At home, I often use some screened audio leads which carry speech signals at about 100 to 200 mV. The leads may carry one or two signals (mono or stereo) and may be about 1m or 2m long. They are not balanced.
Sometimes it's unavoidable but my audio lead has to pass near a PC and also about 6 inches below and in front of a CRT display. I want to avoid electrical noise interfering with the audio signal.
(1) Would it help to clip some ferrite on the audio lead? If so, then at which point on the lead is the best place? (Source, destination, where lead passes near causes of interference). Is it better to use more than one ferrite?
(2) For a given max outer lead diameter of about 3mm (maybe 4mm), I guess some audio leads are better at intereference shielding than others. I'm in the UK; is there a type of decently shielded audio leads you might recommend without getting into really expensive stuff. (I can add the connectors.)
It only helps if there is a problem. Turn the audio gain up high with no music playing, then switch the PC on. Do you hear noises? If not, don't bother.
If you decide you do need ferrite, get the biggest rings you can accommodate and put as many turns of coax through as will fit. Position the ferrite close to the amplifier input.
For decently shielded cable, try some CT100, which is used for satellite dish feeds. It is not terribly flexible so you won't be able to drape it around easily, but it has 100% shielding, and won't let anything in. It is also pretty cheap.
Not clear whether this is a theoretical question, or if you are curently experiencing this electrical interference noise? If it is a practical situation, mentioning some details might prove helpful.
No it will not help no matter how many or where the ferrite beads are. The reason is that any audio interference is *in-band*, at the same frequencies as the desired audio signal. You canot filter out in-band interference without also filtering out the desired signal.
The remedy is to move the cables away from the source of noise. Use longer cables to avoid the problem spot(s). Use better cable with better shielding, etc.
It would be easier to try simply using longer cable(s) to avoid the problem spot(s).
If built properly to CE EMC guidelines (applicable in the UK), generally your domestic PC and connected equipment should not be causing that type of interference.
I run lots of Dell equipment around here next to a stack of Hi-Fi audio components. No problem whatsoever with noise pickup, and if you've ever taken one of their PCs apart you'll see a fairly serious job has been done of shielding and filtering RF interference reaching to the outside environment.
If home built (or it was built for you by a smaller manufacturing fling-it-together concern) then you'll have to try some DIY cures. From experience in working at an EMC testing facility - my number one tip is sorting out these issues is this ...
Ensure that all grounding points on the motherboard are electrically connected through conductive posts to the case (carefully avoiding parts of those posts shorting out against any leadouts or components hanging from the board). Plastic standoffs are a no-no.
Having just part of the motherboard grounded makes other bits attached to it (including _all_ conductors on external cables) liable to pickup and transmit the high frequency switching hash from the processor and graphics circuits. In extreme cases, parts of the motherboard itself will be singing like an aerial sending waves to be guided through convienient slots in the case.
Similary, if you have an expansion cards attached to the motherboard, ensure that the metal backplate of these cards is secure in being grounded to the casework and that there are no rectangular gaps between card and slot. You can get conductive meshed foam to fill the gap before you screw the card in. Any other whole gaps in the case wider than a couple of inches should also be screened across. There should be conductive covers over unused drive bays for instance.
OK so, I'd start there first ... and then ye can implement further measures from then on. Ferrites on external leads is an often used tweak, but it is better in EMC terms to deal with the major problems first at source of the noise and then move on.
It can't hurt to try, but I'm not all that hopeful. Ferrites are useful at blocking radio-frequency interference, but they won't do very much (if anything) to block interference which is actually down at the baseband-audio frequencies. If you're getting interference from a CRT, it's likely that it's from the horizontal and vertical sweep circuitry (and perhaps from modulations in the beam power), and these signals are likely to be in the audio range.
If I recall correctly (possibly not) this sort of interference can result either from pickup of the electrical field, or the magnetic field (radiated from the interferer in either case). A shielded cable helps keep out the electrical field (the more shield coverage the better), but I don't believe it is as effective at keeping the magnetic field out... at least, not at the low frequencies involved here.
You might want to see if you can find some shielded twisted pair cable, and make up some audio leads from that... use the pair for the audio signal (connect both wires at both ends) and ground the shield at one end only. Some people have reported better results with that sort of cable.
You could try using a twisted pair, with a balun transformer at each end.
Belden makes a lot of good types of audio cable stock - the table at
formatting link
shows a bunch of their part numbers for various sorts of A/V cable. Their #1508A is a twisted pair with a 100%-coverage foil shield and a drain wire. Might be worth experimenting with. #1883A is similar, and is specified for use as an audio punchdown cable.
In honesty, though, distance is your best friend in this case. Near-field interference drops off very sharply as a function of distance (I *think* it tends to fall off as the fourth power of distance from the source) and keeping the cables a couple of feet further from your equipment may eliminate the problem or reduce it to the point where it is not harmful. If you have to add a few extra meters of cable to accomplish this, do it - the audio degradation from the extra wire will be negligible.
