Previous in Forum: Spline Cutting   Next in Forum: machining time calculation
Close
Close
Close
3 comments
Rate Comments: Nested
Member

Join Date: May 2008
Posts: 5

power+control shielding

12/06/2008 10:23 AM

hello,

Someone tell me why, when using shielded cables (power+ control), sometimes they only connect to ground in one side (I thought should be both so they could have the same potential.

Thanks

Register to Reply
User-tagged by 1 user
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Good Answers:

These comments received enough positive votes to make them "good answers".
2
Power-User

Join Date: Jan 2008
Location: since 20 Jan 09, the USSA
Posts: 375
Good Answers: 81
#1

Re: power+control shielding

12/06/2008 11:32 AM

It depends on what the shield is doing for you.

It is true that you want both ends of the shield at the same potential. At 50/60 Hz, that will always be the case with one end floated, because the cable is electrically short - it is a minuscule fraction of a wavelength, and any coupled shield potential will not generate currents on it. But if you ground both ends of a 60 Hz shield, and if the far end is very far away such that the 60 Hz potential there is slightly different than at the near end, then you have the potential for current to flow. The shield won't protect as well with current flowing, because the shield isn't thick enough to absorb 60 Hz current, and you can get inductive coupling to the inner circuits if the circuit grounding isn't done properly.

So if the purpose of the shield is to protect sensitive measurements from mains power frequency interference, then you usually ground the shield at one end only.

But if the purpose of the shield is to either contain or protect against radio frequency (rf) energy, then the shield needs to be grounded at both ends, and you have to pay attention to how you get the ground. The difference here is in frequency and wavelength. At 100 MHz - center of FM broadcast band (BCB) - the wavelength is three meters and your shield needs to be grounded at both ends. Potential differences across the shield due to the wavelength will cause currents to flow. By grounding both ends you will keep the currents flowing on one side of the shield (the shield is thick enough to absorb 100 MHz energy). This is due to skin depth.

Skin depth is a phenomenon where the current tends to ride on the outside of a conductor and penetrate just a few skin depths until totally absorbed. Skin depth decreases proportionally to the square root of frequency. At 60 Hz, you would need something like hollow bus bar to get any absorption at all, whereas in the BCBs any conductive braided shield will have lots of absorption.

But keeping those currents on one side of the shield also requires careful attention to the shield termination. If you terminate a shield with a long drain wire, it has no "inside" or "outside" and the careful separation maintained by your expensive braid shield is all for naught - the current that was supposed to be on either the inside or the outside is now on both sides. If you are counting on the braid to keep current on one side of the shield, then you need to peripherally terminate the shield at the connector to keep the shield current on the side you were interested in - inside if you are containing emissions in order to protect BCB radio reception, and outside if you are protecting a sensitive circuit from the effects of an intense electromagnetic field from a nearby transmitter. Note that in many applications you will need to do both: contain radiated emissions to a very low level and protect against external electromagnetic fields at much higher levels. Careful shield termination aids in both these goals.

This is a very top-level discussion of a topic about which entire books have been written. If you are interested in more detail than I have provided, I can give you references.

Register to Reply Good Answer (Score 2)
Anonymous Poster
#3
In reply to #1

Re: power+control shielding

12/07/2008 2:41 AM

In case of low voltage system (24-48 V) where the safety is not an issue, I think that leaving the shield ungrounded in both sides is a perfect solution. No noise currents, shield at the same potential over all length (almost perfect cage). Some care should be taken at the termination ends, but there are a lot of solutions.

Register to Reply
Anonymous Poster
#2

Re: power+control shielding

12/06/2008 11:13 PM

The general rule of thumb for shielded cables is that you ground only one end of the shield to keep electrical noise out, as you would for a low voltage signal cable and you ground both ends of the shield if you want to keep electical noise in, as you would for a servo power cable.

Register to Reply
Register to Reply 3 comments

Good Answers:

These comments received enough positive votes to make them "good answers".

Previous in Forum: Spline Cutting   Next in Forum: machining time calculation

Advertisement