Does making distortion measurement of cable make sense?

Status
Not open for further replies.
You WILL measure a difference at cable ends, (even with a linear load) , simply because you are creating a voltage divider.
We were interested in whether we could HEAR a difference. Hence the DBLTs. Level matching in DBLTs deserve a paper in itself and we say a bit in the AES papers.

Plus slight influence from Sellotape dielectric properties, of course, if you want to nitpick.
As we didn't investigate different insulators for our Lightning Conductor, I can't really comment on this :) In case it isn't obvious to da pedants "Joke Mr Didden. Joke" But the principle remains the same. If I didn't actually check this with DBLTs, I can't comment.

We DID investigate the effect of Tetleys Bitter on the performance our DBLT panel and I think we mention this in one or our AES papers :cool:
_____________________________
But I see I am in the wrong thread.

I assumed the 'question' was if you could hear a difference between cables. The answer is 'yes' and I showed a documented method to investigate this reliably with DBLTs.

I also pointed out one cable which was nearly 'perfect'; at least in our DBLTs. But this doesn't seem to be of interest. :mad:
 
Last edited:
I am not sure skin effect is a non-linear phenomenon. Skin effect is a frequency dependent and not an amplitude dependent. Think of it as a low pass filter. If you input a signal to a low pass filter, the output is still linear (although the amplitude may change).

Now as for can you hear the difference because of skin effect? I am not sure.
 
AX tech editor
Joined 2002
Paid Member
I assumed the 'question' was if you could hear a difference between cables. The answer is 'yes' and I showed a documented method to investigate this reliably with DBLTs.

Allow me to pick my nit.
I believe the question was about distortion in cables, and I assume harmonic distortion was meant.

Audible difference between cables is not the same as distortion in cables.

Jan
 
Maybe someone wants to correct this Wikipedia article!
:D


" The effect of speaker wire upon the signal it carries has been a much-debated topic in the audiophile and high fidelity worlds. The accuracy of many advertising claims on these points has been disputed by expert engineers who emphasize that simple electrical resistance is by far the most important characteristic of speaker wire."

Speaker wire - Wikipedia
 
OK, so here's a real example from consumer electronics. There is a device where an extraneous signal added by the device measures at around 0.001% or -60dB. The innate distortion of the device is ~1%, so the extra 0.001% making 1.001% total is obviously inaudible, right?

The component in question is a cassette deck and the extraneous signal is tape noise.
The answer isn't as straight I think. The tape noise is white noise so it can be partially filtered out by our brain.
 
By definition the difference between the input and output of a system or device is distortion. If the output of a system is not a facsimile of the input, it is distorted.

Harmonic distortion is a sub-set of distortion, and not a particularly good measure of the audibility of distortion.

I think it depends on how you define "distortion". The strict definition is if you put in one frequency and you get the same frequency on the output then there is no distortion. Even if the amplitude is changed, but as long as you only have the same frequency on the output, then there is no distortion.

For example, if you have a pure resistor, then it won't have any distortion even though the output amplitude may change at the output.
 
I don't think skin effect can cause non-linear harmonic distortion. But it can cause linear frequency response/phase response distortion.
+1.
I have to add that if we are talking about 0.001% THD levels then I have to say that copper wires are "non-linear" when they conduct a current. And the larger the current - the easier it is to measure its non-linearity. The reason is TCR. And there is a bit deeper effect - the TCR is temperature-dependent as I know too.
 
Last edited:
By definition the difference between the input and output of a system or device is distortion. If the output of a system is not a facsimile of the input, it is distorted.
Not at all.
If difference is only in level but not in waveform or harmonic content, it is NOT distortion but linear gain or attenuation.

By your flawed "definition" :rolleyes: then ALL amplifiers, preamps, pots, attenuators, etc. DISTORT.

Not sure you are just naturally dull or simply trolling us.

Suspect the second.
 
Any argument that you can't hear a signal aberration, AKA distortion, introduced by system component, e.g. a cable, because it is only 0.01% of the total signal is demonstrably spurious.

When you listen to a jazz recording, for example, it is not uncommon for the high frequency components like breath, shimmer and sibilance to be more than -60dB relative to the fundamental bass tones, yet they are clearly audible and hugely defining to the sense of quality of the recording. Try it: load your favourite FFT app and look at the frequency content of real music.

Why have audio systems with a dynamic range of >120dB if components >-40dB down can't be perceived, as posted a while back? I really wonder if people think through what they are posting when they make the bombastic "it's impossible to hear such and such..." statements.
 
+1.
I have to add that if we are talking about 0.001% THD levels then I have to say that copper wires are "non-linear" when they conduct a current. And the larger the current - the easier it is to measure its non-linearity. The reason is TCR. And there is a bit deeper effect - the TCR is temperature-dependent as I know too.

"we are talking about 0.001% THD levels then I have to say that copper wires are "non-linear" when they conduct a current


Then E is NOT = I * R

You have re-written physics.

The thermal mass of copper is great enough that the current through a conductor used for audio purposes would have insignificant thermal heating effects. It would be most likely in the low ppm range, not hundredths of a percent range. Totally insignificant for audio purposes.
 
Last edited:
I am not sure skin effect is a non-linear phenomenon. Skin effect is a frequency dependent and not an amplitude dependent. Think of it as a low pass filter. If you input a signal to a low pass filter, the output is still linear (although the amplitude may change).

Now as for can you hear the difference because of skin effect? I am not sure.
Skin effect will increasingly attenuate at high frequency in audio application and tops at about 0.5 db at 20KHz. You should be able to figure out if the effect will be audible to you or anyone when playing music.
 
When you listen to a jazz recording, for example, it is not uncommon for the high frequency components like breath, shimmer and sibilance to be more than -60dB relative to the fundamental bass tones, yet they are clearly audible and hugely defining to the sense of quality of the recording. Try it: load your favourite FFT app and look at the frequency content of real music.

Test yourself, then please report back your result Audio DiffMaker example files last item in the table. Note, Bill Waslo is a distinguished member here.

That would make your opinions credible. I’m not holding my breath.
 
Then E is NOT = I * R

You have re-written physics.
No need to re-write physics.

No real cable made out of wire and insulation is just an idealised 'R', even if the conductor part itself is.

When the entire equivalent network of a source / cable / load network with the output stage circuit 's complex impedance, the cable's linear and nonlinear imperfections in dielectric and permeability properties, and the load circuit's complex input impedance are all analysed together, the transfer function of the cable and the reasons why cables sound different can be derived adequately by known physical laws, without any voodoo or new laws of physics.
 
Status
Not open for further replies.