That will work, if you securely ground the shield to the equipment enclosures, preferably on both ends.
One common technique is the "drain wire": gently spiral a bare copper wire around your signal wire bundle, then wrap with foil and tape. Use the drain wire as your ground... it's easier to terminate than the foil itself.
That was done in quite a few vintage radio and audio systems in lieu of real coax. In fact, that technique offers a 100% shield whereas a braid shield varies from 60% to 95%, depending on the coarseness of the braid weave. Lots of current crop of audio, video, TV and satellite coax use a foil shield in addition to a braid shield.
--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in
By the way, is there some reason you can't just use shielded cable in the first place? The shield is much more likely to be effective than your homebrew aluminum foil wrap will be, especially over time as your drain wire oxidizes and stops making good contact with the foil.
You should be aware that shielding is not a panacea - it can make EMI problems worse, as well as making them better. It depends on what the shield construction is, what frequencies of EMI you're being affected by, whether the coupling is capacitive, galvanic, or inductive, what kind of connectors you're using, and other factors. I have here on my shelf more than half a dozen books specific to EMI, shielding, and grounding, as well as innumerable chapters in other books. The one thing I think they all agree on is that there is no "silver bullet" for EMI; you can't just throw shielding at the problem, you have to understand what problems you're trying to solve and how the solutions you're attempting to use work.
Just curious about something. Let's say I have some wires that I want to protect from EMI. Can I simply wrap them in aluminum foil? (and then put some electrrical tape over foil to just keep it all together. Would it offer me decent EMI protection?
I see. Well, I think that I could come up with something, such as wrap a bare wire along the foil and contact it to ground at both ends. Good suggestion.
Thanks. I will save your post. In my application, the particular circuit is designed with EMI in mind, but some extra protection would not hurt.
If you're running low-level signals between grounded boxes, they had better be isolated or differential; AC power safety grounding will ensure lots of "ground loop" potential, and the effective system loop area will be huge, so any magnetic fields will induce additional hum into single-ended signals. In this situation, grounding both ends of the shield generally improves a bad situation. If the sigs are differential, the common-mode magnetic coupling you mention has no effect, and is likely smaller than the inherent box-box hum anyhow. CMRR is where it's at.
I've seen real-life systems in commercial buildings that had volts of AC between chassis grounds in gear a few rooms apart. Connecting a cable shield between the boxes typically reduces that hum a little.
But the bigger issue is EMI and ESD. If you ground the shield at one box, consider the situation at the other: the wires and the shield form an inductor/antenna all the way back to the "grounded" box and the incoming signals are *not* forced to be RF equipotential with the chassis of the local one. So large RF and ESD common-mode potentials will piggyback on all the signals coming into this box, and the shield is virtually useless.
And a good ESD zap onto the shield, near the "ungrounded" box, will couple to all the signal leads and rip straight into its guts.
So, ground a shield at both ends, and points between if feasible.
Why not? If the "ground loop" is caused by AC fields being induced into the overall loop area, it won't make any difference. And, as noted, if small ground-loop potentials will mess up your signals, the sensible thing to do is isolate or go differential
Some people like to hard-ground the shield on one end and capacitively ground it on the other end, attempting to have single grounds at low frequencies and double grounding at HF.
Shields intended for EMI reduction due to voltage fields are best grounded at ONE end only. If they are grounded at both ends, then any potential difference between the ground points will result in current flow through the shield which will cause magnetic coupling to the wires that are being shielded. This is called a "ground loop" and is a frequently misunderstood and misapplied shielding technique.
Which end to ground is the question. Usually the best ground is made near the source of the interference. In some cases, it is desirable to test the shielding by grounding first one end and then the other to see which configuration is better.
If grounding both ends improves things, that indicates that there should be a heavy, separate, ground strap added between the two grounding points. This separate ground strap should not be the shield for a cable.
The drain wire inside or outside the foil would be equally effective. Make the decision based on what's easier to implement.
But why not just ground the shield to each chassis on both ends?
When I step outside of work, in a wooden building, I see pretty much this view...
formatting link
And from my dining room table at home the view includes the antanna farm on Mt San Bruno, across the Alemany Gap. Cheap Radio Shack phones and PC speakers tend to do weird things around here.
It would offer you decent protection from "E" in that "EMI" word. There is no way aluminum (not a ferrous metal) offers any protection from the "M" ;-) If it's the "E" part you're concerned about (high frequency circuit), you can get some benefit of using a shielded cable (provided the shielding is done right - the aluminum foil has a drain wire, then a braid over it, grounded etc.). If it's a power circuit, don't even bother with the aluminum shield.
-------------------------------------
--
Dmitri Abaimov, RCDD
http://www.cabling-design.com
It's hard to shield 60 Hz mag fields, even with a substantial steel box... try it. It would take massive aluminum plates to produce measurable eddy-current shielding at 60 Hz.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.