--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
Not the same situation, but adding this ferrite...
ftp://jjlarkin.lmi.net/Ferrite.JPG
dropped the RF sensitivity of this gadget (a thermocouple signal conditioner) about 30 dB.
My competitor's box, from Oxford Instruments, could be shut down from across the room with an old GR signal generator and a pigtail antenna. Ours was around 30 dB better without the ferrite, 60 with.
You probably don't need them. Computer equipment has many localized switching power supplies running in the 10A range. Even a very solid ground plane will have some RF noise on it. Better computers feed peripheral connectors through ferrite blocks to fix this. Cheap or compact computers may rely on a ferrite cylinder over the cable.
There should be no advantage to putting ferrite beads on cables passing near the computer. If it does, there's probably something broken.
Get the specification for the shield coverage. Musical instrument cables are about 95% copper coverage plus 100% conductive PVC coverage.
Your info about shield coverage reminds me that I once heard someone guess that shielded twisted pair would make good audio cable. Any views on using that?
Twisted pair (whether shielded or not, as the situation dictates) is a great way of reducing noise *IF* you have equipment (and especially at the destination/input end) that has a *balanced* input circuit. If your (still unidentified) equipment has only un- balanced connections, then using twisted pair over such a short distance is likely not worth the effort to try it. Moving the cable away from the noise source is a *much more likely* method of remedying your problem.
It is essentially a theoretical question. I want to avoid later hearing artefacts and/or distortions in the audio due to interference.
I find such audio problems are not always apparent at first, so it's not just a matter of trying it and hearing it. For example, for several months I did not hear the artefacts a particular flash memory recorder I have was adding to its recordings. In this case the problem was bad design rather than interference and it was creating "birdies" at a very low level on one of the stereo channels.
It needed the right audio material and right listening equipment used to make those artefacts apparent.
As I mentioned before, there are times when the cable can't easily take another route so I am interested in the better shielding appraoch which you mention.
Unfortunately, many web sites merrily say that almost any wire will do for a short run of audio cable. I am not so sure!
Not a great plan as it creates a ground loop. OK, it is working against a line level signal so it won't be desperately serious, but it is still technically the wrong thing to do. Making the ground of the signal wire as good as possible is the right thing.
In fact good quality balanced audio cable usually is shielded twisted pair. But not exactly the same as similar computer cable since the requirements are a little different.
Don wrote in news:Xns9BDBE3268DD505D4AM2@69.16.185.250:
That will be an antidote to the audiophools that insist of oxygen free silver or whatever it is they insist on now.
The advice here someone gave on using satellite feed cable is good, it's very well screened. The other thing is to use star network ground schemes for the audio, and to keep digital and analog grounds separate if you can, and where possible, used balanced audio signal lines. Major studios do this, as do many home studios since the practise has been frequently described in Sound On Sound and similar magazines. As they specialise in handling mixed signals from a large range of gear, you can be sure their methods are good. If you do this you'll likely not need to worry if digital and audio cables run side by side, though it's still wise to minimise that, don't tempt fate..
Lacking any additional details, in general I believe you will find that aproach to be fruitless for a situation as you describe. The best shielding and/or filtering known to modern technology won't protect a low-tech piece of consumer equipment from overwhelming EMI at short range. You are seriously underestimating the value of distance in reducing interference. Good luck.
"Mr.T" wrote in news:49cd5dea$0$3253$ snipped-for-privacy@news.optusnet.com.au:
The shielding is often better. If you're running fixed balanced audio lines it might be worth using. As far as I know, the main thing with audio lines is they're made to meet a wide demand for strong and flexible cables with soft coverings.
Given that audio uses a low output impedance into a high input impedance, it doesn't matter what the impedance of the cable is, if it's a balanced line in a twisted pair inside a foil sheath, it's ideal, so long as it's not flexed to damage point. What is NOT a good idea is using audio cables for S/PDIF and other high speed digital transfers.
It is well shielded for its intended purpose (UHF-RF) and it terminates well to RF connectors. But it is nothing special at audio frequencies, and is worse than ordinary audio cable because of its size, stiffness, and poor performance when you try to terminate it with typical audio connectors.
"Richard Crowley" wrote in news: snipped-for-privacy@mid.individual.net:
On the other hand, if you're wiring it in conduit to a stage box where the termination is solder, and the chassis mounted connectors take the strain and adapt to similar connectors in the open attached to durable audio cables, why not? Theatres and clubs have often done it. The cheap foil-wrapped pairs that run a single solid strand of tinned copper along the inside of the foil are ideal for fixed multi-way audio runs. The thinnest paired mic cable of similar ability woult take five to ten times the cross sectioan area. And cost vastly more too.
And a foil sheath really doesn't care if the signal is RF or AF, it's a total faraday cage, what more do you want? I grant that EM induction of signals might be bothersome but it always is, that's why balanced lines matter.
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.