Electromagnetic Interference in Speaker Cables? (video)

[martijn86]

Please don't get caught up in a mud throwing competition Amir. There's no convincing conspiracy theorists.

Personally, I just try to get the optimal thickness for the length of wire and pick a nice style. I did account for the fact that one cable runs alongside a power cable so a foil and teflon shield felt like a sane thing to do. They are visible in the living room so it's worth a little more money to make them look nicer.
If you pick a design with a Chinese vendor, you can message them with all your requests. That way I got a wonderful pair, 2x2m, for €30,-.

I make no claims about audibility but it beats a crumpled red and black CCA string in looks and build quality.

 

[xarkkon]

Let me tell you a short story. Long time ago, I used to have a pair of Dahlquist DQ10 loudspeakers, which had a potentiometer to regulate mid-hi level. With a friend of mine, we listened to some tracks with me tweaking the level until we were both satisfied. Differences were noticeable. Maybe 2 weeks later, I found that the potentiometer wasn't connected on both speakers. But we heard what we heard didn't we ? NO. We heard a difference where we expected to hear one. This is an example of placebo effect.

Biggest differences happen with loudspeakers (macro scale), for the rest it's only micro scale with cables having the smallest influence, assuming they don't have some weard construction as it sometimes happens with highend stuff. And so show you measurements.

i'm sure there's mental placebo effect strongly at play for all. in my case, i didn't know the cables were switched until after the fact, it was just the same song on repeat several times. so no placebo effect.

 

[xarkkon]

What does this have to do with the price of fish (i.e. this video)? I don't think audible changes when dealing with analogue cables/interconnects of different gauge, capacitance etc. is in dispute... there is nothing mystical about that at all.


Why, you want him to show you more bullshit?



JSmith

Even within this thread, there was one person who suggested the audible change I heard was a placebo, which is to be expected given this forum.

As much as I don't agree with Danny's recommendations, I still learn from listening to his arguments and Amir's responses.

 

[JSmith]

As much as I don't agree with Danny's recommendations, I still learn from listening to his arguments and Amir's responses.

Sure, whatever you prefer... but you seem to be trying to conflate simply a change of analogue interconnects with same possibly picking up EMI and comparing audibility of same. What Amir is discussing has nothing to do with what you're referring to, so unsure why you've posted it here.



JSmith

 

[xarkkon]

Sure, whatever you prefer... but you seem to be trying to conflate simply a change of analogue interconnects with same possibly picking up EMI and comparing audibility of same. What Amir is discussing has nothing to do with what you're referring to, so unsure why you've posted it here.



JSmith

because danny's series of videos is about cables mattering to justify his cables.

 

[pma]

Amir doesn't discuss at all, the orientation of the cables to one another (merely winds a random coil on his hand using both cables).

Nobody (not Amir or Danny) in either video, show the actual ability of that lovely tubular/helical woven speaker cable in various orientations/stretch to reduce or reject adjacent EM coupling or interference. Surely, that is what this entire exchange is about?

Hi John,
yes you are basically right and you know well that we have had Faraday induction law here for almost 200 years
https://en.wikipedia.org/wiki/Faraday's_law_of_induction

and that the induced voltage is proportional to derivative of magnetic induction and to the geometric area between the two wires (shorted at one end of course)

so it may be quite easily calculated.
I made a practical experiment of the 2m speaker wire 2x2.5mm2 shorted at the end and laying along the power amplifier power cord, the amp is 2x100W class AB with about 50VA idle power. Switching the amp ON-OFF makes almost no difference and the highest induced peak loop voltage is below -120dBV, so absolutely inaudible. The only way to get some harmful voltage is to put the power cord inside the loop created by the 2 speaker wires, however I do not think anyone would do it.

 

[Thomas savage]

Cable threads are like dog mess on a path , no one walks into it and gets away clean . .

Being disciplined and keeping a vigilant lookout to avoid stepping into one is key to a happy day .

 

[solderdude]

Danny simply made an air-core transformer. Something competely different as 2 cables (with 2 wires having differential currents) being wound as a coil.

