control many stepper motors (over 700)

i was wondering if anyone has run across a controller that can control 100's (over 700) of steppers using an arduino board.

Steppers, or servos (hobby or industrial, for that matter)? There's a big difference...

Ultimately, though, it doesn't matter - for that kind of project, you would want to set up each servo/stepper/whatever to have its own "intelligence" - basically, build a controller for the stepper/servo/whatever using a microcontroller that is mounted on or near the device being controlled. Each microcontroller would be given an "address" and you would need to communicate with each one over a networked bus of some sort.

I2C and SPI probably wouldn't work here - I don't think they have an addressable range large enough (unless you built custom "hubs" or "switches" that had their own addresses, and logically/physically broke the network up so that you sent a hub address and a device address in the packet). RS-485 might be a better option, or you might have to find some other bus standard out there, or roll your own (try to avoid that last one, if you can - its more difficult than you think).

You also haven't specified how close/far apart these motors, whatever they are, will be - this will also determine what network bus/protocol you use. Ultimately, though, in order to set this kind of thing up (even if you were controlling it with a full PC!), each device would need to be on a network of some sort. Its the best and smartest way to handle it, most likely.

One problem I can see with this scheme, though, is how to handle latency and whether or not the devices need to be in perfect or near-perfect synchronization; that question, though, is more in the realm of industrial controls and their management, and not something I have any direct experience in.


thanks for the response. to give you a little more info the steeper motors are NEMA 23 and will be placed about a 1 foot apart in a grid 12 x 36 (i down sized to 432)

Can you give more info on the steppers (coil ratings, current, model number)?

Regardless, they seem to (possibly) be a pretty hefty stepper (though the NEMA number generally applies to the mount hole spacing more than anything else - it does give an idea of what is being worked with).

Have you worked out a power supply system for that many large steppers? Furthermore, do you have the funding for this? I can already tell it won't be cheap. I am guessing (for stepper, power supply needs, networked controller, etc) about $100.00 per stepper (that is if you bought everything new, quantity discounts, etc; doing everything yourself and using surplus, you might get that down to about $25.00 per stepper - maybe). Only a guess, though...


You probably could get a stepper motor driver like below for each motor (if the motors are small), then multiplex the control signals to the motor controllers.

WOW - nice one - how are you intending to power this puppy? That's going to be a project in itself. :)

Edit: Sorry - cr0sh already asked this.

Ha! I am looking to do a very similar thing....although I am newbie when it comes to this stuff. I however will be using a grid 14 x 14 grid of tiny stepper motors to control the rotation of 2" wood cubes. The blocks will be about 1/2" apart.

As of right now, I will be using this motor or one similar

This is the basic operation that I am looking to make happen.

I know exactly how to do it, using the arduino & processing, but I would need many more pins to make it happen.

Is there a way to extend the amount of I/O pins for the arduino? Or a way of dealing with a collection of them? Or am I thinking about this all wrong.

any insight would be greatly appreciated!


I'd go for servos instead of stepper motors. As to doing a lot of individual control, you might want to search the board for "multiplex".

I understand going for servos for the effect in the video, however I wish to use a cube, where the four sides will be stopping points. It will not be as dynamic as the video example. The boxes will rotate to 90degree increments, say, every second. Would you still recommend servos?

Steppers take time to step, so timing and delays need to be considered. You may want to purchase 6 of the stepper motors and experiment. If they perform to meet your requirements, then you can pick up the other 190 stepper motors.

I would have thought that controlling 200+ stepping motors is a lot easer than controlling 200+ servos. This is because the servos have to constantly fed a PPM signal to keep them pointing where they are, where as a stepping motor only needs giving pulses when you want them to move.

However, this sounds like one of those “Art” projects where the person wanting to do something has absolutely no idea of the technical challenges involved.

Grumpy, you are exactly right! It is an art project, and I am just getting into this stuff. I am starting grad school in a month. I have a decent background in programming, but the whole electronics thing is new territory for me, and, for better or worse, is the direction that my work is heading.

I know that I have a lot to learn, but I feel that projects beyond your current comfort level, are the best learning experiences. I have already learned a great deal, and I am up to the technical challanges!

and Thanks zoomkat! I made my first multiplex, with LEDs, and it makes complete sense. My next step is to try and get it to work with a group of stepper motors.

Thanks for everybody's help and input!