# shielding/grounding question

Discussion in 'Electronic Basics' started by tempus fugit, Jan 31, 2009.

1. ### tempus fugitGuest

Hey all;

When a circuit is built inside a metal box to prevent hum pickup (in audio),
does the metal box need to be connected to ground to provide the shielding,
or is it the metal itself that blocks the unwanted interference?

Thanks

2. ### Jon SlaughterGuest

Both. A metal container acts as a shield because no static electric field
can exist inside of it. The electrons distribute themselfs on the surface to
maximize the distance between them(since they repel).

But dynamically fields can exist and by grounding the circuit to it, any
dynamic change on it will cause the circuit to follow it and hence the
potential difference will be 0.

So even if you somehow gave the chasis a potential it wouldn't effect the
circuit.... which it could otherwise.

e.g. suppose you ran a van de graff generator to put a large voltage on the
chasis.... The circuit inside could be effected as the electrons move
through the chasis to attempt to get to reduce the electric field.

Try using google next time? You'll get more out of it in the long run.

3. ### tempus fugitGuest

Thanks Jon

So it's better to have the chassis grounded from a shielding perspective.
Now, does it matter where it's grounded to? For example, I'm trying to avoid
ground loop problems, so I was going to keep my circuit ground separate from
earth ground. Could the chassis be grounded to earth ground with my circuit
ground (or, more accurately, reference) floating from it, or should the
chassis be grounded to the circuit ground?

Thanks again

4. ### Jon SlaughterGuest

Well, if you ground in multiple places you reduce the ground impedence.
Remember that the connection to ground isn't 0 resistance. Hence you end up
increasing the potential difference between the two. I doubt it would be a
problem though.

What I suggest is that you give yourself the ability to make several
connections to the chasis then do tests. One central wire might be good
enough. (if you have a ground plane for example) I've rarely seen more than
1 or 2 connections to the chasis in audio equipment though so it's probably
not a big deal.

I imagine you'll have more issues with the circuit itself rather than the
chasis and a well defined circuit doesn't need one except in rare occasions.

5. ### tempus fugitGuest

Thanks again Jon. Although I will naturally try to lay things out with no or
few ground loops in the circuitry itself, I'm concerned about ground loops
when using the circuit with other equipment. This particular piece of gear
will have 5 guitar effects pedals mounted to it, and hopefully I'll be able
to keep the ground that they will use floating from actual earth ground to
avoid the problem with ground loops (and thus ground loop hum) when it gets
connected to other equipment.

Thanks

6. ### Phil AllisonGuest

"tempus fugit"

** The metal box and audio signal ground must be at the same potential to
provide effective shielding.

Direct connection is not essential, it may be via a low value resistance.

Connection of metalwork to AC supply ground is purely a safety measure - it
prevents the metal acquiring a voltage potential above that of ground.

...... Phil

7. ### tempus fugitGuest

Thanks Phil

Ugghhh... So if I use a metal enclosure, both the signal ground and the
earth ground must be connected to the chassis - the 1st for shielding, and
the other for safety, which introduces the possibility of ground loops with
other equipment. When you say 'low vaue resistance' what type fo values are
you thinking about? A few ohms or a few hundred ohms? Also, do you have any
suggestions for combating the ground loop issue?
Thanks again

8. ### Phil AllisonGuest

"tempus fugit"
"Phil Allison"

** A few hundred ohms is usually OK.

Maybe add a 47 nF cap across it for RFI suppression.

** Power the device with a wall wart.

No need for any safety ground then.

...... Phil