I am controlling two wheelchair motors and it is tank style steering. The motors run about 7 amps no load. I have my eye on this controller: http://www.electronickits.com/kit/complete/motor/k166.htm I'm just not sure I can interface the Arduino with this controller. The controller goes from full reverse when the pot is 0v to full forward at 5v.
Using K166 you'll have to beef up the current handling of the board and terminals (perhaps mount heavy wire from as close to the MOSFETs as possible and directly to the motors.
You shouldn't take too much notice of the mentioned 6V..32V operating voltage. At 6V it won't allow much current and the MOSFETs will run hot with a light load. t anything beyond 12V, it needs at least a minor mod, to keep the outputs of the LM324 from going too high and perhaps killing the transistors.
It's an overall bad and slow switching design, made by someone that obviously don't understand MOSFETs, it could be changed here and there to be almost bearable, but why bother when better designs can be had?
Besides that, If the Arduino has got a D/A-C, you could rip out the potentiometer and feed a the S/A-C output to where the wiper connected.
Better yet, rip out the LM324 and control the bridge directly from where pins 8 and 14 used to connect
Question is... Is it worth paying $25 a pop if you're going to just use the bridge and a little "glue"?
(You'd need 2 kits, one for each motor).
Any suggestions on a motor controller? I'm trying to follow this guy: http://www.rediculouslygoodlooking.com/site/lawnmower.html
He has got a schematic for a H-bridge on that site. This bridge is flawed as well though, the pull up resistors are of a way too large value and the way he's controlling the N-channel devices are beyond redemption, as the switch on times of the low side transistors will be horribly long, making them white hot. The suggestion of using 25kHz for the PWM makes it get even hotter, than if using a more sensible frequency for DC motors of around 3 kHz tops.
I didn't read through the entire page very carefully, so i don't know how he used the PWM, but you should only PWM the lower side devices, as (when used correctly) N-channel devices in a (more or less) matched pair are faster to switch.
If you are able to follow a schematic and perhaps etching your own PCBs, I could post a more efficient design.