Author Options:

Stepper Power Help??? Answered

I trying to design a cheap & easy (not ideal) bi polar stepper driver using the following components:
1. (1) A Bi-Polar motor rated at 12V 1A.
2. (4) 10A 250V Dual Throw, Single Pole Relays.
3. (1) Arduino UNO Microcontroller.
4. (1) ATX Power Supply.

I have the switching logic all correct from the UNO to the relays, and I have all of the wiring correct on the power side of the relay.
The problem comes down to shorting the PSU.My thought is the following, and I hope you gurus can either confirm, or correct my thought process (pretty please, and thank you).

Please see the Attached Pic for a basic schematic of what I'm doing.

12v = 1A * 12OHMS
12v * 1A = 12 Watts

Does this mean that somewhere within the power loop, I need a 12 Watt 12 OHM Resistor, to eliminate shorting out the PSU?
If this is true, what would be a practical solution other than using a big exposed spring, bar, etc... that would by its sheer mass resist all of this power?




6 years ago

I don't see what you are trying to do ? A relay based stepper driver ???

Whatever, an ATX supply would work, but only if you put ENOUGH load on the other rails. The 3V3 rail needs to see at least a couple of amps of load to let the rest of the supply work, so you need to add a resistor to allow that, and the 12V too, probably need to ALWAYS see at least an amp of load.

I can't help but think this "easy" stepper driver will cost you a lot of time and more money than a transistor solution, and your maximum frequency of driving the relays is going to limit the speed to less than 1 RPM.


Answer 6 years ago

Sorry if the diagram is too basic...

Along the bottom of the diagram, the first three components are the focus of the question.

The 1st component on the left is the PSU. Next to that is the first relay, and via the microcontroller, it is triggered to switch on turning this relay state into a +12V (that's what I'm attempting to represent via the red [should be yellow to reflect actual] horiz. wire), this in turn runs up the yellow leg of the stepper, through the coil out to the blue leg, through the "OFF" relay (-12V / GRND). The red/green combination will mimic the yellow/blue combo.

I wanted to clarify the diagram, but I think you may already understand the basics of what I'm attempting to do.

I'm not so much trying to make this in lieu of a solid-state solution, or even say its going to give me much performance, I know that there is going to be mechanical limits, and in turn timing/switching limits. I just want to prove to my self that I can take the relays that I already own (cheap for me anyway), and make my stepper move.

I do appreciate the help, and I know that you are largely trying to steer me away from making a bad choice in a stepper driver (thanks for looking out). All that being said, I'd still like to get this all working :)

So instead of struggling (I would because of lack of experience with ATX PSU) with applying load to the other rails, would it be easier to switch to a 12V 1A output From a 110VAC wall adapter? What kind of resistance should I be putting in place???

Thanks for the help!


Answer 6 years ago

There are stages in the stepper sequence when two windings are on, so you need at least a 2A supply. You won't need resistors.


Answer 6 years ago

Not if I don't do any half/quarter stepping
Just strictly a one step per pulse in the following sequence:

+ - - -
- + - -
- - + -
- - - +


Answer 6 years ago

Why not just replace the relays with suitable darlington driver transistors or a suitable FET?

I can see the relays operating fast enough to get any real speed from the stepper or lasting long as they will be operating many times a min.

AND will need a fair amount of power to run them.

This isn't a sensible idea.


6 years ago

Ignoring the question of whether or not your method is the best for operating a stepper motor and assuming that it is an exercise in education (tinkering).

I don't see any case where the power supply will be shorted out.  Unless I am not understanding your diagram, all the commons of the relays are connected to coils for load.  So, I don't see any way that  the power supply can be shorted.

Can you draw another schematic showing how it would be shorted?


Answer 6 years ago

I didn't see where it was logically going to short out either, but the ATX was failing to turn on when hooked up. when I unhooked the connection to this circuit, the PSU fired right up. This made me second guess if I actually needed some resistance based on the power formulas listed above.

I think I'm going to just try and find a wall adapter, from some old speakers or something.

Thanks for the reply

YES you hit the nail on the head, just having fun tinkering making stuff move without having to spend my time/money calculating power, resistors, transistors / Darlington arrays etc...

I am just trying to get this to work, no matter how sensible the method.
It's all about the simplicity of the stuff I have in my possession.


Answer 6 years ago

If anybody that really knows what they are doing & is willing to test this circuit on an ATX PSU, please confirm my results. I hooked nothing else up to the PSU just the four relays hooked to the stepper.

I googled how to short out the PSU without having to have it hooked up to a motherboard, and it all fired up as expected. I took a volt meter and tested that it was producing voltage before hooking it up to my circuit.

After hooking it up, the fan stopped, and I braved it out to check the voltage (which was nothing at that point) before turning it off. I unhooked it from the circuit, and all was working as it was before. I checked, double checked, and triple checked the wiring and all was good.

I even hooked up the small 5V power from the UNO to the relays in place of the ATX PSU, and got the motor to start moving.

However just getting it to move was not the complete desired result, as this had zero torque and stopped being able to move under the load of me pressing my fingers together against the motor shaft.

I proved the basic logic in the circuit is basically correct, it's just coming down to finding the right power balance that I'm really struggling with here.

Thanks for your help.

And for any who wish to just badger me about how bad my idea is, that doesn't really help! I'm not upset, but I just don't see any benefit at all to flood comments that are not relevant to my problem at all.

I'm not planning on changing my goal of getting this circuit, with my current components (less maybe the power source) up and running at the right rating for this motor.

Once I have a properly moving motor, then I can address migrating to transistors, as I do realize that relays are going to wear faster, not provide a very fast cycle speed, have loud clicking noises, not regulate current, etc...