If you want to reverse the direction of a DC motor, how would you do it? One of two ways:
- Change the positive voltage to a negative voltage
- Swap the two power leads
The first method is the simplest, but requires a dual-ended power supply–two sets of batteries, or a more complex circuit to split the voltage or generate the negative voltage from the existing positive. Either way adds overhead to the power supply.
The second method turns out to be the easiest to implement in practice, although it’s a bit tricky in an all-electronic solution. If you’re using relays to control the motor, you just use a DPDT relay with an off position, wired like so:
[insert relay schematic here]
To make the motor go forward, activate the FWD coil; to make it reverse, activate the REV coil.
The trick comes when implementing this in electronics. Most electronic switches (transistors) only pass electricity in one direction, so you can’t use a single switch to change the direction of current flow. The solution is an H-bridge, a set of four power transistors arranged like the upper and lower legs of an H, with the motor wired across the middle.
[insert H-bridge schematic here]
To make the motor turn forward, activate transistors 1 and 4. To make it reverse, activate 2 and 3. To burn out your transistors and power supply, activate 1 and 2 or 3 and 4.
On my CD-bot, I built an H-bridge from scratch out of (salvaged) small-signal, bipolar transistors. I wanted to do it from the ground up once, to know that I could. I also wanted to honor the goal that everything in the bot was made from salvaged materials, and I’ve never salvaged an H-bridge (as far as I know).
[insert picture of CD-bot motor driver board and 754410]
Since then, I’ve discovered the SN754410 quad half-H driver, which I purchased from Acroname. It’s a 16-pin DIP that provides four half-H-bridges, i.e. two complete bridges. It’s a piece of cake to use; and because one input controls an entire side of the H (transistors 1 and 2 or 3 and 4 in the diagram above), you can’t short it out with a simple mistake in the control logic. It’s a wonderful chip, I’m delighted to have found it, and I wonder why it’s not mentioned more often.