Let's Make Robots!

Stackable SN754410 Motor Driver Boards, 1 Amp per board, 2 Motors per Stack


This board is what everyone needs! It can operate as a single Motor Controller, Or Stack them for increased power! You can stack as many boards as you want on top of each other, and the headers simply connect to each other. Requires 6 Lines (or 4 if you tie the Enable lines high) for control over all sides of the H bridge. This allows High side and Low Side Braking, as well as coast. Every stack of drivers controls 2 motors, and for more amperage just stack the boards on top of each other. Every board can handle one amp independently, and has adequate traces for that power. Since the amperage will be flowing through the doubled up headers, It Should be able to handle up to 12 amps (if every pin header is rated at 3 amps). This means you can control very beefy motors with your surplus 1A SN754410's! This board is available for sale now, shoot me a PM and you will recieve a quote for the price of the board. I plan to have a stock of these available soon, and I hope to have them ready within a couple weeks!

Comment viewing options

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

This is a great idea Krumlink but as usual I have some questions about the trace width. Have you done any calculations on this. 4A is quite a lot of current running through those thin traces.

Thank you so much for that link! I have updated the board to utilize the proper thickness traces (40mil) so it can handle the full 4 Amps. Usually what I do Andersen is just layout the board so that all the connections are connected. Then I redo the board, learning from the previous layout and making the traces as thick as they will fit on the 2nd revision.

That sounds like a good approach.
When you do your layout, do you start with a schematic and then do the layout via rats nest so that Eagle can validate the layout, or do you just start routing?
I think this board will get very hot if run under full load. You probably should try to get as much surface area connected to the GND center pins of the SN754410 as possible to disperse the heat.

There is a large ground plane with all of the center gnd pins connected. That should provide some heat sinking. I start with the schematic, finish it then route the board and rats nest along the way.

OK, Krumlink. You can just shoot me now for this suggestion.

How about options to either use the motor drivers in parallel or to run separate motors? I happen to love boards that give me many options, but maybe that is not the direction you want to take. That's cool.

But if you could give this board the flexibilty to run:

  • One pair of motors with 1, 2, 3, or 4 drivers in parallel
  • Two pairs of motors each with either 1 or 2 driver in parallel.
  • Three pairs of motors with each driven by a single chip, or one pair optionally driven by two in parallel.
  • Four pairs of motors with each driven separately.

Now you have a generic motor driver board that could be the center of many different types of robot projects.

OK. Shoot me and/or ignore this suggestion. ; j

This would be tricky, Every IC would require jumpers, lots of them. And Headers, etc. I can look into a rough layout but man, you crazy!

I know. I should just start designing my own stuff instead of messing with yours.

It's very possible to do that suggestion, but the amount of traces and wires and etc you would need, it would be like a 4 layer board! Or just very physically large with lots of traces. Not totally impractical, but very bulky. It would be easier to build boards that stack on top of each other for your power needs. Each board has vertical headers with female headers below the board, so it parallels all the connections. Very simple to do.

Actually I think that a stackable H-bridge board could be quit handy. 
If you need more current, just slap another one on top and your ready to go. If you need a separate motor controller, solder on a 90 deg header for the control lines but still use the power headers and slap on.
Humm… that is something that I have never seen…

I have to agree. I worked on a board that had the option for 2 different variants of a chip, I2c and SPI. It wouldn't have been bad if all of the pins lead to different pins on the proc, but some were shared. It came out looking pretty good, but I don't know if I will be making a board that supports both again. I think the stacking concept that GA mentioned would be interesting to try to come up with.