criticize/debug my PWM to 0-5V analog converter

[bschwand]

Hi all,

I am looking for feedback on a circuit I made, which works, but doesn't :-)

Background: I have some small BLDC motor with integrated driver that has a potentiometer to adjust speed from 0 to 100k rpm.
The BLDC driver is supplied 12VDC. The potentiometer has outside legs connected to the driver, those connections I have measured provide 0 and 5V. The wiper leg is a 3rd connection.

I want to control the BLDC motor with a PWM signal, about 1kHz and duty cycle varying between 0 and 100%, at 12VDC. (that is just what the other circuitry I want to interface with provides)

Here is the circuit I designed, I simulated in LTspice and worked well enough.

The actual circuit, when not hooked to the BLDC driver, outputs effectively from about 0.2-4.8V when fed a PWM signal with varying duty cycle.

When I connected my circuit to the BLDC driver, I noticed the 0 and 5V pins should have a resistor across (like the original potentiometer) otherwise the motor will not spin at all.

The 0-5V output however does not reach the full range. Given a duty cycle of 0%, motor is off, and speeds up as duty cycle is increased until about 40%.
At that point, any increase does not increase the voltage output and the motor speed does not increase. I noticed when I change the duty cycle, the motor briefly (sub second) speeds up a bit, then settles back to the same speed as with 40% duty cycle.

I unfortunately have no doc or schematic of the BLDC controller, and the IC on it has no markings.

Any ideas what is going on, why this is not working as expected ?

[bschwand]

replying to my own question

somehow, everytime I try everything I can think of, finally give up and turn to the forums for help...
right after I click send I have another idea... which solves the problem.

The issue was that cables from the circuit to the motor driver were very long, and my power supply was limiting current to the motor at a max of 350mA. When hitting the higher speeds, this motor takes up to 1.5A.

So the circuit works, pretty happy about that. Also happy that the LTspice sim results and real-life circuit are very close.