I have a desire to control some DC electric motors using a microcontroller.  Under the covers these are normally H-Bridges, so that the motor can run in both directions. I also require that the motors be able to undriven – to freewheel without electric braking.

The cheapest motor controllers will only deliver 800ma (per motor), which might be enough, but possibly isn’t. I think the L9110 datasheet (also known as the HG7881) hints at a peak of 2amp, but my Chinese is insufficient. There is no information on the switching frequency. These controllers drive the motors in either direction, or brake – but they will not disable the motors.

The next cheapest drivers are based on the L298 motor controller chip. The parts number variants indicate the packaging type; L298N is the vertical version, and a L298P is surface mount – which I can’t get my head around, how are you meant to dissipate the heat with a surface mount version? Anyways, the L298 can switch at up to 20KHz according to the L298 motor controller datasheet, and delivers a total of 4A, 2 amp to each channel, and a total of 25W maximum, presumably about 12.5 watts per channel. I’m looking at motors of around 6V, so the maximum wattage isn’t a consideration, but I’d have to start thinking about the power limits if they were 12V motors.  I’ll need to supply more than six volts because the L298 module sucks up 0.7V doing its business.  Even the cheapest modules using this chip expose the enable line for each motor, and they include a jumper to permanently enable (which I don’t want, but that’s just a matter of removing the jumper from the enable pin); more expensive modules expose the current sensing pins on the L298 (which I don’t think I have a use for). This unusual one is a l298 with a connector for stepper motor. Solabotics has an informative (if poorly spelled) L298 instruction sheet with examples hanging off their product page.

If I could switch at 30KHz, that’s beyond human hearing and thus “silent”, but beyond these two controllers there’s not a lot of commonly available choice, so I’m going with the ever-so-slightly more expensive and actually lets-me-do-what-I-want choice.

Higher frequencies seem to require MOSFET technology, which means dollars – and it seems higher voltages for the motors, which means more expensive motors, and more unusual power supplies (more cost).  I recall that higher switching frequencies lead higher efficiency. Oh well.