h-bridges motor control

I am using a bipolar h-bridge ZHB6718 made by Zetex as part of a speed controller for a small electric motor - (a model train). The bridge is driven by a pwm signal from a microprocessor. The circuit works, but not as well as I would like, and it does do some odd things, like the motor suddenly changing direction. I don't have any test equipment apart from a small digital multimeter. Measuring the pwm output I get a reading, which I assume is a measure of the average dc value of the pwm signal. With the motor disconnected this value increases smoothly from about 0.7V to about 4.5V, but with the motor connected it only changes from about 0.7 to 1.0V, and the motor does not achieve its maximum speed.

There are four inputs to the microprocessor for forward, reverse, slow down and emergency stop. These come from a 433Mhz receiver. There are two outputs to the motor - pwm at about 100Hz and zero. These change for forward and reverse. The two leds from RC4 ad RC3 are forward and reverse headlamps.

Does anyone have any thoughts?

Since you get an output of only 1V instead of 4.5V then the poor little H-bridge is overloaded.

It's possible it is noise. Add the capacitors recommended for the regulator you have. You might also add a large cap (100 Ufd. or so) to the +12.

Is the low reading from the micro PWM?

Q1 and Q4 don't seem to be driven by anything specific. I could be wrong, I'm just back from the pub.

If it is more than a very small motor you may need to change the 1.8ks to 220 ohm and the 5.6ks to 510 ohms. Then you can drive a few 100 milliamps.
I should ask what transistors you are using.

Yes it is a small motor and only takes about 10mA at 12V. The h-bridge is rated at 2.5A. There are decoupling capacitors. Just forgot to include them on the schematic. Tried your idea ronv of lowering the value of the resistors. It does appear to be more stable, but I still can't get the speed I was hoping for, and yes the low reading I am quoting is at the output of the micro.


Replacing the motor with the multimeter I get a nice steady increase in voltage from 0V to 11.5V in both directions. I think I might try building a discrete component bridge. The integrated version seemed a nice idea, but I am not sure what it is doing.

Motors can be very noisy. I would try to put a 1K load where the motor goes. If that is ok it is probably noise getting into the micro.

Built a discrete component version, and got the same result, so I am back with the integrated h-bridge.

Tried a 1k resistor and it worked fine. I also tried a different motor. I have a small 12V motor that takes between 20 and 60mA depending on the load. It was driving a small pump so I suspect it was fairly loaded so that the current would be towards the 60mA end. It worked fine, with the voltage across it rising to about 10.5V.

The problem seems to be the motor that I want to use. It is a Mashima can type 12V.

Take a look at this site.

https://www.electro-tech-online.com/custompdfs/2011/04/motor-noise-reduction.pdf