Society of Robots - Robot Forum

Software => Software => Topic started by: lcab on April 07, 2010, 04:44:17 PM

Title: revoluations
Post by: lcab on April 07, 2010, 04:44:17 PM
is there a way to measure a gear motor revoluations if i am not useing a encoder because we are using sabertooth motorcontroller and an axon microcontroller
Title: Re: revoluations
Post by: waltr on April 07, 2010, 05:40:17 PM
Not that I'm aware of. You need some kind of sensor to measure the rotations.
Title: Re: revoluations
Post by: Soeren on April 07, 2010, 06:14:59 PM
Hi,

As I usually say: Everything can be done with electronics  ;D

And I didn't even have to break Eagle out this time, so I'll just point you over to Roman Blacks encoder free speed encoder (http://www.romanblack.com/encoder.htm)
Title: Re: revoluations
Post by: waltr on April 07, 2010, 08:03:48 PM
But doesn't the sabertooth controller use PWM for motor speed control whereas that circuit requires a "smooth" DC voltage?
Quote
Sabertooth's transistors are switched at ultrasonic speeds (32kHz) for silent operation.
Would that circuit still work?
Title: Re: revoluations
Post by: Soeren on April 07, 2010, 09:24:00 PM
Hi,

But doesn't the sabertooth controller use PWM for motor speed control whereas that circuit requires a "smooth" DC voltage?
Quote
Sabertooth's transistors are switched at ultrasonic speeds (32kHz) for silent operation.
Would that circuit still work?
Well spotted!

However, I don't think it will be much of an issue, because the 32 kHz should be fairly easy to filter out of the equation and I have a hunch that the poles will show through it.

If it fails, a Hall switch or other method may be the way to go.


Perhaps lcab could post some photos of the motors?


I wonder why they went that high in frequency on this controller, as even young people (i.e. kids) cannot hear 25kHz (even 10kHz would probably not make much noise, as the noise comes from armature windings pushing and pulling their neighbours and the mass of the copper will limit their movement increasingly as the frequency goes up.

Further, at 32kHz, there is no PWM at the motor (due to the armature inductance), but an LC smoothed pulsating DC, so the only joy they'll get is reduced power loss - This may or may not be valid for high speed outrunners - I haven't ever tested one, so perhaps their inductance is adequately low to enable PWM at that frequency.

Yet another bad thing about going up in frequency, is that the switching loss increases, as the switching transition times (which is an unchangeable feature of the power transistors) will occupy a larger relative part of the period.

To make a good PWM control, it should be kept at around 1..5kHz (depending on motor specs.), as on high frequencies, the part of the waveform with full voltage, which is what gives it good torque, is "integrated" down to (close to) the mean voltage.