DIY cable thinking

In 1983, Dr Malcolm Hawksford published a seminal and entertaining analysis of AC signal transmission within wire conductors, in The Essex Echo series in HiFi News & Record Review magazine.

(This article, updated, was re-published in Stereophile Oct 1995.)

Dr Hawksford offers plausible explanations of what I was hearing in my humble experiments a few years later, listening to different configurations of single strand copper wire as interconnect and loudspeaker cabling.

Ideal conductor size:

Referring to known electromagnetic theory, the phenomenon of Skin Effect means that AC currents (music signals, for example) flow at the outer ‘skin’ of a conductor.  Also, the ‘depth’ of this layer reduces with frequency.


For audio frequencies, Hawksford argues, in order to minimise high frequency attenuation (roll-off), the conductor wire diameter should not be too large or small.

“These results suggest a copper wire with a maximum diameter of between 0.5mm and 1mm is optimum if a uniform current flow across the conductor is to be maintained over the audioband…..

At a diameter of around 0.8mm, the conductor becomes closer to a low-valued ideal resistor at audio frequencies…..

The time taken for the field to propagate to the skin depth δ is longer at low frequencies. Thus, thick conductors would appear more problematic at low frequencies, showing a greater tendency to time dispersion….”

In my own  listening experiments, I liked the bandwidth and tonal balance using copper wire of 0.65-0.7mm diameter (22swg/22-21awg).

To my ear, ‘single strand’ wires (for interconnect and loudspeaker cabling), sound tonally ‘complete’ with full harmonics and also clean and ‘grain-less’ – compared to multi-stranded cables.


Additionally, Dr Hawksford explains, that AC signals travelling along/within a wire, are accompanied by electric and magnetic fields with amplitudes at 90 degrees to the signal propogation.


“It appears cable defects have their greatest effects under transient excitation rather than within the pseudo steady-state of sustained tones. Transient edges are effectively time-smeared or broadened (albeit by a small amount), where this dispersion is a function of both the signal and the properties and dimensions of the conductors.”

This possibly explains some other listening observations that I have made:

  • Signal conductors sound best when thay are un-sheathed, un-wrapped – they sound best when surrounded only by air. (Dynamically more ‘open’, less ‘constrained’.)  – sheathing materials likely interfere with the signal fields.
  • Signal wires also sound best when separated – not bunched, twisted, plaitted, etc. – The signal fields of conductors in close proximity to each other, likely interfere with each other.

Multiple conductors:

“Stranded conductors without individual strand insulation appear to be a poor construction when viewed by this model, as the loss field propagates against the strands and experiences discontinuities in air/copper boundaries that are inevitably random. This is comparable to a large-scale granularity…..

…..multiple, separately insulated (Litz) strands can usefully lower the series impedance due to the internal field in the conductors. Whether the overall impedance is lowered depends on the conductors’ geometry and spacing, both of which define the series inductance due to the external field components.”

Depending on your system sensitivity, you may need more conductor cross-sectional area than provided by a single strand wire, so multiple wire cable configurations may be required.

This raises the question of Cable Geometry…..

Preferred loudspeaker cable configuration:

Depending on system sensitivity & cable lengths, you could use 1 or 2 or 3 wires for each +/- polarity.  Alternating polarities seems to result in more dynamic ‘freedom’ – possibly something to do with magnetic fields reinforcing/coinciding instead of clashing/opposing.


Copper wire spiralled around a plastic tube, alternating +/-/+ polarity.


0.7mm diameter, coated ‘winding wire’ was used .  (Uncoated wire would be ideal, but difficult to obtain.)

Sound: This cable provided a big, spacious, warm sound, with very good sense of ‘depth’ and ‘width’. For a more ‘explicit’ treble tone, you can use 0.65mm diam. wire, or a mixture of 0.7mm and 0.65mm.

Preferred interconnect cable configuration:


1 x 0.7mm diameter copper ‘winding wire’ for each +/- polarity.

Summary of key ideas:

  • Copper wire conductors 0.65 – 0.7mm diameter.
  • Air-separated conductors, avoid sheathing or screening if possible.
  • Alternating +/-/+ polarities, if you are using more than one conductor in each polarity direction.
  • Devise a method of mechanically holding wires rigid to minimise vibration – in my case, using a plastic tube.


A selection of other conductor configurations tested & rejected:


Air-spaced conductors – nice but lacking some dynamic punch – the wires are loose and free to ‘vibrate’.



Wires routed inside a tube do not sound as good as wires which are held mechanically rigid and spaced apart – eg. by spiralling around the outside of a tube.


Conductors in close proximity (eg. tight twisting, plaitting or braiding) can sound ‘impactful’ but harshness, ‘confusion’ and ‘graininess’, are introduced. Less ‘transparent’.


Configurations with wires crossing each other, add ‘bite’ & ‘sharpness’, but also edginess and listening fatigue.


As above, this configuration produced punchy bass, but rolled-off, absent treble – lacking ‘openness’ and ‘air’. (Significant parallel capacitance?)


Approximation of ‘Golden Ratio’ wire diameters (a la Cardas) – tonally impressive with extension top and bottom.  Has merit, worth re-visiting.

Maybe it’s about time I repeated these listening tests and carried these experiments further…….




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s