Running cables in parallel that both have 2 wires each with currents and voltages in opposite direction is something entirely different.

The issue in practice would not be inductance or capacitive coupling (too little capacitance for mains frequencies and its audible harmonics) but as mentioned before; reception of RF.
That would be a common mode thing so that signal would be over both wires where a speaker signal would have currents in both directions.

 

[b1daly]

Even within this thread, there was one person who suggested the audible change I heard was a placebo, which is to be expected given this forum.

As much as I don't agree with Danny's recommendations, I still learn from listening to his arguments and Amir's responses.

Every playback of the same exact audio signal will sound different because our auditory perception is constantly changing. Both at the level of attention and in the involuntary level.

There is a simple test which will prove how profound these effects are.

Find a playback system with a tone control. Listen to 30 seconds of a track. Then crank the bass and listen for 30s. Now put the tone control back to flat and listen. It will sound like the bass is gone, as if someone actually lowered the bass.

Drawing conclusions about the audible effects of equipment needs to be done very carefully to be valid.

If you are not purporting that some piece of gear is awesome and just want to be happy with your setup it doesn’t really matter.

 

[mk1classic]

Several feet? Where would I read the high-end speaker wire manufacturers saying that the scenario for their cables?

There is nothing real-world about that. AC wire coming out of an amp has no tendency to run parallel to speaker wire with zero distance for several feet. And if that is the scenario, then don't run it several feet that way.

Let's remember that I am trying to debunk Danny's test where the exposure was the diameter of the coils which was just an inch or two, not several feet.

A side note. Maybe you should invite Dave Rat over for some testing with real equipment...
He does simplified tests on mile of cables in series coiled up. This is how they do it in PA world

 

[Lambda]

As I watched the video, the thought that I had is that I would have taken a somewhat different fundamental approach (not with the measurements per se). I would start by working out the minimal voltage required by a typical speaker in order to produce sound pressure that humans can hear. I would use this voltage threshold to assess whether the amount of noise picked up in speaker wires will be audible.

At the reference voltage of 2.83 V, SPL for a typical speaker will be about 85 dB. This information is found in most any ordinary plot of sensitivity vs. frequency (i.e., “frequency response”) for most any speaker that anyone has measured in the past half-century or more. It is also available in the spec sheet for any off-the-shelf driver. The 85 dB value is typical, and although there are notable exceptions, the great majority of commercially available speakers are within a few decibels of this value. What this means is that at this reference voltage, 2.83 V rms, the sound pressure from a typical speaker will be +85 dB relative to the threshold of audibility, for a young person with perfectly normal young-person hearing, and at frequency near the peak of our hearing sensitivity. (It means this partly because the standard reference value for sound pressure in decibels, which determines the 0 dB point, was chosen to be 20 uPa, which value had been deemed the threshold of audibility for sound pressure.)

Now we need to figure out the voltage corresponding to 0 dB, given that 2.83 V corresponds to 85 dB. In other words: what voltage is -85 dB relative to 2.83 V? We use this equation: -85 = 20 x LOG(V/2.83). Solving this equation for V, we have V = 2.83 x 10^(-85/20). The answer is 159 uV. Thus, in order for a typical loudspeaker to produce a sound that a young person with perfectly good hearing can hear, the RMS voltage presented to the speaker needs to be around 160 uV at least. You could probably be generous and call it 150 uV, but there isn’t any readily identifiable reason to make it lower than than this.

That’s a good start, but the sensitivity of our hearing is about -35 dB at 60 Hz compared to 1 kHz, in the peak-sensitivity range. This means that the voltage presented to the speaker at 60 Hz needs to be +35 dB relative to the voltage presented at 1 kHz, in order for the 60 Hz hum to be at or above the threshold of audibility. The equation we now want to use is: +35 = 20 x LOG(V/160E-6). Solving for V, we have V = 160E-6 x 10^(35/20). The answer is 9 mV. At 60 Hz, the signal to a typical speaker needs to be at least 9 mV in order for the 60 Hz hum to be loud enough to be heard by a young person with good hearing in ideal, quiet conditions.

