Did that expert reference provide any measurements of the amount of such feedback from a reasonably typical environment and how effectively shielded loudspeaker cables improved this?
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Shielded speakers cables
- Thread starter sbronf
- Start date
You can check out the AES site or any of numerous audio club sites at which he has does presentations, or look at his publications page and scroll down. He is well-known and highly regarded for his EMC work among other things. Don't know if it would answer your exact question.
The problem of RFI leaking into the speaker output terminals and being coupled to the input through the internal feedback network has been a problem for decades. How bad depends upon the consumer's environment as well as component sensitivity to RFI. A lot of times it is fairly insidious, e.g. just a little added buzz or hash from a dimmer, fluorescent/LED light power supply, etc. so not always an obvious noise source like a nearby radio transmitter. We used to live near an airport and the ground control radar would inject a noise burst into some components every time it pointed our way; that was fun to track down.
IME shielded speaker cables did little to nothing to help reduce RFI, but it was a long time ago when I dealt with it, though have helped others and read threads recently indicating it is still an issue for some folk. No idea how many or what percentage though it is not an uncommon topic on various forums, don't really care, it's a problem to them. I generally used a snubber filter (series RC circuit) to prevent loading the amplifier's output in the audio band.
Last edited:
Sorry for being late to the party and, I confess, I am well and truly confused. I have recently invested in some shielded speaker cable that I have yet to terminate. Most say not to connect the conductors to shield, some say conduct negative conductors to shield, some say cross connect the internal shields and conductors to the outer shield (mind boggles), some say the shield should connect to an earth on the pre amp, but no one seems to have a definitive answer.
To set the scene, my power cables are circa 3-4ft shielded at one end, with no hum/buzz. Interconnects are shielded XLR's. Speaker runs will be 3 & 4mtr respectively. I typically have no issues with any sort of humming or buzzing from speakers/amps. There is a low level hum from the power amps that does not transfer to the speakers so it's not a ground loop. The power amps are 700W monoblocks with dual hi power toroidal transformers, so I suspect it's just in their nature to hum little. No signal or power cables run in parallel to each other for any notable distance.
In the opinion of folks far more learned than I, should I bother connecting the shield on the cable at any end given my setup.
Unless you have severe RF interference, shielded speaker cables are unlikely to do anything sonically, and increase the risk of accidental shorts. As for the connections, much depends upon the amplifier topology. Some amplifiers cannot handle a grounded output terminal (either terminal, plus or minus), others may not mind but it may not be optimal for shielding, and some may connect the (-) output directly to ground same as any shield. If the amplifier's speaker output terminals are floating, connecting the shield to the (-) terminal may actually make noise coupling worse by providing an additional "antenna" for noise. So without knowing the amplifier's output topology nobody can provide a definitive answer.
The safest approach, assuming you actually need to shield the speaker cables, is to connect the cable shield to the amplifier's chassis ground and connect the speakers to the internal wires, leaving the shield open at the speaker end. This will provide a path for coupled noise to chassis ground and away from the outputs, but the open end acts as an antenna that can actually increase noise coupling at certain frequencies related to cable length and characteristics. The chassis connection should be made with a very short, very thick (preferably braid -- I have often used solder wick * ) wire.
Toroids tend to hum more than conventional transformers because there is no air gap in the core, making them more susceptible to DC (or DC-like) signal content that can overload the transformer's core. You could try a DC block like Emotiva's CMXn or similar, but it could also be mechanical vibration unrelated to the incoming AC power.
* Like this, many options available: https://www.amazon.com/Lesnow-No-Cl...8-1-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&th=1
Or use a zorbel network for radio frequencies at one or both ends of the cable.
100 ohms and .01 UF works.
Thank you. I have the amps connected to a DC blocker that can handle up to 32amps so all good.
I had a feeling not to connect the conductors to ground. As regards the output topology of the amps, all I can tell you is that the outputs are BRIDGED and should NOT be grounded on either speaker output terminal as both red and black speaker terminals are effectively “hot”, i.e. carry signal.
Speedskater
Major Contributor
Cable pickup of Radio frequency interference is a very situation specific type of problem. You need:
a) a source of the interference.
b) a cable of a length and placement to act as antenna.
c) an audio component susceptible to interference at that frequency.
Many if not most interference sources are not 24 by 7. So does the noise level change from time to time?
a) a source of the interference.
b) a cable of a length and placement to act as antenna.
c) an audio component susceptible to interference at that frequency.
Many if not most interference sources are not 24 by 7. So does the noise level change from time to time?
DSJR
Major Contributor
I have nightmares of 1980s Naim amps which had no output protection, needed their higher-than-normal-inductance cabling over several metres each ideally, and basically uncrseened interconnects between pre, supply and three core mains cables to the power amps (DIN one end and XLR the other and wired L, R, and Return). Not sure how much has changed today, although some of their smaller amps acquired inductors on the output and some of the posher interconnect cabling now costs a LOT of money. I discovered a lot of the sonic properties of said amp systems was actually in the pre to supply/power-amp interconnect, which subjectively widened a central image and contracted extreme L-R information, this confirmed when swapping back to a proper twin-screened interconnect equivalent.
To those lurkers really new to this stuff, there's NOTHING WRONG with a simple stranded copper (not copper plated aluminium) fig-8 cable of around 2.5mm conductor cross section (not total external diameter as Linn seem to put in their specs I noticed recently) for most runs up to several metres each (maybe a 4mm type for longer lengths or more powerful systems). Higher price is NOT an indicator of quality here, as snakeoil and very high profit margins rapidly comes into play as regards dealer-purchased cables - and I have first hand experience of this!
To those lurkers really new to this stuff, there's NOTHING WRONG with a simple stranded copper (not copper plated aluminium) fig-8 cable of around 2.5mm conductor cross section (not total external diameter as Linn seem to put in their specs I noticed recently) for most runs up to several metres each (maybe a 4mm type for longer lengths or more powerful systems). Higher price is NOT an indicator of quality here, as snakeoil and very high profit margins rapidly comes into play as regards dealer-purchased cables - and I have first hand experience of this!
Said new cables are 6mm cross section and I buy bulk and complete them myself.
Again, on the shielding topic, any reason to shield on one end or both of a short run (1.5mtrs) of power cable? Again, I have no noise in my system bar small amount of toroidal hum on the power amps.
DSJR
Major Contributor
I'd not bother...
Which most amps already have right before the output.
Yes, but if you want to make sure no RF is getting back into the feedback network (directly connected to the input devices) then a small additional zorbel hurts nothing but makes sure of zero RF getting back in from your antenna (speaker wires). Not necessary in most cases, but if you live close to a radio transmitter could help.
Similar threads
