Warszawa

The sound of speaker cables

Recommended Posts

3 hours ago, Nopiano said:

‘Thus, at audio frequencies, a cable less than 2,000 ft long is no more complicated than its series resistance and parallel capacitance. As the cable becomes longer, or as frequency increases, the cable will begin to behave as a transmission line.’From http://audiosystemsgroup.com/TransLines-LowFreq.pdf

This is a great old paper by Audio Engineering Society, Fellow and interference expert Jim Brown.  But that is in reference to very long balanced interconnects.

Share this post


Link to post
Share on other sites
Super Wammer
2 minutes ago, Speedskater said:

This is a great old paper by Audio Engineering Society, Fellow and interference expert Jim Brown.  But that is in reference to very long balanced interconnects.

So do you think it’s not applicable to speaker cables, given the orders of magnitude here?  If you can find something more explicitly related to speaker cable that would be great.
 

I don’t know where the figures of hundreds of Ohms come from in the Townshend paper, when most domestic speaker cables barely reach 0.1 ohms in normal lengths, and often much less. 

Share this post


Link to post
Share on other sites
3 hours ago, rabski said:

The characteristic impedance of a speaker is as much down to the drive units as to the crossover and there are also effects resulting from the electrical reaction of drivers. None of this can be modelled in a dummy load and every speaker is different. Without accounting for that, a blunt instrument method of measuring is inherently flawed. It proves that one specific property can be measurably different in a laboratory situation, but that cannot be extrapolated to real-world conditions.

Skin effect? No idea other than theoretical. The theory of 'it cannot have an effect at audio frequencies' has been questioned, and also for other effects. Once we understood the actual way electricity flows through a wire, it became clear that things like skin effect and dialectrics can and do have an influence at any frequency. As before, however, the magnitude is another thing entirely.

Nobody has yet looked at the issue with the scientific rigour it would need, because it is far more complex than it appears at first. There is a great deal of published work about all sorts of cable and dialectric properties at single AC frequencies, and especially high frequencies, as there are obvious large-scale commercial interests for power distribution, radio transmission and the electronics industry. The problem with audio is that it is a signal with a near infinite combination of frequencies and magnitudes and it is also at overall relatively low levels, which adds layers of complication for measuring. The other problem is that the actual commercial applications are relatively extremely small scale.

I'm a bit confused: does this mean you will or will not be selling those nice Audio Note speaker cables you scored in the classifieds the other day? 😁

Edited by MotherSky
  • Haha 4

Share this post


Link to post
Share on other sites
1 hour ago, Nopiano said:
1 hour ago, Speedskater said:

This is a great old paper by Audio Engineering Society, Fellow and interference expert Jim Brown.  But that is in reference to very long balanced interconnects.

So do you think it’s not applicable to speaker cables, given the orders of magnitude here?  If you can find something more explicitly related to speaker cable that would be great.

Given the 2000 foot length and the impedance's involved it's best not to make the mistake of trying to relate it to speaker cables.

* * * * * * * * * *

As for reasonable loudspeaker cables for most conventional loudspeakers, it comes down to:

Total end-to-end (series) resistance of the cable with respect to the frequency impedance curve of the loudspeaker.

However there are some very nice unconventional loudspeakers (think Apogee & Martin Logan) where the total end-to-end inductance may affect the high frequency response.

Share this post


Link to post
Share on other sites
Super Wammer

I'm boooored :zzzz:and it's only page 3.

Ooh page 3, better bring on the glamour girl photo. 

effbbc5ec040fd805c710e09b058b0db.jpg

  • Like 1
  • Haha 2

Share this post


Link to post
Share on other sites
Super Wammer
55 minutes ago, Speedskater said:

Given the 2000 foot length and the impedance's involved it's best not to make the mistake of trying to relate it to speaker cables.

* * * * * * * * * *

As for reasonable loudspeaker cables for most conventional loudspeakers, it comes down to:

Total end-to-end (series) resistance of the cable with respect to the frequency impedance curve of the loudspeaker.

However there are some very nice unconventional loudspeakers (think Apogee & Martin Logan) where the total end-to-end inductance may affect the high frequency response.

Thanks!  That sounds close enough for me to the the earlier quote which said that it ‘is no more complicated than its series resistance and parallel capacitance.

