Trick of the Light

Cable Asylum

Interconnects, speaker wire, power cords. Ask the Cable Guys.

I originally thought that the two braids (the Kimber braid and the one I was doing) were different, but they're not. Turns out to be just a trick of the light as the Who might say.
When I replicated the Kimber braid, I just brute forced it from the patent drawing. Seeing as it was two intertwoven twisted pairs with each pair on a different axis at 90 degrees to each other, I did the braiding on two different axes.
I used two wires of the same color on one axis (left/right or x) and two wires of another color on another axis (up/down or y). I took two same colored wires and crossed them back over themselves on their axis and took the other two wires and did the same on the other axis.
The end result is that one color wire forms a pair on one axis and the other color wire forms a pair on the other axis.
When I do four wire braids, I start out with alternating colors (in this case, blue/white/blue/white). Doing it this way gives the same geometric result, but instead, each pair is made up of two different colors rather than one color which gives it a different appearance than if you started out with paired colors (i.e. blue/blue/white/white).
Also, there's one difference between the Kimber braid and the Inter-8 Weave.
When Kimber ties together two wires to form one of the two composite conductors, they use the two wires on the same axis. In other words, one axis forms the positive conductor and the other axis forms the negative conductor.
The Inter-8 Weave forms the conductors using one wire from one twist on one axis, and the other wire from one twist on the other axis. This is how I connect mine. Using the alternating colors at the start of the braid makes it easy to make the connections using just two colors. Just connect like colors together at each end, i.e. blue to blue and white to white.
Technically, the Kimber approach is actually a step backward in terms of magnetic field interference.
The purpose of using one wire from one of the pairs on one axis and another wire from one of the pairs on the other axis is so that each resulting conductor presents substantially the same cross section to the interfering fields (which are vector fields). In this way, the induced voltage across each conductor is also substantially the same (i.e. common-mode).
Only when the noise voltages are the same across each conductor can the differential input, which amplifies only the difference of the voltages on each conductor, reject them. Any difference in the noise voltages results in their difference being passed through and amplified.
In other words, if you've 1 volt across one conductor and 1 volt across the other, the differential input subtracts them giving you 1 -1 = 0. If on the other hand you have 1 volt across one conductor, and say 0.95 volts on the other, you get 1 - 0.95 = 0.05 volts which slips right past the differential input just as if it were a signal.
By forming each conductor from the same pair on the same axis, each conductor presents a different cross section to the interfering fields which will result in a different induced voltage across each conductor, reducing the effectiveness of the differential input it's feeding.
In terms of maintaining interfering voltages in common-mode, even a simple twisted pair would outperform the Kimber approach. So if noise reduction is important to you, the Inter-8 connection scheme would give the better result.
se

This post is made possible by the generous support of people like you and our sponsors:

VH Audio

Topic - Trick of the Light - Steve Eddy 00:26:27 10/03/02 (0)

You can not post to an archived thread.