Let's Make Robots!

OctoDriver 8A Dual Motor Controller

Vendor's Description: 

I've been working on a doggy door for a while - while the software is straightforward, finding a suitable motor controller has been more difficult.

The problem is:
  • A lot of the inexpensive motor controllers are too meager to control anything but the most dinky motors. ladyada's Arduino shield only does 600mA continuous, 1.1A peak.
  • The beefier controllers either:
    • are really expensive (Pololu's DMC 01 can do 13 amps, but it's $100), or
    • only control a single motor, or
    • have complicated control schemes

So I've been on the lookout for a medium power, inexpensive, and easy to control circuit - I was flipping through starlino's website and I spotted exactly what I was looking for. He calls the circuit the 'OctoDriver', it combines 2 h-bridges to provide 8 amps peak, 4 amps continuous to each motor and with 4 chips, you can control 2 motors. I asked him if I could put it on a PCB, and he thought it was a great idea. I couldn't think of a better name, so I call my version "The OctoDriver".


You can find out more, pick it up, download files, etc, on the project page. I also did a pretty detailed howto. And like all the other Prop Platform Modules I've done, layout files are available under the MIT license.

What do you think?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Thanks OddBot! My real gripe is with the $20 or less 'motor controllers' that were primarily designed to hit a price point instead of actually work. I'm sure many beginners who order the $20 Arduino shield are pretty dissapointed & turned off when they figure out what they've bought.

I hope the OctoDriver offers builders an inexpensive option that will perform and let you do something cool with a powerful motor.


This is Ryan from Pololu.

Have you actually tested this board at 4 A continuous with different chip pairings? There isn't a good reason to suspect that these h-bridges can be paralleled, and if they can be paralleled, it's likely that one bridge will be working a lot harder than the other one, so it's not as simple as multiplying by 4 to get the continuous current. Also, with all outputs low, the board is going to burn over 300 mA without the motor doing anything.




Pololu sells a $40 dual MC33926 motor driver carrier (the top picture on the right) that can do few amps. It also is way more compact, works with different logic voltages, has reverse protection, and has current sense outputs.

I don't understand why you are comparing this OctoDriver to the Pololu TReX dual motor controller DMC01. The TReX takes serial (RS-232 or TTL) or RC inputs, it can mix the RC signals, and it has an additional unidirectional auxiliary channel. If you wanted to have a more direct comparison to the claimed specs you might look at the motor driver category of Pololu's website. We sell the dual VNH3SP30 motor driver carrier MD03A that is only $10 more than the OctoDriver. It is much more compact and has nicer motor driver chips on it (9 A continuous, 30 A peak per channel).

I0J1191.200.jpgf you really want a beefy motor controller with a serial interface, we sell the Pololu qik 2s12v10 dual serial motor controller (13 A continuous, 30 A peak per channel) for $75. The qik (the bottom picture on the right) is controlled with RS-232 or TTL serial and has additional features like automatic baud rate detection, a demo mode, acceleration control, and current control and feedback.

 - Ryan

Firstly - I'm a big fan of your products. You guys have been making some great, high quality motor controllers. Some your products are expensive but they're quality designs with a lot of extras. It is difficult to service or repair surface mount designs, but if saving space is the top priority, it's the only way to go.

Also, thanks for asking - yes, I actually have tested it on several chip pairs with multiple loads and they've performed within the design limits specified.

But I am confused when you say there's no reason to think these H-bridges can be parallelized.  If you haven't tested running these in parallel and you have concerns about doing so, why does your website suggest this?





Ha, that does sound a little contradictory. When we mention that the SN754410 can be stacked, we do say you can get "almost" 2A. It's a little kludgey and we wouldn't do it in a product. The more you stack, the more kludgey it gets. We might recommend duct taping something, but you probably wouldn't like it you bought something and found it duct taped together. (We do have a piece of electrical tape on the bottom of the 3pi, though.)

The SN754410 and similar chips are kind of nice, but they really have a lot of drawbacks compared to more modern components. The idle current draw is a good example.

We're not using surface mount parts just to keep the boards small. It's also so we can take advantage of more modern parts. Parts like the MC33926 do not come in a through-hole package, but they are the kind of parts that get used in demanding applications like cars.