Navigation: DIY Audio Projects / DIY Hi-Fi Audio Cable Projects / DIY Braided Speaker Cables

DIY Braided Speaker Cables

Sreekanth Nagaraj  India Flag To email Sreekanth, type out the email address.

DIY Braided Speaker Cables

Here is my simple do-it-yourself (DIY) speaker cable effort. The length of this cable is 3 meters to suit my requirements. However, this cable design is of low capacitance (C) and resistance (R) and relatively low inductance (L) so the length may be increased (within reason) to suit your required length.

I first experimented with "Twisted Pair" designs. The main problem with this is the Figure-Eight braid, so I tried hard to avoid this by using three pairs of wires. After a lot of trial and error as well as testing with various patterns of braiding, this is the one which gave me the best results. Six 16-gauge stranded copper wires were used per speaker cable (three per conductor) which result in an aggregate gauge of about 11.25 AWG per conductor. See the American Wire Gauge (AWG) Table for a reference of conductor size.

Close up of Six Wire Braid

Photograph 1: Close-up of Six Wire Braid

Close Up of DIY Braided Speaker Cable

Photograph 2: Close-up of DIY Braided Speaker Cable

It is easy to keep resistance low as that is simply a function of conductor size. What you do have to pay attention to with DIY audio cables is inductance and capacitance. For speaker cables high inductance will affect the frequency response, while high capacitance may cause some amplifiers to oscillate and become unstable. This cable design results in very low capacitance while also marinating the inductance at an acceptable level. The measurements for my 3 m (~10ft) DIY speaker cable are C = 0.013 nF and L = 0.002 mH.

DIY Braided Speaker Cable

Photograph 3: DIY Braided Speaker Cable

Table 1 shows a comparison of this inexpensive DIY speaker cable with some (expensive) commercial speaker cables.

Table 1: Comparison of Cable Properties

Cable Capacitance Inductance Resistance Gauge Cost (Pair)
DIY Speaker Cable
12AWG Zip Cord
Cardas Quadlink-Five C*
Kimber Kable 4TC*
Kimber Kable 8TC*
Goertz MI 1*
1.3 pF/ft
18.0 pF/ft
19.0 pF/ft
44.2 pF/ft
100.1 pF/ft
500.0 pF/ft
0.20 uH/ft
0.19 uH/ft
0.16 uH/ft
0.09 uH/ft
0.04 uH/ft
0.01 uH/ft
0.013 ohms/10ft
0.016 ohms/10ft
0.018 ohms/10ft
0.023 ohms/10ft
0.011 ohms/10ft
0.022 ohms/10ft
11.25 AWG
12 AWG
12.5 AWG
13 AWG
13 AWG

* Commercial cables measurements are from the respective manufactures website.
** Approximately $500 for a complete terminated pair of 3 m (10 ft) cables.

So you can see that there is a wide spread in the measurements. The capacitance of my DIY cables is anywhere from about 15 to 350 times lower than some of the "exotic" cables, so amplifier stability is not an issue. I am not sure that an ultra high capacitance will give the desired sonic results (and could cause amplifier instability). I find the bass response from high capacitance cables to be affected; of course this is my opinion.

SPICE-Based Analog Simulation and Discussion

To give you an idea of the of how the frequency response of speaker cables is affected, the SPICE-Based Analog Simulation software TINA-TI was used to simulate the cables. The plots below show the simulated frequency response for 10 foot (3 m) lengths of the DIY speaker cable presented above and the Kimber Kable 8TC. For the simulations 28.3 V signal are applied and the end load is 8 ohms which would result in about 100W of power.

Spice Based Simulation of DIY Speaker Cable

Figure 1: Spice Based Simulation of DIY Speaker Cable

Spice Based Simulation of Kimber Kable 8TC Speaker Cable

Figure 2: Spice Based Simulation of Kimber Kable 8TC Speaker Cable

As you can see, the simulations show that there is virtually no difference between these two cables within the audible range (usually quoted as 20 Hz to 20 kHz). Relative to a perfect conductor (zero loss), the DIY speaker cable is -0.018 dB at 20 kHz and the Kimber Kable 8TC is -0.012 dB at 20 kHz (0.006 dB difference between the two). The majority of those losses noted are uniform insertion losses (i.e. cable resistance). One last note about the simulations, those losses noted are at about 28V or 100W of power. For typical listening levels (about 1 to 10W) the losses and differences would be even smaller.

Now to give those tiny numbers some perspective, the average threshold of human hearing is typically considered to be about 3 dB. Some audiophiles with well-trained ears claim to be able to hear a change of 1 dB. It is hard to imagine that one can hear a differnece of 0.006 dB. See the The DIY Speaker Cables thread on the forum for more information or to discuss this project.