Now comes these questions: What voltage did Amir measure? Did he take the measurement in a way that was realistic? The second question is difficult, and I’ll leave it for others to answer. As for the first question, the display on his analyzer indicated -130 db referenced to 4 Vrms. We can calculate the voltage by starting with this: -130 = 20 x LOG(V/4). Solving for V, we get this: V = 4 x 10^(-130/20). The answer is 1.3 uV (microvolt, or 1.3E-6 V). [There is another way to solve this, making use of the fact that each increment of +/- 6.02 dB corresponds to a doubling or halving of voltage. Since -130/6.02 = -21.6, the measured voltage is 4V / (2^21.6) = 1.3 uV.]

It is useful to use decibels to compare the measured value (1.3 uV) to the 9 mV threshold value. This: 20 x LOG(.0013 mV / 9 mV) = 20 x LOG(.00013). The voltage that Amir measured is -77 dB relative to the voltage for which a 60 Hz signal played through a typical speaker will just barely be audible.

*

Excellent addition for this thread. Thanks!

Looking at Amirs test with 140W load and 6ich cable we see -128dB peaks and -130dB Up at ~300Hz
Why? Because the current is not alt all 60Hz sinewave and for a Transformer it can look something like this:

At 300Hz the threshold is not +35dB but +15dB therefore i would allow myself to subtract 20dB from your excellent calculations bringing us -57dB before Before audible threshold.

With exceptionally sensitive speakers in the room corners this might be -50dB?
I would assume doubling the load as well es the cable length would add about each 6dB

This brings up the questions of what is needed in terms of cable length and power assuming bes/worest case contitions and the data from Amirs Test.

So increasing the power to resalable 2000W gives us log2(2000/140)=3,83 "doubling" = 23dB
increasing the the cable form 6inces to 2m gives us log2(78.7/6)=3,7=22dB

Giving us +50dB with a 2m cable and 2000W load
Strange enough this adds up 0dB ( i swear it came out to this after i made up the estimated values)

2m cable and 2000W load amirs test conditions and best case high sensitivity speakers it might become audible.

 

[MrPeabody]

The question being considered here is precisely and properly whether the voltage that is likely to be picked up in speaker wire is great enough such that a typical speaker with typical voltage sensitivity would be able to produce sound at or above the threshold of audibility. It is manifest that speaker sensitivity plays a pivotal role in this fundamental question. Yet, it is difficult to find, either in Amir's video or in most of the comments posted here, any clear insinuation that the amount of sound you get from a speaker in relation to the voltage you feed it has anything to do with the question. If the goal is to enable Joe Audiphool to understand when and why noise picked up in speaker wiring would matter, the only good way to this end is to make it mainly about the fact that voltages below a specific threshold are too weak for a speaker to produce sound at or above the threshold of audibility.

In order for a typical speaker to produce sound pressure at or above the 1 kHz threshold of audibility, the voltage threshold is in the ballpark of 150 uV, more or less depending on the sensitivity of the individual speaker.

In order for a typical speaker to produce sound pressure at or above the 60 Hz threshold of audibility, the voltage threshold is in the ballpark of 10 mV, more or less depending on the sensitivity of the individual speaker (at 60 Hz).

The interference voltage that Amir produced and measured is at -77 dB relative to the threshold for audibility (at 60 Hz) for a typical speaker. In order for the exercise being undertaken here to be usefully productive, it is highly desirable for the voltage Amir measured to be put into this particular kind of perspective, i.e., expressed relative to the voltage corresponding (for a typical speaker) to the threshold of audibility.

Should anyone desire to know how I obtained any of these numbers, the answer may be found in my prior post in this thread, a couple of pages further back in time and/or space.

 

[Lambda]

60 Hz threshold of audibility, the voltage threshold is in the ballpark of 10 mV, more or less depending on the sensitivity of the individual speaker (at 60 Hz).

Looking at Amirs test with 140W load and 6ich cable we see -128dB peaks and -130dB Up at ~300Hz

But the over tones have about the same level but a much higher frequency therefore they appear louder.

