How to connect bi-wire speakers

As an engineer, I will simply state that the concepts and equations behind my statements can be found in any text on electromagnetic theory, and, surprisingly, are pretty simple.
The Director for the Center for Audio Research and Engineering, and Director for Postgraduate Studies within the Department of Electronic Systems Engineering at Essex University in the UK, has written articles which very neatly explain how a binding post can dramatically affect sound quality. While these papers are more specifically about cables, the same concepts apply to the binding posts as they are just a continuation of the conductor chain. Many of his papers can be found here:
http://www.essex.ac.uk/ese/research/audio_lab/malcolms_publications.html#PhD%20Thesis

Hopefully some of those articles can answer your questions. Happy Reading!

Jeffrey Smith
Silversmith Audio

Jeff,

Can you point me to a specific paper as there are more than thirty listed at the URL you gave.

I agree wholeheartedly that all these concepts are in engineering textbooks, at least they were twenty years ago. I am not questioning any of these engineering principles.

However, waveguides and effects such as you describe have long been regarded (by most electrical engineers and professors that I know) as having negligible impact to the signal at audio frequencies and cable lengths (meters not kilometers) used in most hi-fi system configurations.

It is the application of these theories to the range of audio frequencies in the configuration of a typical hi-fi system where I beg to differ/question whether they can have such a large affect as to be one of the worst offenders.

Thanks to point me to the specific paper that shows a binding post (or cable) can be the worst offender.

When those engineers and professors discuss waveguides and their implications in audio cables, they usually just show characteristic impedance, LCR filter effect losses, impedance losses, etc. They rarely mention phase effects, group delay, etc, and I've never seen any mention of the interaction between the electromagnetic wave traveling at a few meters per second in the conductor material, and the electromagnetic wave traveling at nearly the speed of light in the dielectric.
This article is a "readers digest" version which discusses much of this and has been printed in Stereophile and HiFiNRR in the UK:
http://www.essex.ac.uk/ese/research/audio_lab/malcolmspubdocs/G3%20HFN%20Essex_Echo_(cables_1985).pdf

Here is a much more in-depth and technical paper (long - 12mb):
http://www.essex.ac.uk/ese/research/audio_lab/malcolmspubdocs/G6-9%20Unification%20(4%20parts).pdf

Best Regards,

Jeffrey Smith
Silversmith Audio

Jeff,

Thanks I managed to get the first link ( even though it was a little garbled on the thread )

In case anyone else is following this, it should read

http://www.essex.ac.uk/ese/research/audio_lab/malcolmspubdocs/G3%20HFN%20Essex_Echo_(cables_1985).pdf

I fully agree with the discusion and math starting from Maxwell's equations ...who can argue with that!

I also agree with the calculations in terms of the math and the application of wave propagation theories to highly attenuative media like copper. There is no doubt that these losses and distortions do occur.

Again where I beg to differ is the magnitude of these EM effects in relation the very large primary signal.

The numbers suggest that the secondary distortionary effects of EM field propagation generated by the current flow in copper media can be ignored compared to the size of the primary signal (current flowing down a cable and responding principly to LCR lumped circuit theories, current being a bunch of electrons)

Nevertheless, your references and material is very interesting and I plan to read up more on this subject and the links you have provided. Definitely Happy Reading there! Thanks

Shadorne, I think you've hit the nail on the head! The key to all of this is the phrase "at *audio* frequencies". Jeff brings up the group delay bugaboo but fails to mention that this delay is on the order of only a couple hundred nanoseconds between upper and lower frequencies in a typical speaker cable length (say 8ft.), with upper frequencies arriving sooner.

In addition, this difference amounts to changing the placement of the tweeter 16 microns behind that of the woofer. Wanna bet you can't measure that difference in any practical way?