Share this post


Link to post
Share on other sites
Super Wammer
3 hours ago, Nopiano said:

I don’t know where the figures of hundreds of Ohms come from in the Townshend paper, when most domestic speaker cables barely reach 0.1 ohms in normal lengths, and often much less. 

I think that's part of Keith's point that science is being misapplied here.

Share this post


Link to post
Share on other sites
Super Wammer
55 minutes ago, Lurch said:

I'm boooored :zzzz:and it's only page 3.

Ooh page 3, better bring on the glamour girl photo. 

effbbc5ec040fd805c710e09b058b0db.jpg

I coudn't find an "ooh, err, boing" symbol so went with "like". Nothing there that a good ironing wouldn't sort out.

Share this post


Link to post
Share on other sites
Moderator
3 hours ago, Nopiano said:

So do you think it’s not applicable to speaker cables, given the orders of magnitude here?  If you can find something more explicitly related to speaker cable that would be great.
 

I don’t know where the figures of hundreds of Ohms come from in the Townshend paper, when most domestic speaker cables barely reach 0.1 ohms in normal lengths, and often much less. 

You're looking at resistance. not characteristic impedance.

  • Upvote 1

Share this post


Link to post
Share on other sites
Moderator
2 hours ago, MotherSky said:

I'm a bit confused: does this mean you will or will not be selling those nice Audio Note speaker cables you scored in the classifieds the other day? 😁

No.

The AN cables are multi-conductor Litz construction, which may well have an effect. Sonically, I wouldn't like to call it, but to bi-wire with approriate lengths of Van Damme 'hi-fi', which is usually my cable of choice' would cost not much less.

And I don't care about the biwire argument. The designer of the speakers I use states they were designed to run biwired, and whist not everyone agrees with him, I'm happy to take his word for it, considering the sort of music and presentation he enjoys.

Share this post


Link to post
Share on other sites

To explain further the Characteristic Impedance of a cable is made up of the resistance, inductance, capacitance and conductance per unit length. The calculation equation was invented by Heaviside long before the advent of audio.  The value of CI is dominated by the resistance per unit length at low freqs and then gets to the final value as the other active elements come into play as the freq rises. So for example 75ohm coax is not 75ohm until around 100khz. This would happen with a speaker cable.

Mismatches between the CI of a cable and its load will produce reflections. The cable can be any length for this to happen. It happens with PCB traces with high speed digital circuitry (ask me how I know). The reflections travel back down the cable/line/track until they reach the source where they can be reflected again if there is another mismatch. Its common to to use what's called a "back match" to absorb this ie 75ohm resistor is commonly used in RS485 signal lines. 

However these reflections are travelling at a very high speed, 50% of the speed of light comes to mind. So even with multiple reflections its not easy to see how they could affect audio. Thats the bit that has to be explained.

Hope that wasnt too boring.

Ps Those AN cables will probably be high capacitance/low inductance being Litz and if the resistance per unit length is low will be lowish CI. What did you think of them?

Edited by zeta4
  • Like 1
  • Upvote 1

Share this post


Link to post
Share on other sites
Super Wammer
24 minutes ago, rabski said:

You're looking at resistance. not characteristic impedance.

Yes, I was waiting to get caught out on that!  I realise I have no idea what characteristic impedance is, and since I can’t see its relevance to my 2.5metere cables I had better dip out now.  

Share this post


Link to post
Share on other sites
Super Wammer
3 minutes ago, zeta4 said:

Mismatches between the CI of a cable and its load will produce reflections.

Thanks for the helpful explanation.  How relevant is it then to compare the composite measurement that is CI with the impedance of a speaker, as the Townshend paper seeks to do?   Is it more meaningful than saying the the static impedance of a typical speaker is 5.5ohms* and the cable total resistance is (say) 0.05ohms, therefore the cable is irrelevant? 


* I wouldn’t have known this, but Alan Shaw of Harbeth states that it is typically the case, across the red and black terminal on his speakers.  

Share this post


Link to post
Share on other sites

Don't feel bad about it. 'impedance' is complicated enough by itself (ask anyone just learning electronics). But Radio Frequency Characteristic Impedance is way more complicated, it confused the paper's author. And yes, it has no relevance to speaker cables.

Edited by Speedskater
typos
  • Like 1

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.