Measurements of speaker cables in frequency and time domain

[Chris M]

We have already used Cat7 installation cables here and it works, some said it would also sound better. Well we know.

I wouldn't call 16 AWG "pretty thin" over any distance less than 25 feet.

Almost everywhere in Europe 1.5 mm2 cable is used in house installations. At 230 Volts in Germany and 16 Ampere standard fuse, that's 3680 W. Wow. That should fit 16 AWG !? May be...

The amplifiers that I saw from the inside have 40 to 70 Volts output on the power pack; if 30 to 60 Volts of it arrive at the speaker terminals ... how big is the current at e.g. an, oh so great, Hypex or Purify device with e.g. 500 W?

Hmmmmmm .... we average and take 50 Volt, then we already have 10 Ampere. (Basis P = U × I, not complex) I have no idea whether the voltage at the loudspeaker output terminals is lower, significantly lower and / or dynamic. Then the maximum possible current would increase more and more.
This would mean that the current load of a rather thin wire or a stranded wire would reach its limits. From a technical point of view, as far as I know, it doesn't do any harm to take 2.5mm2 ~ 13AWG or more ... and you don't get poor either.

Made with Translator

 

[Chris M]

I think it's great to see the cables make a difference. This is also communicated by reputable manufacturers such as Beldon or Lapp and can sometimes be looked up in the technical data. However, this is of little use to the technically interested hifi fan, because the interaction with the loudspeaker can only be guessed at. Who has audio precision or other measuring devices at home? Maybe a Motu M4 or something ... That's why I'm grateful to see these measurements here and look forward to the competent and friendly discussion.

also made with the translator

 

[Ashley Salmond]

If done correctly, the twisted weave cables tend to have lower inductance and higher capacitance compared to single pairs. As long as the equivalent gauge is the same (i.e. DCR is similar), the net result should be LESS deviation in FR. There's an older aes paper that compares cable LCR characterisrics and I think the twisted weave design was included.

 

[pma]

Who has audio precision or other measuring devices at home? Maybe a Motu M4 or something ...

You made a good point and I can make you sure that the Motu is fine. It is rather about understanding of physics and electricity than about fancy instrumentation. One may have the AP but this sole fact is not enough, though it may help in some measurements. Not necessarily in case of cables.

 

[pma]

Note that my Audio Precision analyzer is far more accurate than OP's scope measurements. Digital scopes are made for speed so their dynamic range is quite poor (usually 8 to 10 bit digitizers).

I did not use a digital scope for measurements of FR of the real cables loaded with real speakers, sorry. My measurements of speaker cables are at least as good as yours and on contrary they show real life situations. Measurements of a speaker cable FR loaded with 200k input impedance of the AP is useless. Same applies to attempts to measure “cable distortion”.

And yes, if we need to measure cable reflections, the AP is close to useless and the digital scope with rise time in ns is a perfect tool. This is a different story and different tools are needed than in case of audio band frequency response.

 

[BluesDaddy]

We have already used Cat7 installation cables here and it works, some said it would also sound better. Well we know.



Almost everywhere in Europe 1.5 mm2 cable is used in house installations. At 230 Volts in Germany and 16 Ampere standard fuse, that's 3680 W. Wow. That should fit 16 AWG !? May be...

The amplifiers that I saw from the inside have 40 to 70 Volts output on the power pack; if 30 to 60 Volts of it arrive at the speaker terminals ... how big is the current at e.g. an, oh so great, Hypex or Purify device with e.g. 500 W?

Hmmmmmm .... we average and take 50 Volt, then we already have 10 Ampere. (Basis P = U × I, not complex) I have no idea whether the voltage at the loudspeaker output terminals is lower, significantly lower and / or dynamic. Then the maximum possible current would increase more and more.
This would mean that the current load of a rather thin wire or a stranded wire would reach its limits. From a technical point of view, as far as I know, it doesn't do any harm to take 2.5mm2 ~ 13AWG or more ... and you don't get poor either.

Made with Translator

Speakers draw no where near the amperage of power, by an order of magnitude. This is information you can easily look up.

 

[BluesDaddy]

I think it's great to see the cables make a difference. This is also communicated by reputable manufacturers such as Beldon or Lapp and can sometimes be looked up in the technical data. However, this is of little use to the technically interested hifi fan, because the interaction with the loudspeaker can only be guessed at. Who has audio precision or other measuring devices at home? Maybe a Motu M4 or something ... That's why I'm grateful to see these measurements here and look forward to the competent and friendly discussion.

also made with the translator

You may be drawing the wrong inference from these test results. While cables CAN measure differently and, when not used appropriately in a given circumstance (amplifier, speaker load, distance) - all things being equal, different cables will NOT sound audibly different.

 

[Pdxwayne]

You may be drawing the wrong inference from these test results. While cables CAN measure differently and, when not used appropriately in a given circumstance (amplifier, speaker load, distance) - all things being equal, different cables will NOT sound audibly different.

Hmm, what about this article?
https://benchmarkmedia.com/blogs/ap...KwHM2ZqkqRlb7LVNrQNXnHOC2-OKx3nRoCMagQAvD_BwE

One should be able to find a case with certain amp and speakers combination where a long 20+ ft 16awg vs 8 ft 10awg would indeed causes sound balance difference that IS audible.

 

[BluesDaddy]

Hmm, what about this article?
https://benchmarkmedia.com/blogs/ap...KwHM2ZqkqRlb7LVNrQNXnHOC2-OKx3nRoCMagQAvD_BwE

One should be able to find a case with certain amp and speakers combination where a long 20+ ft 16awg vs 8 ft 10awg would indeed causes sound balance difference that IS audible.

