>
> At 11:56 AM 5/8/97 -0700, you wrote:
>
> >I've finished my first project, a servo controller.  It mostly
> >works, but there is some unreliability, which I suspect is from
> >motor noise and power dips when motors start up.  _Mobile Robots_
> >suggests that these problems can be solved by putting a capacitor
> >from power to ground, but they don't specify the size.
>
> I've done quite a bit with servos.
>
> Make sure you have VERY heavy traces to provide power to the servo.
> Include a series resistor on the signal out of the PIC.  270 ohm minimum,
> max varies with servo brand, 10K is reasonable for most, would go as high
> as possible if I were you.  If in an RF environment, I suggest using
> standard strategies to reduce noise - chokes, caps, shielding, etc.
>
> Andy
>
Is the purpose of the serial resistor to lower the voltage of
the PWM signal?   I've been using a PIC to control 3 servos on
my Inchworm robot for the past 9 months and I don't use a
serial resistor, but I do use a higher voltage for servo power
than for the PIC, which naturally makes the PWM output a lower
voltage than the servo power.  Initially, I was observing
instability in the servo when the PWM signal was the same
voltage as the servo power.  Then I checked out the PWM signal
with an oscilloscope and discovered that it was about 0.3 volts
lower than the servo power voltage.  So I started using 5.3
volts as the servo power volgage and the instability was gone.
Of course, I now use 5.75 volts as the servo power voltage
because I want to get more power out of the servo.  I'm
thinking of going to 6 volts on my next design which, I've
heard, is the absolute max for servos.  Can you confirm this
Andy?

Keith

Keith D. Kotay
Ph.D. Candidate
Dartmouth College
kotay@cs.dartmouth.edu
http://www.cs.dartmouth.edu/~kotay