A short while back @marconi asked for the surge impedance of some 25mm SWA, so here are some related measurement results.
For those who don't recognise the term, "surge impedance" is the same as "characteristic impedance" as in the 50 ohm cox cable you see advertised. It can best be explained by the thought experiment of having an infinitely long length of 2 core wire. If you can avoid salivating at the copper to be nicked, then ask the question "If I connect a battery, how much current will flow?"
As the cable has no end, it will not depend on the load but on the cable itself, and from that we can derive a model and some equations as covered by this article:
en.wikipedia.org
Anyway, I happened to have a LCR meter, some off-cuts of cable, and a bit of time on my hands this dull wet day, so I made some measurements of the open circuit capacitance, and the short circuit inductance, so we can compute an estimate of the impedance.
At 1kHz the simple SQRT(L/C) approximation is not very good, but at 100kHz and above it is close. Also you see the R increase with frequency due to the skin effect, typically that is negligible at power frequencies unless you have very big conductors (few hundred mm^2).
TL,DR; Mains cable in the 30-70 ohm range, not far off coax cables.
For those who don't recognise the term, "surge impedance" is the same as "characteristic impedance" as in the 50 ohm cox cable you see advertised. It can best be explained by the thought experiment of having an infinitely long length of 2 core wire. If you can avoid salivating at the copper to be nicked, then ask the question "If I connect a battery, how much current will flow?"
As the cable has no end, it will not depend on the load but on the cable itself, and from that we can derive a model and some equations as covered by this article:
Characteristic impedance - Wikipedia
Anyway, I happened to have a LCR meter, some off-cuts of cable, and a bit of time on my hands this dull wet day, so I made some measurements of the open circuit capacitance, and the short circuit inductance, so we can compute an estimate of the impedance.
| Length (m) | freq | C | D | L | R | Z (from L/C) | Z (from |L+R|/|C,D|) |
Split concentric, 16mm | 5.5 | 1.00E+03 | 1.32E-09 | 0.008 | 2.00E-06 | 0.014 | 38.92 | 47.6 |
| | 1.00E+05 | 1.31E-09 | 0.005 | 1.36E-06 | 0.098 | 32.22 | 32.3 |
| | | | | | | | |
2.5mm SWA 2-core | 0.9 | 1.00E+03 | 1.44E-10 | 0.015 | 7.00E-07 | 0.013 | 69.72 | 123.2 |
| | 1.00E+05 | 1.38E-10 | 0.018 | 5.10E-07 | 0.053 | 60.79 | 61.2 |
| | | | | | | | |
2.5mm Prysmain T&E | 3.1 | 1.00E+03 | 4.50E-10 | 0.062 | 1.90E-06 | 0.067 | 64.98 | 155.3 |
| | 1.00E+05 | 3.60E-10 | 0.077 | 1.65E-06 | 0.165 | 67.70 | 68.2 |
At 1kHz the simple SQRT(L/C) approximation is not very good, but at 100kHz and above it is close. Also you see the R increase with frequency due to the skin effect, typically that is negligible at power frequencies unless you have very big conductors (few hundred mm^2).
TL,DR; Mains cable in the 30-70 ohm range, not far off coax cables.
