I am trying to build a stepper motor driver, based on the A4989 chip, to control NEMA 17 and NEMA 23 stepper motors.

Stepper motor specifications:
NEMA 17: 0.4 Nm, 1.5 A/phase, 2.1 ohms.
NEMA 23: 1 Nm, 2.4 A/phase, 1.2 ohms.

The circuit will be:from datasheet

I made some calculations and found the requirements of this circuit as follows:
For NEMA 17: U = 3.15 V/phase and power is P = 4.725 W.
For NEMA 23: U = 2.88 V/phase and power P = 6.9 W.

The external MOSFETs are IRFZ44N

I have some questions:

  1. Firstly, for choosing proper PSU, if I am half-stepping or micro-stepping, both windings will be energized so the power will be multiplying x2, meaning the current drawn is x2. Is that correct?

  2. If I power the dual H-bridge from 12 V or above (as per the datasheet) with a resistor in series with the drain, will my stepper motor still work and how will it be affected? Also, this chip is specified for 30..500 W steppers so should I dissipate power through a resistor?

  3. Is the MOSFET suitable for such a small coil current, compared to its ID(max)?

The MOSFET has a much higher max. drain voltage than needed but the current is important and I can limit it with Rsense (ITRIP(max) = VREF/(8×RSENSE)), can't I?

Another thought on this was if I could power the chip with 12..50 V, as per the datasheet, but power the H-bridge from a lower voltage.


1 Answer 1


Let's take these in order.

  1. While both coils are energized in microstepping mode, they are not both fully energized. This means that you are not driving the full 1.5 or 2.4 A per phase into both phases at the same time. Microstepping will divide that up so that only a portion goes to each phase. So, for 1/16th of a full step, 1/16 of the current will go to phase A and the other 15/16 will go to phase B. This means you don't need a power supply that will handle 2x the rated current.

  2. Yes, you can put a resistor in series. It won't affect very much, particularly since a current sense resistor is typically a very small number of Ohms. The driver chip and motor windings are current controlled devices, the voltage only matters in that a higher voltage can energize the windings faster and thus allow a faster transition from one step to the next. There's usually a curve provided by the motor manufacturer which lists the maximum speed relative to supplied voltage. Just make sure that your resistor can handle however much power its dissipating. I^2*R and all that.

  3. Those MOSFETS look fine. Using a relatively "low" current won't hurt anything.

Finally, there is no reason to power the Allegro driver with one voltage and the H-bridge with another. Just use a 12V supply that can deliver at least 5A and you're good.

  • \$\begingroup\$ i understood how it will work in microstepping and thanks for that, but in halfstep mode ? And you say about my supply that needs to deliver 5A at least, why 5A ? Because for Nema 17 i need only 1.5/phase, wouldn;t hurt my motor with such a high current ? ( For Nema 23 i know that i can drive it ) \$\endgroup\$ Commented Sep 29, 2017 at 10:02
  • \$\begingroup\$ Ok, you mentioned the minimum specs of PSU, so that it will work well, isn't it ? \$\endgroup\$ Commented Sep 29, 2017 at 10:10
  • \$\begingroup\$ Having more current available is not a problem. I was thinking of the NEMA 23 you mentioned at 2.4 A/phase and by having a 5A supply then that would be enough for either motor. The rated current is the maximum available, that's not how much it will be driving out all the time. Allowing extra head room on the power supply will keep it cooler and running more efficiently. \$\endgroup\$ Commented Sep 30, 2017 at 22:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.