SN754410 H-Bridge leaking on one side?

I've got a simple robot using a SN754410 to drive two small DC motors. I have 4xAA batteries supplying power to the motors through the motor power pin.

The motor driven from the pins on the right side of the SN754410 stops and starts properly.

The motor driven from the pins on the left side of the SN754410 run constantly. When my robot receives a command that operates this motor, the motor (already running) perks up and is more active for the duration of the command. This seems to happen regardless of what's happening with the motor logic pin on that side.

Any ideas? I've looked for soldering mistakes around the SN754410 pins but can't find anything that looks like it'd be shorting out.

A "floating" control pin might cause a problem.

A "floating" control pin might cause a problem.

Thanks for your help zoomkat! I think this answer got me on the right track.

I wasn't sure what you meant by "floating," but t3h google says it might be a return current path to a non-earth ground? That got me thinking about grounding issues.

It looks like my problem was in software; every iteration through my motor control loop, I was turning my motor control pin ON, but never turning it OFF. The result was a motor that spun constantly.

(For those spectators unfamiliar with the SN75441, sending HIGH to a control pin enables its motor, sending LOW disables it.)

So now it's working, until I break it again with the next "improvement" to the project.