$13,000 power cable

[Hero999]

I was being hypothetical.

The 1/4 wavelength for 20kHz is:
[latex]\frac{1}{4}\lambda=\frac{c}{20000 \times 4}=3747m[/latex]

In practice you would never use such a long cable 3.75km long to transmit audio power and if you did you would use a matched 600hm: telegraph type system.

 

[Nigel Goodwin]

In practice you would never use such a long cable 3.75km long to transmit audio power and if you did you would use a matched 600hm: telegraph type system.

Still no good, you can't feed a speaker signal down that!

Like I said twice before, high voltage transformers each end would make it perfectly possible though.

 

[Hero999]

Yes, I would use two 8hm: to 600hm: impedance matching transformers and a 600hm: transmission line.

But an audiophile wouldn't like that as the transformers would ruin the frequency response, even if you went to the trouble of

 

[yngndrw]

I would use a gold plated fiber optic cable driven with a CO2 laser which would then feed into a 512bit DAC followed then into a 1Mhm: £50,000 solid platinum cable to power my iPod branded headphones.

 

[Nigel Goodwin]

Yes, I would use two 8hm: to 600hm: impedance matching transformers and a 600hm: transmission line.

Too much loss down the cables!

That's why I would prefer a higher voltage level.

But an audiophile wouldn't like that as the transformers would ruin the frequency response, even if you went to the trouble of

Good audio transformers have decent frequency responses, but aren't by any means cheap! - don't Quad electrostaics use them?, and they are seriously well regarded speakers, even after more than 40 years.

Funnily enough I took a new TV to a guy the other day who had a pair!.

 

[Hero999]

Too much loss down the cables!

That's why I would prefer a higher voltage level.

But that won't solve the problem with standing waves creating distortion or will it?

If we use a high impedance cable then using impedance matching transformers will increase the voltage like like you say we need to.

Let's say we're driving 4hm: speakers from an amplifier with a peak output of 12V.

[latex]\frac{N_S}{N_P} = \sqrt{\frac{Z_P}{Z_S}} = \sqrt{\frac{600}{4}}=12.25[/latex]

The RMS output voltage will be:
[latex] \frac{12 \times 12.25}{\sqrt{2}} = 103.9V[/latex]

Look, we've killed two birds with one stone!

 

[Nigel Goodwin]

But that won't solve the problem with standing waves creating distortion or will it?

I think you are probably imagining such distortion?.

If we use a high impedance cable then using impedance matching transformers will increase the voltage like like you say we need to.

Let's say we're driving 4hm: speakers from an amplifier with a peak output of 12V.

[latex]\frac{N_S}{N_P} = \sqrt{\frac{Z_P}{Z_S}} = \sqrt{\frac{600}{4}}=12.25[/latex]

The RMS output voltage will be:
[latex] \frac{12 \times 12.25}{\sqrt{2}} = 103.9V[/latex]

Look, we've killed two birds with one stone!

That's basically only normal 100V line, I'd like to see it a lot higher for the distance you were talking about.

 

[Hero999]

I think you are probably imagining such distortion?.

I'm not imagining it, it does exist, it's basic transmission line theory.

Try connecting your scope to a signal generator, with a 50hm: output impedance running at just 5MHz, via a 10m long twisted and the pulses won't look very square, you'll notice a lot of ringing. Now repeat the same experiment using the same length of 50hm: coaxial cable, the ringing will be greatly reduced. This is because it's acting as a transmission line at the harmonic frequencies, before with the twisted pair which has an impedance of 600hm:you were getting standing waves and transmission line resonance but this has all disappeared with the 50hm: cable because it's matched to the source.

Now normally this isn't a problem for audio, this is because the length of the cable is short compared to the wavelengh but this isn't the case with the example I used in my previous post as the cable is so long it needs to be considered as a transmission line. If I wanted to run a cable to distribute audio from my home in Bedford to you in Derbyshire, stepping up the voltage wouldn't be good enough, I'd need to match the impedance too, if I wanted to keep the ringing and standing waves at bay. As I have illustrated previously, in practise this would mean increasing the voltage anyway as practical transmission lines have a higher impedance than most speakers.

That's basically only normal 100V line, I'd like to see it a lot higher for the distance you were talking about.

That was lucky, if the had used an amplifier with a peak output of 24V I would get twice the voltage but it isn't about the voltage at this distance, matching the impedance is far more important. In the old days, before fibre optics came along, telephone lines used to be matched to 600hm: to prevent these sorts of problems.

 

[Nigel Goodwin]

I'm not imagining it, it does exist, it's basic transmission line theory.

Try connecting your scope to a signal generator, with a 50hm: output impedance running at just 5MHz, via a 10m long twisted and the pulses won't look very square, you'll notice a lot of ringing. Now repeat the same experiment using the same length of 50hm: coaxial cable, the ringing will be greatly reduced. This is because it's acting as a transmission line at the harmonic frequencies, before with the twisted pair which has an impedance of 600hm:you were getting standing waves and transmission line resonance but this has all disappeared with the 50hm: cable because it's matched to the source.

