Analysis of speaker cables

[billshurv]

And the resonances I am talking about don't have much detail, but they envelop you in a live environment. If you have ever listened to a choir in a decent size chapel or Cathederal you will understand the effect. Almost impossible to create at home from a 2 channel source, but need a challenge.

[Markw4]

Sure, I get it. Some of the music I also want to be able to reproduce well has complex synthetic ambience. That type of decay can be more detailed, depending. Still working on it. Its a challenge for sure.

[Evenharmonics]

Originally Posted by scottjoplin

I'm pretty sure that those are subjective statements of preference, ie liking, so no one will blame you for ignoring them too.

They claim x, y, z and when challenged, some do realize that those are just subjective observation and lack the objectivity such as bias control, some will double down on their claims as if their income depends on it. Then there are couple other types but I won't bore you with details since they are minority in numbers.

[bucks bunny]

And there are some that respondl to postings different to their dubious opinions with insults. Revealing a profound lack of reasons and style.

[EddieT]

I just use el cheapo 18ga extension cords and clip the ends off. They are cheaper and faster than buying lamp cord at the big box store. I figure any compromise in speaker cable is infinitesimal in comparison to speaker compromises I have already made.

[vacuphile]

It is most certainly not cables that bind us :-)

[Hans Polak]

Suddenly I realised that when comparing the step response for the Townshend cable with 1 segment versus 10 segments, I should have reduced all components with a factor 10 to end with the same LRC for both.
What I did was adding 10 equal segments, meaning effectively that I made the cable 10 times as long.
For the Char Imp this makes no difference, but for a step response it does.

In the image below I have corrected this, where I still have to mention that I can't find whether the specs that Townshend provides are per foot, per meter or for their shortest cable which is 2 meter.
Anyhow, with their figures you see the step response for resp. 4, 8 and 16 Ohm for one segment and for 10 segments, where you see a faster settling time but also a larger attenuation as with one segment.

Now I did the same for the earlier mentioned Nordost Valhalla cable, also for 4, 8 and 16 Ohm
If settling time being an important feature for an LS cable, this cable behaves excellent.
Apart from somewhat more attenuation, the version with 10 segments responds almost identical to the 1 segment version.
At audio frequencies the step response is a first order low pas filter determined by the R and the L, where the C plays no role.

Hans

[stvnharr]

That is a nice step response graph for the Nordost. You may be the only person on the whole forum with Nordost speaker cables.

[Hans Polak]

That is a nice step response graph for the Nordost. You may be the only person on the whole forum with Nordost speaker cables.

What a nice idea, I would love to be able to get the opportunity to try them once, but unfortunately no, I don’t have them.
It was just to show that Char Imp is meaningless at audio frequencies because the wavelength is way larger as the cable length and therefore it cannot even behave like one.
So Townshend’s claim is not at all relevant.
But I won’t claim to have solved the question why cables may sound different.
It’s still simply black magic.

Hans

[Elvee]

Originally Posted by Hans Polak

What I did was adding 10 equal segments, meaning effectively that I made the cable 10 times as long.
For the Char Imp this makes no difference, but for a step response it does.

It does make a difference for the characteristic impedance: although you can stack any number of elementary, identical LPFs and get a consistent Zo (which surprised me as a small miracle), you cannot change the lump size and still get an identical result. Miracles have limits.
I showed it in one of the previous sims I posted.
For audio, it is purely academic of course, but it matters from a theoretical POV.
The green trace is the impedance of the lumped version; the red trace shows the values split in 12 segments.
The theory of TL applies when the elements are infinitesimally small, but it diverges with discrete, lumped elements.