At 300Hz the threshold is not +35dB but +15dB therefore i would allow myself to subtract 20dB from your excellent calculations bringing us -57dB before Before audible threshold.

The question i'm trying to answer is how long can the cable be before its possible to hear something under worst case conditions.

Base on your Assumptions and Amirs test i estimated

2m cable and 2000W load

But if you insist on -77dB and not taking the overtones into account then its more like 16m cable and 4000W

I give Joe Audiphool the benefit of having super sensitive speakers and he is probably also putting his ear right next to the speaker to check if there is noise.

 

[MrPeabody]

... My main issue with Danny's videos so far is that they have not shown me how the pricier cable is better than the cheaper cable. He has shown basic scientific effects and extrapolated their effects into justifying cables. I'm open to his videos as I do want to see the science from all angles, and I'm hoping he can show more. ...

I would rather that he does not show more, because his regular practice is to falsely interpret the significance of the stuff that is genuine. For example, I recall that not too long ago he connected some cables of some sort to the antenna input of some tuner or receiver and demonstrated that they varied in their ability to pick up noise (or possibly genuine radio transmission), then from this he extrapolated (your word) the results to a conclusion that was patently bogus. This is the kind of thing he does routinely. This is the sort of thing that is all too common among audio tricksters. They start with something that is technically correct, but then twist it around in a way that isn't logically coherent, to argue something that is patently false. Danny does this in such a routine manner that it is ill-advised to pay any attention at all to any of his claims. Another practice of his is the practice of modifying the crossover of some popular off-the-shelf speaker, changing the sound in a way that is audible and possibly more accurate though not necessarily so, and then claiming, with no justification at all, that the apparent improvement in the sound (which isn't even necessarily an improvement) is attributable to the very costly crossover components that are unlikely to sound different from components that cost only a fraction as much. There is a thread here on ASR where this was discussed in some detail. Bottom line, it isn't wise to trust anything that he says. His YouTube followers are to be pitied.

 

[MrPeabody]

But the over tones have about the same level but a much higher frequency therefore they appear louder.



The question i'm trying to answer is how long can the cable be before its possible to hear something under worst case conditions.

Base on your Assumptions and Amirs test i estimated



But if you insist on -77dB and not taking the overtones into account then its more like 16m cable and 4000W

I give Joe Audiphool the benefit of having super sensitive speakers and he is probably also putting his ear right next to the speaker to check if there is noise.

I apologize, I had not intended to be responding to what you wrote in the post prior to mine above, it was just timing. I did read your prior post but didn't follow it all. I will give it another read, but one comment I'll make now is that the -77 dB is specifically for the 60 Hz interference relative to the threshold of hearing at 60 Hz and assuming a speaker with typical sensitivity. I really wanted to emphasize the point about the speaker sensitivity not having been given the proper role in the treatment of the question, and I now fear that this point has again been obfuscated to where it was a waste of time for me to have tried. But sometime later today I'll take another read of both of your posts here and hopefully I'll be able to follow it.

 

[Teeter]

Am I okay with 16ga, 2cond twisted pair, 99.9%pure copper, stranded wire, Shielded, and jacketed??? lol knock yourselves out

 

[redstang]

...
There is nothing real-world about that. AC wire coming out of an amp has no tendency to run parallel to speaker wire with zero distance for several feet. And if that is the scenario, then don't run it several feet that way.
...

There most certainly is. Just look at the photos of those cable runs through those cable management trim panels. Someone could very easily be running a power cord through there. Right now, in my system, the power cord for my subwoofer is running along my left channel speaker cable for about 2 feet. Do I hear anything? No (but TBH I've never scrutinized it either). And his argument just isn't about power cords is it? It's that just the speaker wires themselves can create induction. Regardless, to me, the point is to prove that the effect isn't audible much less even measurable, right? It's very plausible that someone might run the power cord for their sub along many feet of speaker cable, and it is VERY common to run many feet of multiple speaker cables and/or HDMI and line level sub cables all together.

And I agree - if that causes a problem, simply don't do it, but I was hoping you would show if it is even possible for that to measured, much less be an audible problem. That's why seeing a measurement on your gear would be interesting.