One can ALWAYS find examples on the extreme fringe of things, but that's the point of "all things being equal". The differences in the example you cite would certainly be measurable. As was recently said on another thread, cables can DEGRADE the sound due to improper use, but cables that measure the same will NOT improve the sound and will NOT have an audible sound difference.

 

[Pdxwayne]

One can ALWAYS find examples on the extreme fringe of things, but that's the point of "all things being equal". The differences in the example you cite would certainly be measurable. As was recently said on another thread, cables can DEGRADE the sound due to improper use, but cables that measure the same will NOT improve the sound and will NOT have an audible sound difference.

If using Amir's method of measuring, do you think we will see any "audible" differences between a 20 ft 16awg vs 8 ft 10 awg?

 

[Plcamp]

That Benchmark article says it all IMO. It also links a calculator to take into account your particular amplifier’s damping factor, and your speaker’s min/max actual impedance. It presumes 0.5db delta across the freq range as an audibility criteria, and concludes in its typical case analysis that 10’ of 11awg needed in that case.

 

[BluesDaddy]

If using Amir's method of measuring, do you think we will see any "audible" differences between a 20 ft 16awg vs 8 ft 10 awg?

In what circumstance?

 

[Pdxwayne]

In what circumstance?

Regarding how Amir tests cables.

I suspect his way of measuring will show 20 ft 16awg would measure similar to 8 ft 10 awg cable, where differences found would be so low it is not audible.

Would you expect his measuring method to show any audible difference?

 

[MrPeabody]

If you do the same kind of measuring with an ideal speaker cable that has no reactance, in general you will see the same type of effect. The strength of the effect will depend on the cable's resistance value. If the resistance is adequately large in comparison with the reactance of the speaker, you will see the same type of effect on the voltage at the speaker terminals, for a cable with no reactance.

As such, the measured response curve for a given combination of speaker and cable should be compared to the theoretical response curve for a non-reactive cable in combination with that same speaker. Ideally, this would be done not for just one ideal cable with one particular resistance value, but rather for a variety of cable resistance values that cover a realistic range. Not to suggest that this would be easy, only that without doing it this way, measurements of this sort aren't all that meaningful, because you are left asking, "How great would the DC resistance of a non-reactive cable need to be, in order that I would see this same anomaly in the response? To what extent does the cable reactance contribute to this anomaly?"

Also, if I have interpreted your graph correctly (the first one), the response deviation seems to be about +/- .05 dB. Maybe I have misinterpreted your graph, but if I didn't misinterpret your graph, you are implicitly suggesting that this much response deviation is significant. Isn't there a point where it isn't audible, or should we be concerned about the different response deviations for different cables (used with a given speaker) no matter how mild the response deviations?

In passing I'll also mention, as others have mentioned, that as far as these measurements are concerned, the only effect of the resistor placed across the ends of the cable was to diminish the impedance peaks of the speaker. I have no idea why, for these particular measurements that you took, that you thought it would be appropriate to use that resistor. Also in passing, when you use square waves for something like this, all that you accomplish is to make many people on this forum ignore what you have to say.

 

[BluesDaddy]

Regarding how Amir tests cables.

I suspect his way of measuring will show 20 ft 16awg would measure similar to 8 ft 10 awg cable, where differences found would be so low it is not audible.

Would you expect his measuring method to show any audible difference?

You're still not specifying "audible" under what conditions, "how Amir tests cables" is only relevant to the measurements, not the audibility in a given situation.

 

[Pdxwayne]

You're still not specifying "audible" under what conditions, "how Amir tests cables" is only relevant to the measurements, not the audibility in a given situation.

Did you read the article or try the table downloadable from the article?

What is in "red" at the right most volume of the table is considered audible.

 

[BluesDaddy]

Did you read the article or try the table downloadable from the article?

What is in "red" at the right most volume of the table is considered audible.

Is there a reason you don't want to set the parameters under which you want me to put forward an opinion? Referring back to the article is just shifting the discussion.

 

[Pdxwayne]

Is there a reason you don't want to set the parameters under which you want me to put forward an opinion? Referring back to the article is just shifting the discussion.

You are the one who makes claim. Learn to read the article, then get back to me if you care.

 

[MrPeabody]

One thing that I'm still a bit bothered by is that shortly after Amir posted the review of the speaker cable that started all this, pma wrote this:

The test as was described and posted makes absolutely no sense with regards how to test a speaker cable.

What bugs me is that pma has not explained this. He has followed up with many posts and even started this new thread wherein he explains the proper way to do this, but unless I missed it, he hasn't ever given an honest explanation of what it is that he thinks is wrong with what Amir did. Even after Amir posted in response, at possibly two separate points in the three different threads, explaining that he used the cable to connect a signal source to a real speaker, measured the voltage at both ends of the cable and produced a graph of the difference, I do not think that pma has responded to this, to explain exactly what he thinks is wrong with how Amir took measurements of the speaker cable. pma demonstrated that you get different results with different combinations of cable and speaker, and he showed how you can obfuscate the thing by showing what square waves look like and by putting a 47-ohm resistor across the speaker terminal, but none of this serves as an explanation of what it is exactly that he thought was wrong with the way Amir took those measurements.

 

[pma]

I did. The statement was addressed to the post #1 in the Morrow cable review thread. That review included only measurement of cable FR with AP (40 ohm source impedance and 200 kohm load impedance) and measurement of “cable distortion”, the phenomenon that does not exist. This review then resulted in anecdotal comments without any background. This is what I criticized and I would do it again.