However disturbing my contribution to designing and manufacturing High-Fidelity cables may be, never was I led by any controversial purpose: some aberrations about cables from leading manufacturers are not worth mentioning.
My approach took into account all the design that, to my mind, partly solved the problem of transmitting a low frequency signal between two links of a high quality music system.
Accordingly, I concentrated on three cables that favoured technological advances and improved our knowledge of all the phenomena ruling the science of music transmission.
As for Isoda cables, the solution they brings to the design of L.F. cables is not usable hereafter.
In spite of the experts‘reluctance Audioquest proved, with its Fms Blur, that Litz‘s wire endowed it with a musical dimension hitherto unknown.
Besides Audio Research proved that with Litz‘s wire, pure long crystaled copper and an unequalled mechanical hold of conductors, a major problem was actually the nature of the materials transiting the signals, since a significant part of the electrical power transits the cables insulator.
Mit was the first to introduce corrected cables in the electronical principle that has been currently used in telephony since 1914, according to the physical principles discovered at the end of 19th century and applied by professors Pupin and Krarup.
But on the one hand Mit solution is far from industrial besides the fact that, in a transmission line, all the parasitic elements are scattered along the line and thus unlocated.
On the other hand the manufacturer's error (a youthful error, I hope so) was simply using an ordinary cable with a view to ulterior correction instead of designing an „idealised“ cable, so to say, with a view to reducing further correction to a minimum afterwards.
Taking the above-mentioned experimets into acount, we had to find a specific structure in order to solve the problem by eliminating first the coaxial topology so as to reduce power in the insulators by parting both conductors using air as a possible insulator.
We also have to drive each signal (+ and -) along exactly the same geometries and hold both conductors as tightly as possible.
Besides, using Litz's wire (or equivalent construction to avoid skin effect) in order to eliminate Eddy current in copper is an absolute necessity and we must select the highest pure and long-crystaled copper (OFHC), the effect of which is explained hereafter.
The final operation consists of measuring the resulting parameters in a given structure and correcting the lumped section according to its data and length on an impedance analyser from DC to 1 MHz.
To achieve the correction, let us elicit mathematically the parameters needing to be corrected so as to obtain a distortionless cable.
First I think it's relevant to mention that my approach was not in the least motivated by scientific exactness or interest but by sheer intuition after listening to music on existing cables which proved inevitably unsatisfactory (except for Mit and Isoda) and designing several cable prototypes to test my hypotheses.
Therefore you should not regard my work as a scientist's attempt to have his formulas coincide with actual sound, but as a music lover's attempt to solve a concrete problem by using, somewhat artfully, his own scientific knowledge.
On the whole, we can write the characteristic impedance of a transmission line thus:

We can see at once that LG = CR
That was English scientist Heaveside‘s condition ( end of the 19th C.). If the condition is fulfilled, the result is that a is independent from frequency and the signal is also transmitted undistortedly.
But in practice and usually, he product LG is always very low, compared with RC, thence the „Pupinisation“ of telephone lines which consisted of inserting a few mH inductor a mile in the lines. The improvement and quality of the messages increased to the detriment of the wave velocity and the current in the lines. But High-Frequency experts consider rightly that R = 0 and G = 0 and obtain a simplification of the charateristic impedance:

But this formula is aberrant for L.F., especially if its origin is unknown since in L.F. parameter G is not disregardable and R is about 1/100 ohm

Therefore, in order to fulfill Heaveside‘s condition we have to:
ø Reduce R : OFHC copper allows to do so, but it is not enough.
ø Reduce C : It is hardly impossible since it depens on nature and permittivity of the insulators.
The only parameters we could possibly modify are the line conductance and its inductance. Therefore we have to elaborate a technique increasing both inductance and conductance by „weakening“ the insulators and or inserting a reactor along the line as well.
In conclusion, let us introduce the magnetic material „Carbonyl Iron Powder“ used in that survey. This material first appears as a powqer, in the shape of pure iron particles perfectly spherical from 1 to 8 mm in diameter. Not only is it efficient in frequency up to 300 MHz but it is chatacterised by 1,9 Tesla induction of saturation, no remanence, very high temperature stability and, contrary to most magnetic materials, its resistivity and its very low permeability about 10.
Consequently, Carbonyl Iron has proved satisfactory and plays a significant part in my research for high quality cables, and the best way to appreciate them is just listening.


Jean Fadel
Paris France April 1988