If the motor inductance is high enough and the PWM frequency high enough, this actually works fine - I have used this many times before. You end up with a zero average current (at 50% duty) but a small ripple current. One advantage of this ON forward-ON reverse mode over the more "traditional" ON forward-OFF mode is that there is no significant portion of the PWM cycle where the FET body diodes (or inductive clamp diodes if using BJTs) conduct, so the losses can be lower in the switching devices. Since the ripple current causes some additional losses in the motor, this isn't usually an overall efficiency benefit but it can reduce the heatsinking requirements of the switches. In the above analysis, I am assuming that the PWM frequency, in any PWM mode, is high enough that the current does not decay to zero during any part of each PWM cycle if the average current is high. This is much more efficient than using a very low PWM frequency and relying on mechanical "averaging" of the motor torque. There are several other possible PWM modes, like ON forward-Shorted, where the off time shorts the motor, allowing the current to continue to flow, driven by the inductance of the motor. This has advantages and disadvantages as well. On Mon, Nov 19, 2012 at 7:05 AM, IVP wrote: >> Assuming active-high outputs, with a 0% duty cycle, PA1 is off, >> P1B is on, and the motor spins one way. At 100% duty cycle, >> the opposite is true. A 50% duty cycle implies an alternating (but >> balanced) forward/reverse action > > Oh, I see what you mean. But using two outputs of an L293 is > actually an H-bridge isn't it, as each output has a high- and low-side > transistor. What you're suggesting would be the same as alternately > turning on/off the (S1, S4) and (S2, S3) pairs of discrete switches. > /Enable =3D 0 of the L293 performs the same function as turning all > devices off in an H-bridge > > As you say, the nett effect would be no movement although I suspect > there will be (quite) some heat to dissipate in the motor, as its > effectively > receiving full power. This wouldn't be good for a fan-cooled motor. And > perhaps heating also in the drivers if the noise generated and/or switchi= ng > waveforms stops them operating as the intended digital devices > > Joe > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .