It will make very little difference, because the
current is mostly DC. Even if you are using
PWM to drive the motor, the inductance of
the coils will even out the current flow.
Do you actually have a noise problem, or
are you just thinking premptively? I have
never taken any special precautions to
reduce RF noise from my motors, and I
have never had any problems.
Here are a few things you can do that
should be more effective than shielded
1. Use good motors, not cheap toys ($).
2. Use motors with graphite brushes ($$).
3. Use brushless motors ($$$).
On Sun, 12 Mar 2006 02:59:26 -0800, firstname.lastname@example.org wrote:
If you have encoders or other sensitive TTL level signals, shielding the
motor leads will help. PWM from MOSFET H bridges can have rise times in
the 50 nS region = ~1v/nS with 48V power = pretty nasty for electrostatic
coupling to nearby wires. Also if you have a brush noise suppression
capacitor mounted at the motor, a large current spike will happen
eveytime the HBridge switches...
It should help. At the very least use zip wire [close-parallel wires]
or twisted cable, and/ or shielded multi-conductor cable.
First off, the motor brushes in brushed motors sparking over generate a
huge amount of high-frequency noise, which can be propagated as radio
waves. Try running your electric drill, and watching the TV screen.
PWM'ing also will generate large spikes at the on-off points.
Secondly, this translates into current noise in the leads to the motor
controller, those wires will have significant inductance if not
extremely short and heavy, and you will have a nice inductive loop
which will then act as an antenna, and propagate the current noise as
RF. This can be especially detrimental to your controller board sitting
a few inches away. You can reduce this by using twisted pair to reduce
the inductive loop, and further shielding the pair to contain the
residuals. Normally the shield is only tied to ground on one end, to
prevent ground loops, and you can experiment regarding which end works
best for your situation.
Thirdly, besides this you also want to do all the other things to
reduce the noise generation in the first place. Caps or snubbers on the
motors, bypass and reservoir caps on the motor controller, etc. The
following site has a wealth of info ....
- dan michaels
OK, shielded cable can help reduce small amounts of noise, but you need to
reduce the noise as much as possible first.
(1) Put a small ceramic (0.1uf) capacitor at the motor terminals. This will
kill a *lot* of noise and a *lot* of the big spikes.
(2) If it is practical for your circuit, design an L/C filter to put between
the PWM driver and the motor. This will smooth out the current spikes of
the PWM and put mostly DC over the wire. (See Note)
(3) Shielded is good, but twisted pair will do just as well.
There is a big debate here over the use of L/C circuits between PWM drivers
and motors. This is pretty much and engineering question, and it has been
argued to death. Suffice to say, assuming permanent magnet motors, torque
is directly related to current, and voltage at the motor represents the
back EMF (and internal coil resistance times current) of the motor. An LC
circuit driven by a PWM source becomes very near a current source and it is
ideal for driving motors.
I ran shielded cable to my drive motors, but found that what really helped was
to put a ferrite bead around the motor leads, as close to the motor casing as
possible. This made my microcontroller a lot happier.
It would still reset occasionally, until I found that the real problem was a
ground loop (acting as an antenna) on a reset line.
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.