pcmhacking.net

I've seen alot of people with speedway cars use electric water pumps and fans etc, just activated by a switch, but the beauty of this setup is you never forget to turn it on or off! once its setup it just works!

The reason why we hard wire them is so as there are no switches to forget, the PCM is smarter than me

It's a good idea but I would have thought that a variable voltage supply to regulate motor speed would be better than pulsing and electric motor on/off as each time it's energised it would create a spike in the initial current draw?

yep

its a bit more complicated than that tho.... remember a motor draws more current the slower its going.... the frequency is fast enough that the motor is still turning and generating a reverse voltage when the current is reapplied. so its not as big of a current draw as the initial startup. remember this exact type of PWM speed control is used to control your battery drill, radio control cars, pretty much all variable speed motors use PWM over straight out voltage regulation.

Using the davis craig controller we saw 10amp at low speed and 3amp at full...

PWM speed controllers are not that inefficient. whatever effect you saw im certain is wasnt a result of using PWM to control the speed, it was something else causing it to draw more current, like a problem with the speed controller or too low of a speed causing the motor to stall.

The high current draw at lower speed and lower current draw at higher speed is normal.
For a given supply voltage, a DC motor at lower speed will produce more torque (hence more current) than the same motor at high speed where it will produce less torque (less current) but it's running at more speed, if that makes sense.
Output torque of the motor is directly proportional to current.

http://www.me.umn.edu/courses/me2011/ar ... -tutorial/

Higher current draw at lower speed has nothing to do with PWM inefficiency.
PWM is very efficient as long as the frequency is high enough.

yeah one thing I found was I think anything below 30% duty cycle and the pump wouldn't turn ....

I also found when testing fuel pump with pwm, it was drawing a lot going slow but it was for half the time if that makes sense.