...
Let's remember that I am trying to debunk Danny's test where the exposure was the diameter of the coils which was just an inch or two, not several feet.
...

Yes, while the point of your video was to debunk his test, I guess I'm also looking to just put the entire argument to bed. His overall but convoluted intent is to say that induction, regardless of how it gets created, causes audible problems. So he shows us an example of induction, and then implies that then it is also happening to your plain speaker wire in normal use cases (classic "Audiophile logic"). So while the logic of his test is horribly flawed, the question of whether induction can be created by parallel wires and then be audible is what really needs to be investigated/debunked.

He's using that same flawed logic in his FM "test". His overall method in these videos is to show that a certain kind of interference can happen with cables/wire, and then make the huge leap that therefore it HAS to be having a negative effect on your audio - while never showing that that interference a) can even happen under a normal use case and b) is actually audible. Cable manufacturers are doing exactly that - showing examples of real world interference that can appear on wires to get you to believe that this interference, even at unmeasurable levels, is somehow bringing up the noise floor of your system and that you, as an audiophile, will be able to hear it when you buy our super-duper cable to filter out that noise.

 

[redstang]

Come to think of it, I have first hand experience with a much worse scenario than this.

I've installed the current speaker system at the gym that I work out at. The speaker cables have varying lengths between 20 and 30 meters. All the cables are laid out alongside the power cables for the lighting in the gym. That lighting draws about 5kw of power, and there are three long rows of lighting, so let's say that there is 1,7kw running along each line.

Now, what does 30m of speaker cable laid alongside 1,7kw of lighting sound like?

Nothing. Nothing at all.

That's an excellent anecdote, but it would be immediately dismissed by a cable believer because they're going to say that a) whatever speakers are in that gym aren't good enough to hear the difference, and b) the acoustics in a gym are terrible so you're never going to hear the subtle nuances their nice cables provide. They will then proceed to mock that example because of how irrelevant it is to them.

Their arguments have become somewhat predictable to me. I've spent too much time in these cable debate threads.

 

[fmplayer]

I've seen so many crazy anti-blind test arguments. Yep, you have the stress angle. Then you have the simple statement that a blind test simply isn't a normal listening situation - that when you listen over the course of longer times you can tell that things sound different/better after changing cables.

Stress like being in darkness?
Want a true blind test with no additional stress to the listeners? Ask some blind people. They rely completely on their hearing which is exacerbated compared to ours. If there's something to be heard, they will.

Another huge one is that if you can't hear a difference, then your system isn't "resolving" enough (i.e. expensive enough).

Or maybe there's no difference to be heard... Who says there is or must be some difference? What theory ? Seriously, what makes the system resolving ? price ? snake oil ?

Then there's also the one where "the entire system needs to acclimate" to the cables and settle in. Sort of a burn-in type of argument that somehow it takes some time (hours to days) for the amplifier and cables to come into a harmonious relationship, so fast switching cables doesn't work. It always amazes me the theories people come up with. I think that's part of the reason why this debate can be so entertaining.

... and amps and cables started to take their hands and fell in love with each other ? ... and after some time there were little babies called idio(t)philes. Obviously, if you switch too fast, they don't have enough time to fall in love. QED lol

 

[xarkkon]

Every playback of the same exact audio signal will sound different because our auditory perception is constantly changing. Both at the level of attention and in the involuntary level.

There is a simple test which will prove how profound these effects are.

Find a playback system with a tone control. Listen to 30 seconds of a track. Then crank the bass and listen for 30s. Now put the tone control back to flat and listen. It will sound like the bass is gone, as if someone actually lowered the bass.

Drawing conclusions about the audible effects of equipment needs to be done very carefully to be valid.

If you are not purporting that some piece of gear is awesome and just want to be happy with your setup it doesn’t really matter.

i use cheapo cables that came with the gear and am super happy with my cheapo cables. i think you're probably addressing the placebo point in my post.

I was only addressing someone else that said everyone here knows basic cables structure (gauge, etc) cause audible changes. As you've shown, many here think that all changes are entirely placebo, which is a perfectly fine position to take.