stallion, Mang."

ISTAR Scope Club

Or just measure it on the potentiometer wiper as that is a bigger target to probe.

---

[www.hydraraptor.blogspot.com]

Quote
nophead
Or just measure it on the potentiometer wiper as that is a bigger target to probe.

I've always found that easiest. I attach a alligator clip test lead to a jeweler's screwdriver and can easily adjust it as I'm reading the value.

A4983 boards can be insurted two ways by 180 degrees is this critical or not

Edited 1 time(s). Last edit at 02/11/2015 09:29PM by midnight.

Extremely critical.

You have 12v and 5v on that device. Put it in backwards and you can put 12v onto the 5v rail. destroying the driver, the mega and if your really unlucky your computer.

hi davex
In full step mode, the current through the coils is limited to 70% of the current limit. how to include this in calculation..

Edited 1 time(s). Last edit at 02/13/2016 06:14AM by rizulthkr.