Optimum Gauge for Networks

[rolandr]

Steve:

No problem, personally I like to bring controversial subjects to the forum. It has actually been a reasonable response since this apparently challenges accepted teachings on the matter. The attack has been mainly of the numbers. But I am disappointed that a closed mind attitude has taken over blocking the ability to appreciate the fact those numbers can only, if at all, occur at MAXIMAL skin depth therefore suggesting cupper bar type cable!!! Also ONLY Dan considered looking at the article and actually was thinking of testing the "theory" instead of just attacking it. Kudos for him. This despite the fact that Prof Hawksford is no slouch in this field!!. My original comment of keeping cable thin while at the same time "keeping resistance as low as possible was also totally overlooked". In general the concept of keeping a scientific open mind was not applied!! It seems that once a theory is accepted nobody else considers checking it. We must remember that theories come out of empirical research and not the other way around. You can not attack a theory just by quoting another theory. The article was sent by Allen Wright and is available.

[rolandr]

Gentlemen:

Prof Hawksford 18 page article HAS BEEN FORWARDED TO TODD hoping he can post it for the reading pleasure of everybody. Surprisingly after doing significant analysis the prof feels that there is possibility that fat cables may have an effect at audio freqs and actually recommends use of thinner cables for audio. It looks like I will be standing by my original recommendations for thinner cable use as described here earlier.... Again do not kill the messenger..... my wife is worried!!!

[marcuswilson]

Steve,

There is no personal abuse going on.

Roland,

My experience in electronics over 30 years has shown me that signals do not travel at such slow speeds and there are situations in which certain frequency components are delayed in cable, but that is explained in other ways than skin effect. I think you are quoting figures and possibly making assumptions out of context. It is perfectly measurable that skin effect has no time delay associated with it. Furthermore, in high frequency work, one uses larger diameter wires than normal to account for skin effect and they are usually hollow as the centre is useless. There is no frequency related delay in these systems.
The accepted understanding states can be no way that the skin effect can cause different frequencies to travel at different speeds at audio and supersonic frequencies. You need to produce a valid explanation to substanciate such claims.

Theories may come from empirical research, but not always. A theory is an unproven assumption and is yet to be absolutely proven. After all they only seek to explain what is observed. Many have been proved wrong.

[Richard C.]

Hi all,

I see that this "revolutionary theory" paper of Prof. Hawksford has been out for review from the whole audio world for more than 18 years. IF it has ANY merit at all, I would think that it would be accepted by now and have a great following. Theories aren't accepted overnight in this world and most are never accepted at all; they are simply passed over quickly. There isn't enough time to study every theory to the ninth degree; not when we have proven over and over again how skin effect really works. Most of these theories are like opinions - everybody has one! Could there be a good reason why the audio world has overlooked this article?

It's too bad that there aren't more followers (with lots of spare money for cables) to buy a "special cable" from the Prof. One that only his small company with a few highly devoted employees could produce. This is truly a missed marketing opportunity!
Richard C.

[rolandr]

After again obtaining a copy of the Prof's article
and reading it numerous times, his points become
more clear.

He states that axial transmission occurs superficially on the wire and is close to immediate in speed. But a smaller component of the transmission penetrates deeper into the actual wire. The more the conductivity of the metal the more this occurs. This small component of the signal gets significantly delayed by traveling deep in the wire as opposed to the immediate transmission of the superficial traveling component. This delay is freq dependent
because the lower the freq the deeper it can penetrate. He points out that this delay is significant as listed in the calculated freq/skin depth/velocity #s given above. This freq smearing
may be most noticed during transient sounds. Once a sound stops the large axial superficially traveling signal component rapidly stops since it is just about immediate BUT the deeper slower freq dependent component is still traveling and in effect causes a smearing which may be audible at the end of transient sounds. The fatter the wire the deeper and slower this in wire traveling component can be. Therefore he recommends thinner wire anywhere from 0.5 to 1mm for most audio work. As far as I know he does not have any economic interest in any cable manufacturing company. The Prof has an excellent reputation and impeccable credentials. He is well regarded and well published at the AES and JAES. Please keep criticisms where they belong, that is, strictly based on your disagreement of the material he presents and away from personal attack. Win your arguments with facts and not with slander!!!!

The Prof's article was E mailed to all in this particular discussion except to the one that actually requested it Dan since I did not have his E mail. Please let him know it is available.

Roland

[AllanCT]

I just finished reading it.

Anyway I just need to know what the loss field would be in an interconnect cable of 2 meters or better yet 6 feet. You know the low end monster cable ones that fit the jacks okay.

I didn't see the part where he assumed or calculated the exact loss field for a decent preamp. and high impedance amplifier which I will assume we all have here.

The significance of that loss field would be paramount to his argument assuming his other calculations are correct.

If the loss field insignificant like I shall presume his theory becomes rather moot.

If there are ways to quantify those fields in a simple interconnect I would be happy to review that discussion.... I think.

[airboss]

Hi Roland:
I have kind of moved on to other topics and interests as it turns out. Seems like we flogged this one to death some time ago. Anyway, I'll take a look at it.

Dan

[airboss]

Greetings Gang:
Well, we've been duped it seems. Prof. Hawksford telegraphs the whole thing by letting us know in the first instance that the date is April Fools Day. These sorts of techno-babble articles are published in many journals in the April issue. QST has one every April for instance. I recall even on Ebay the April 1st auction of a radio chassis lieing in the bottom of a fruit box with old newspapers described as "classic hallocrafter (sic) bannana boat radio with original papers". Great fun but not a word of truth to it. Figure 1 is breathtaking. Sinusoidal waveforms at right angles IN a conductor? Kirchoff's law maybe? Roland this is good stuff!

Thanks
Dan

[marcuswilson]

I was scratching my head for a while, reading it at 1 in the morning.

I think his conclusion is wrong, because stranded cables have an advantage in that they would capture the loss component. The loss component from one strand of wire travels radially, so will leave this wire and go into the adjacent wires. Being loss components they will then go at right angles when they enter the second wire and travel along axially. In this way, the loss component will be reclaimed!

The only problem is to make sure the loss component turns at right angles to the correct direction (ie. forward). This is, of course, why cables have the direction of signal travel marked. You will notice extra loss if you use these cables backward, because of the reversing of the reclaimed loss components trying to travel back toward the amplifier!!! The impedance of the reversed cable, measured at the output of the amplifier (in the case of a speaker cable), would be different from the same cable the right way round.

It is essential that you know in what direction the cable was extruded when it was manufactured, so it can be used the correct way. Only cables which have all of the strands laid the correct direction around can truly be low loss.

I wondered why MOH would bother writing such a paper, but I have now noticed how easy it is.

[rolandr]

I give up!!! I guess if you can not win
with facts you must WIN with slander. I also noticed everybody is an engineer of what I am not sure. Janitors are engineers, bathroom attendants also. I personally don't think this matter is that important but I am totally blown away by the low brow closed minded attitude of people in this discussion. I just wished someone with more knowledge could put this article in proper perspective. The prof is no joke. He sits in the review committee of the AES. He just points out that there may be more to cables than the simplistic aproach we all were raised to believe
and that fat cables may hurt not help. Why all the personal attack??? If you disagree just blow the article off and continue using your multistranded fat whatever. We all agree on not paying a fortune for whatever cable it is we use.
The only argument is wether to use thin or fat.
This discussion is over as far I am concerned. You have the article figure it out for yourselves.