I'm well aware of the reasons for impedance matching, I just disagree on the need for it at audio frequencies like this

While it's quite true that with long enough wires standing waves will occur, the actual amount of high frequency in normal audio is extremely low, so I don't see as it would be a problem, and overall distortion wouldn't be affected to any noticable degree - less that the distortion caused by the losses in the wires themselves, and the transformers!.

 

[Hero999]

Well at 3.75km I would agree.

How far you you away from me?

About 75 miles?

That's 120km which is one full wavelength at just 2.5kHz. Now given that things get funny when the cable length even approaches 14 wavelength, I think you'd agree that an unmatched audio link between us would sound pretty bad.

Light really travels too slowly doesn't it?

 

[Nigel Goodwin]

Well at 3.75km I would agree.

How far you you away from me?

About 75 miles?

Run it to my younger brother's instead, he's not far from you!.

That's 120km which is one full wavelength at just 2.5kHz. Now given that things get funny when the cable length even approaches 14 wavelength, I think you'd agree that an unmatched audio link between us would sound pretty bad.

Now that would be getting more problematic!.

Still, this is far more interesting than $13,000 leads!

 

[CheapSlider]

This is getting back more to the OP topic make your own audio leads

(I like the 'audiophile solder')

 

[philba]

Actually, what I find interesting is that there seem to be countless companies making these "snake oil" products and selling them to a never ending supply of high end, high wallet, low brains consumers. It truly amazes me that not only do they line up to be cheated but they defend the products with religious ferver and fully reasoned pseudo-scientific arguments. I particularly love the guys that claim they can hear sub 10 uS jitter to justify super expensive digital cables.

Well, at least there is one consolation, none of them have turned into suicide bombers... Maybe we could start a high-end audio movement in the middle east.

 

[Hero999]

Run it to my younger brother's instead, he's not far from you!.

Now that would be getting more problematic!.

Still, this is far more interesting than $13,000 leads!

My previous post contained an error, I meant 1/4 wavelength, not 14 wavelengths.

A cheap twisted pair with an impedance matching transformer would do at better job at connecting an audio link between you and me than that overpriced cable would. Also 75 miles of the twisted pair would be cheaper than 6ft of the premium cable.

 

[yngndrw]

I'm not imagining it, it does exist, it's basic transmission line theory.

Try connecting your scope to a signal generator, with a 50hm: output impedance running at just 5MHz, via a 10m long twisted and the pulses won't look very square, you'll notice a lot of ringing. Now repeat the same experiment using the same length of 50hm: coaxial cable, the ringing will be greatly reduced. This is because it's acting as a transmission line at the harmonic frequencies, before with the twisted pair which has an impedance of 600hm:you were getting standing waves and transmission line resonance but this has all disappeared with the 50hm: cable because it's matched to the source.

Now normally this isn't a problem for audio, this is because the length of the cable is short compared to the wavelengh but this isn't the case with the example I used in my previous post as the cable is so long it needs to be considered as a transmission line. If I wanted to run a cable to distribute audio from my home in Bedford to you in Derbyshire, stepping up the voltage wouldn't be good enough, I'd need to match the impedance too, if I wanted to keep the ringing and standing waves at bay. As I have illustrated previously, in practise this would mean increasing the voltage anyway as practical transmission lines have a higher impedance than most speakers.

Or you could used a standard balanced signal cable - You know, the XLR cables that use differential signals. Then it doesn't really matter what interference gets onto the line because it will be subtracted on the far end. Or have I got this all wrong ?

 

[Hero999]

You're not wrong, youprobably would use a balanced line.

I think you're missing my point about impedance matching and transmission lines. At low frequencies it doesn't matter what the impedance is on either end of a piece of wire but as the cable length starts to approach 1/4 of a wavelength all sorts of funny effects start to take place. A signal might reflect of one end of the cable and come back to the source, resonance can also occur which will appear as ringing on a squarewave. All these effects can distort the origional signal or, kill the transmitter, if it is designed to drive a specific load. These problems are solved by terminating the transmission line by its characteristic impedance at one of its ends.

Normally impedance matching isn't required very low frequency signals, as the wave lengths are so long, but if you need to transmit over a long distance it stards to become important. This is why people normally only talk about transmission lines for RF.

https://en.wikipedia.org/wiki/Transmission_line
https://www.tpub.com/neets/book10/41.htm

 

[yngndrw]

Ah, I'm getting vague memories of what I learned in A Level Physics, with waves.

Thanks for the explanation.

 

[Super_voip]

As long as the wire is thick enough not to mask the damping effect(source impedence) of the amp by too great an amount the rest is whooHaa. Tests have shown that people can differentiate 0.1dB difference in level with the louder ie 0.1dB more sounding better, so all these cables have to do is increase the level at the speaker by 0.1dB and so justify their exsistance.
BTW to those that aren't up to scratch with their dBs 0.1 is a very small amount.

 

[CheapSlider]

The cable at the beginning of this thread is only a mains cable.

For the speakers the company offers this **broken link removed** at a mere $30,344 for 6 feet.




I notice they have vacancies for 'cable artisans'

PS
Snake oil ? Did you mean **broken link removed**

 

[Hero999]

That's a joke, so there can be many miles of cheap PVC insulated cable between the amplifier and distribution transformer but that last few meters of cable will improve the quality of the supply that much. :rolleyes.