Measured (!) speaker cable difference!?

[Raindog123]

Hi Folks,

Within the latest speaker-cables yack ("yet another kabels conversation") mostly filled with typical fluff, a rather intriguing piece of data was posted. An ASR member @Balle Clorin presented his old FR (frequency response) measurements of various commercial speaker cables.

The measurements are rather interesting and even unique because (a) they were performed with an actual speaker (rather than a resistive dummy-speaker load), and (b) they clearly capture [frequency-dependent] signal variation between measured cables – of 1/4db @50Hz, 1/2dB @300Hz, and 1.0dB @10kHz (in 4m, 12-foot cables).

So, we would like to reopen the conversation and invite everyone to discuss the findings. Eg, concentrating on:
- Confirming/invalidating the results (eg, through own experiments, public data, suggestions for future tests);
- Interpretation of @Balle Clorin's results (eg, hypotheses explaining them vs potential test-setup flaws);
- Audibility of such signal spectra difference, if validated…

@Balle Clorin might be onto something here... or not. So, let’s get to the bottom of it! (And Adam is committed to "assist". )

 

[Balle Clorin]

I measured the frequency response for different cables at the speaker INPUT (i.e. speakers connected). and got response differences. I wish someone could repeat this since the results surprised me..

And with s 8.4ohm resistor at the cable end it looks like this.

Speaker impedence/phase

The yellow arrows below show the frequency response difference at the amplifier end and the speaker end of the cable

 

[Raindog123]

Additional info from @Balle Clorin:

 

[amirm]

I am not clear on the test. Is this measuring the frequency response of the amplifier but at the end of the cable connected to a speaker? If so what speaker?

And do really mean a resistor in parallel with the speaker????

 

[amirm]

Also is the amp class D?

 

[MaxBuck]

The Audio Critic had an interesting feature (I've tried unsuccessfully to find it again) in which the effect of cable impedance (and of that, largely inductance) on amplifier frequency response. I think this was back in the 1980s, so amp design varied from today. Bottom line is that high-inductance cables could have pretty sizeable impact on the FR of the power amp.

 

[Raindog123]

Also is the amp class D?

Some answers are added above as post #3.

 

[Raindog123]

As a reference, Archimago's cable measurements basics...

 

[Balle Clorin]

Amirm
I am not clear on the test. Is this measuring the frequency response of the amplifier but at the end of the cable connected to a speaker? If so what speaker?


Yes at the end of cable at the speakers terminals. The speaker connected was connected as the resistor in the figure
This speaker
TQWT- (troelsgravesen.dk)
And do really mean a resistor in parallel with the speaker????

Yes , to reduce the voltage level to PC/interface I made a voltage divider > 10kohm in parallel with 8ohm speaker will not effect anything,,.

Amp is Class A..push pull 15 watt with 220 000uF powersupply

'

 

[Cbdb2]

You say the amp output impedance is .1 ohm. I don't think it stays there at high freqs. or the plots into a restive load would be flatter.

 

[Raindog123]

You say the amp output impedance is .1 ohm. I don't think it stays there at high freqs. or the plots into a restive load would be flatter.

Yet the “no cable” curve is pretty flat…?

 

[daftcombo]

What puzzles me the most is that rises in FR seem to fall again after 10 kHz or so on the first graph.
Is it really FR or not rather impedance measured?

 

[Cbdb2]

Does anyone else have a problem with a signal being 3db hotter with a cable than without? Somethings not right with these measurements. Lets assume a cable does not add power. Then all those plots at the cable end can not be above the no cable plot. If you fudge the results so the max signal at the spkr is never above the amp out (has to be) then the low freq plots of the different cables will be 3db apart, not buying it.

 

[Balle Clorin]

Yet the “no cable” curve is pretty flat…?

Output impedence is not a constant value, it vaies with frequency...

 

[Balle Clorin]

What puzzles me the most is that rises in FR seem to fall again after 10 kHz or so on the first graph.
Is it really FR or not rather impedance measured?

FR, impedance was some months before and not part of the cable measurement. Impedance measurement also requires a completely different set up.

 

[Balle Clorin]

Does anyone else have a problem with a signal being 3db hotter with a cable than without? Somethings not right with these measurements. Lets assume a cable does not add power. Then all those plots at the cable end can not be above the no cable plot. If you fudge the results so the max signal at the spkr is never above the amp out (has to be) then the low freq plots of the different cables will be 3db apart, not buying it.

Yes , strange things, maybe you can repeat the setup . Cannot recall the exact levels used or if i did something different for the no cable test, it may have been at a different level if it was done not the same day. But the LOAD is different the speaker is not 8.4 ohm as the resistor load is.
Anyone is free to do the same measurement setup and correct any errors in mine..

Maybe a measurement with speaker and at the amplifier end and speaker end tells something. This test was at same levels within the same hour

 

[MakeMineVinyl]

Speakers present a reactive load, and any cable has a reactive component. That this should alter frequency response is not surprising.

 

[Cbdb2]

Yes , strange things, maybe you can repeat the setup . Cannot recall the exact levels used or if i did something different for the no cable test, it may have been at a different level if it was done not the same day. But the LOAD is different the speaker is not 8.4 ohm as the resistor load is.
Anyone is free to do the same measurement setup and correct any errors in mine..

Maybe a measurement with speaker and at the amplifier end and speaker end tells something. This test was at same levels within the same hour
View attachment 147874

At 10k That shows more voltage at the speaker than the amp puts out! Can't happen.

 

[Cbdb2]

Speakers present a reactive load, and any cable has a reactive component. That this should alter frequency response is not surprising.

Agreed but 3db is excessive for low output impedance amp.

 

[MakeMineVinyl]

Agreed but 3db is excessive for low output impedance amp.

Obviously there is a variable at play which has not been addressed. Either the measurement is faulty or there is something going on with the test setup which is not known.