Andy Shaw wrote:
.......
  I found that a series RC snubber across the motor contacts worked
>> fairly well - R = 15 - 100 ohms, C = 0.047 - 0.1 uf or so.
>>
>
>As I said in an earlier post I'm more of a software kind of guy. Could you
>describe what the arrangements are for an "RC snubber", and what the
>connection arrangements are with respect to the motor contacts! Also if you
>are feeling particularly educational any chance of explaining how the thing
>works (or are we well into the black art of hardware design here...)!
>

Andy, thus starts a 3-week long seminar [just kidding - not really
black arts].

Basically, as indicated above a simple snubber is just a series
R-C directly across the motor terminals. However ---

Clasically, snubbers have been used across "relay contacts" in order
to protect the contacts from the large voltages created when you
interrupt power to an "inductive" load - E = L di/dt - when you break
current quickly by opening the relay contacts, di/dt is huge, and
so is E. This tends to burn the relay contacts and impact reliability.

It also produces loads of EMI [high-frequency noise] that gets
propagated onto the rest of the system, and can disrupt operation.

For DC relays, you can use an upsidedown diode for spike suppression,
but for AC classically they have used RC snubbers. The R actually does
the snubbing [ie, provides a low-impedance path for the inductive spike
to discharge through], while the C blocks the AC power from appearing
directly across the R. Typical R values are 10 - 100 ohms - and you
don't want this right across 220VAC, thus the C in series.

I have taken some great photos in the past of relay contact arcs with
and without snubbers. Without, you get a few msec of high-freq, hi-V
spiking on contact open. With a proper snubber, you get a slow wave
of just a few volts.

Nowadays, some people probably use bidirectional transient suppressor
diodes instead of R-C snubbers.

In the DC servo motor situation being discussed, the application of
the snubber is a little different. Here, it is basically snubbing
the noise generated by the brushes in the motor - they are, of course,
making and breaking contact and generating noise continuously.

Here, the snubber goes across the motor terminals. For DC servos,
they appear to simply use a small electrolytic cap, and this is
probably fine, and simply acts like a low-pass filter. I have also
used just a small R across the motor, and this also works, but
of course draws load current.

However, for the case of the Tamiya motors I mentioned, I am
running those with a PWM-controlled H-bridge chip, so the DC is
interrupted at a hi rate, and using a plain cap across the motor
ends up causing large current spikes:  I = C dv/dt - where dv/dt
is large. Therefore, I used a small R [10 - 100 ohms] in series
with the cap for best effect. The R snubs the inductive currents,
and also limits the capacitive currents. The C blocks the R
from DC loading the power source.

So, in this case, what I ended up with looks just like the R-C
snubber you find across AC relay contacts.

hope this helps,
- dan michaels
www.oricomtech.com
===========================

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics