Cable Shielding: Foil or braid

[qu1nn]

Hello

I am under the impression that if you wanted to protect a sensitive
electronic device from external coupling (aka nearby cables near a
switch or a sensitive cable near a noisier one) that it would be best
to use shielded twisted pair with 100 shielding. Generally when you
have 100% shielding it would be a foil with a bleeder cable strands.
However if you get a cable with braids you only get 90% coverage.

So if one is to ignore the ENDS of the cable and you are only looking
at protection from nearby em fields would it not make more sense for
100% shielding? or is it that the foil shield does not really "perform
as well"?


Thank you for your time
Qu1nn

 

[Bob Myers]

Hello

I am under the impression that if you wanted to protect a sensitive
electronic device from external coupling (aka nearby cables near a
switch or a sensitive cable near a noisier one) that it would be best
to use shielded twisted pair with 100 shielding. Generally when you
have 100% shielding it would be a foil with a bleeder cable strands.
However if you get a cable with braids you only get 90% coverage.

"Shielding" a cable is actually a bit more complicated than this. Not
only are you trying to prevent capacitive coupling of noise into the
signal path (which is what a "shield" does), you're also trying to
prevent inductive coupling (achieved through minimizing the open loop
area of the complete signal path, both "forward" and "return") AND
you're trying to keep noise from entering the system through at least
one other path - noise voltages/currents showing up on the system
ground via the "shield" connection. If the "shield" of a cable is
tied to the "ground" at both ends, it is a possible path for currents,
including noise currents - and if that path shows a significant impedance,
those currents result in a noise voltage (potential difference) between
the two "grounds," which in some circuit configurations can lead to
problems. So it's not just a matter of the physical coverage of the
signal conductors by the "shield," but often also a question of the
resistance
or impedance seen by currents within that ground connection. Foils
may not be very good in this respect, so often the best performance
will be obtained from a combination foil/braid "shield."

Bob M.

 

[cornytheclown]

"Shielding" a cable is actually a bit more complicated than this. Not
only are you trying to prevent capacitive coupling of noise into the
signal path (which is what a "shield" does), you're also trying to
prevent inductive coupling (achieved through minimizing the open loop
area of the complete signal path, both "forward" and "return") AND
you're trying to keep noise from entering the system through at least
one other path - noise voltages/currents showing up on the system
ground via the "shield" connection. If the "shield" of a cable is
tied to the "ground" at both ends, it is a possible path for currents,
including noise currents - and if that path shows a significant impedance,
those currents result in a noise voltage (potential difference) between
the two "grounds," which in some circuit configurations can lead to
problems. So it's not just a matter of the physical coverage of the
signal conductors by the "shield," but often also a question of the
resistance
or impedance seen by currents within that ground connection. Foils
may not be very good in this respect, so often the best performance
will be obtained from a combination foil/braid "shield."

Bob M.

Just curious......but are there any circumstances where grounding the
shield at both ends would be required or helpfull. I seem to recall
reading some material on shielding and I seem to remember there were
some instances where on would connect both ends of the shield.....

 

[Rich Grise]

Just curious......but are there any circumstances where grounding the
shield at both ends would be required or helpfull. I seem to recall
reading some material on shielding and I seem to remember there were
some instances where on would connect both ends of the shield.....

With RF, and coax connectors, the shield is connected, but it's a
sort of different animal than your original question, I think. I don't
know the relative merits of braid vs. foil, but I'd bet that foil is
considerably cheaper.

Cheers!
Rich

 

[Bob Myers]

Just curious......but are there any circumstances where grounding the
shield at both ends would be required or helpfull. I seem to recall
reading some material on shielding and I seem to remember there were
some instances where on would connect both ends of the shield.....

Consider that when you connect the "shield" of any cable to
the ground at both ends, what you're really doing is tying the
grounds of the two pieces of equipment together. This means
that the "shield" may no longer BE just a shield, but possibly
is a return current path. Sometimes, that's what is wanted, and
sometimes it isn't. Connecting the shield at one end only means
that it really IS just acting as an electrostatic shield (and if there
isn't a return path already provided within the cable assembly,
then you likely have problems because that current IS going to
return SOMEHOW, or the thing just won't work). Both
conditions (connected at one end or connected at both ends)
have their uses.

Bob M.