Thoughts on motor control, efficiency/heat, PWM + Voltage adjustment

So I was thinking about the motor controllers and motion is handled with Maslow.

The L298 is having to turn on and off quickly when getting to the final position. Each time the motor is turned on there is demand of inrush current needed. This high amperage at the supply voltage is somewhat of a waste. What if instead we could not exclusively PWM our way to position but have an ability to lower the supply voltage? This would slow the motors speed and allow less on/off cycles. This would lower the times the inrush current was required and since the supply voltage would also be lower during these times, less power would need to be dissipated by the H-Bridge.

My idea is what if we pulled another signal from the arduino, passed it into an LM317 then passed that into an NPN power transistor like an 2N3055. We would then have dynamic control over the input voltage to the H-Bridge allowing the motors to run at higher voltages/speed (24 volts) in transit & homing, while at the same time allow lower voltage/speeds (6 volts) during fine movement?

The problem is that when you lower the voltage through a transistor, that
transistor has to dissipate the heat for the power that is not getting through.

The motor controller is (or should be) using mosfets which have very low
resistance when turned on, and get turned on very quickly so they spend very
little time in the bad intermediate state.

nobody controls motor speed by adjusting the voltage with a transistor,
everybody uses PWM. If we are having problems, it’s because the devices selected
are smaller than they should be.

1 Like

I see your point @dlang. I also feel the BEST option would be an upsized H-Bridge. :grin:

The transistors would have to take all that powerless but they could be capable of higher power output.

This idea was originally just a mind experiment with of speeding up rapid (long non cut) movements and getting accuracy with requiring less switching the motor on/off. :