A4988 driver maximum stepper speed?

Other Hardware Motors, Mechanics, Power and CNC
1

I would like to hear what is the maximum stepper motor speed that people have actually achieved using an A4988 stepper driver and what motor they were using.

I have some of these motors and with a 24v power supply the max I seem able to get is about 1200 full steps per second or 360rpm.

I realize that motors with low resistance coils would probably work faster at the same voltage but they may take too much current for the A4988 driver and I am interested to find out what is possible with it.

...R

2

No point trying with high-impedance motor, the inductance of the windings is far too
large (for low voltage operation at any rate).

Stepper max speed depends on degrees-per-step, supply voltage, winding impedance.

You might get some good results with the A4988 driving a 15 deg/step printer motor
of a few ohms from 30V supply, but you'll lose the resolution...

For really high performance you need a proper stepper driver with discrete
switch MOSFETs and capable of 120V+ supply. Once you get to this sort
of regime steppers actually start to become efficient since most of the electrical
power is going to overcome back-EMF rather than just energising the windings.

I think the future of small position-control systems will be BLDCs with integrated
encoder+controllers - technology like magnetic encoder chips will allow a single
PCB solution to a built-in motor controller. Expect the prices to drop as these
start to displace hybrid steppers from the market in coming years.

3

Thanks Mark,

I know there are better options both for motors and drivers. I bought the A4988 and the motors I have as the best that would fit my budget. At the moment I don't wish to invest in better hardware.

I had been hoping that someone who has experience of operating motors at higher speed with an A4988 would share their experience including the type of motors and other relevant information.

With so many views and no other replies I'm inclined to assume that nobody else has motors controlled by A4988s that are operating faster than mine. Which is interesting in itself.

I might ask the same question on the RepRap Forum.

...R

4

A lot of reprap's are using the 1.7 ohm NEMA17's these days, but Repraps don't require
high speed as they are belt drive for XY. The Z performance is the only place it matters,
and then only for startup really.

Its lead-screw applications like CNC milling machines where motor speed is crucial -
something like a little PCB mill would be the sort of thing.

5

I am seing the same limitations with the A4988 driver here too. My nema17 motor is supposed to run up to 4500rpm according to the manufacturer.

MarkT - Could the H-bridge SN754410 be the answer here? H-Bridge Motor Driver 1A - COM-00315 - SparkFun Electronics

As it is shown on this article here: http://arduino.cc/en/Tutorial/MotorKnob

6

You do realize that steppers lose torque the faster they go and 4500rpm is probably a marketing thing? 4500rpm at what torque?

No, a straight non-chopped H bridge is the opposite of what you want.

7

No doubt this is acheived without any load. Now they claim doing 4500rpm using a particular controller. I am just wondering on the design of that controller.
The motor I am talking about is here NEMA-17 Bipolar 5.18:1 Planetary Gearbox Stepper - 3317_1 at Phidgets

8

I see 900rpm on that page. And that is divided down by the gearbox.

9

I don't have any concerns about the competence of the A4988 to produce fast steps if you have a motor that can run very fast at the current and voltage limits of the A4988.

It was suggested elsewhere that the motor might run faster using microstepping and (of course) an appropriately higher step rate. I haven't tried that yet.

...R

10

Yes, microstepping can overcome resonance problems. So can weights or rubber disks.

11

polymorph:
You do realize that steppers lose torque the faster they go and 4500rpm is probably a marketing thing? 4500rpm at what torque?

A lot.

A low-impedance stepper driven from a 200V chopper drive will go like the
clappers if the winding insulation is up to it.

The lack of torque is caused by slow current slewing and/or back-EMF, both of which
are improved proportionately with higher voltage supply and lower inductance windings.

Or in equation form:

dI/dt = V/L (current change in inductor)

V = Blv (back EMF depends on speed).

12

Wow. Seems like a lot of work. Large CNC mills use brushless DC servomotors to avoid the stepper problems.

13

of course there is a limit. about 25 times the motor rated voltage. alas, the real world has a way of raining on the parade.

the iron loss will burn up the motor if you try to go too far. double the voltage and get 4 times the heat.

if you need power at high speed, don't bother with steppers. get a servo with encoder. they are designed for higher speeds.

also, the motor detent torque effects available power. that is why the largest steppers you will ever see are NEMA-34

e-bay, search for NEMA-23 over 1,500 hits.
search for NEMA-32, 3 hits.

here is some great information, the guy who makes those units is a genius in regards to understanding motors.

14

polymorph:
You do realize that steppers lose torque the faster they go and 4500rpm is probably a marketing thing? 4500rpm at what torque?

No, a straight non-chopped H bridge is the opposite of what you want.

you will loose half the torque as you double the speed. high speed steppers are not a good application.

the other problem is sending the steps to the motor correctly. a very clean signal is required. as you attepmet to make steppers spin faster, you need to make sure the power to the motor clean. choppy pulses will cause the motor to miss steps.

as note about micro-stepping, when you get into the low speeds, the micro-stepping helps the motor move much smoother. if you are limited by output frequency, cut your micro-stepping back to 4 or maybe 8.

15

polymorph:
Wow. Seems like a lot of work. Large CNC mills use brushless DC servomotors to avoid the stepper problems.

Well usually now they are PMSM (permanent magnet synchronous motors),
but the difference between that an a BLDC isn't much, basicall less torque ripple
as the drive is close to sinusoidal.