Re: [CAD_CAM_EDM_DRO] Re: PWM Back Emf

Well, I said that badly. Nevermind me, listen to Mariss. The essential is
the same, though, you're measuring the difference between the command signal
and the drive signal to read the back EMF.

On 1/13/07, Mariss Freimanis <mariss92705@...> wrote:
>
> That -V 'dip' is caused by the inductive energy stored in the motor.
> You can read the motor back EMF only when this current decays to zero.
> This negative pedestal will get wider as you apply an increasing load
> to the motor.
>
> This puts unacceptable restrictions on your circuit:
>
> 1) It only works at low speeds (low PWM duty cycle)
> 2) It only works at low PWM frequencies (<<1kHz).
> 3) It only works at light loads (<10%).
>
> A much more effective technique is to use "IR" compensation. It
> removes all the above restrictions. IR compensation adds positive
> feedback by increasing the PWM duty cycle proportional to motor load
> and can regulate your motor speed to within +/-2%. Do this:
>
> 1) Use a current-sense resistor to generate a voltage prop. to load.
> 2) Filter and amplify this voltage.
> 3) Use a trimpot to attenuate (2) to the exact value needed.
> 4) Sum (3) with the PWM command voltage using an op-amp.
>
> Adjust the trimpot to where the motor speed doesn't change with load.
> Too little and the motor will slow down, too much and the motor will
> speed up with increasing load.
>
> Mariss
>
>


[Non-text portions of this message have been removed]