5

Sometimes we might need to know the exact length of the wire pairs inside a cable. One can be interested in calculating voltage drop or some other calculation regarding the exact length.

Because the wires inside the cable are twisted, their actual(electrical) length is greater than the cable jacket length.

Is there any practical method to calculate the actual length?

I couldn't find a duplicate, but if there is I will delete this question.

user16307
  • 12,049
  • 52
  • 184
  • 320
  • If you want to determine volt drop then measure the loop resistance. – Andy aka Nov 11 '19 at 12:39
  • loosely twisted wires will not be "as long" as tightly twisted pairs. I've seen this effect in using a drill to create tight twists, and during the drill's rotation the wire is continually CONTRACTING. – analogsystemsrf Nov 11 '19 at 13:23
  • What method you chose should take into consideration why you need the answer. Don't forget obvious things like looking at building plans. – Chris Stratton Nov 11 '19 at 18:49
  • Time domain reflectometry is a technique where you send a short pulse and measure the travel time to the open terminated end of the cable and back to the source. Half of this time is your effective cable "length". – Captainj2001 Nov 11 '19 at 22:25
  • That's a very good question. I'm wondering whether we can get a good estimate by measuring the twisting pitch and the individual wire insulator external diameter. If you are the one actually twisting the wires, a practical solution is to simply measure the length before and after the twist and come up with a fraction. If you're buying an already twisted cable, you could do the opposite and untwist a small segment of the cable. – joribama Nov 12 '19 at 06:14
  • 1
    @joribama I have submitted a geometric math themed answer that you might find interesting. – Shadetheartist Nov 13 '19 at 22:58
  • 2
    @Shadetheartist - I had noticed (and up-voted) already. That's exactly what I had in mind :) – joribama Nov 14 '19 at 05:02

4 Answers4

5

From a purely geometric perspective you could calculate the length using the helical length equation.

enter image description here

Where H is the length of the twisted wire and

enter image description here

Where R equals the radius of the turns in the wire. Basically from the center of the twisted pair assembly to the center of one of the wires.

enter image description here

So if the wire makes a complete rotation around the center in 10mm, and distance between the center of the twisted pair and the center of a wire is 1mm, then were you to untwist the wire and straighten it the length would be

enter image description here

Shadetheartist
  • 341
  • 1
  • 9
3

Cut off a 1 meter piece, remove and straighten one of its conductors, and measure the actual length per meter.

Update:

Straighten out one of the conductors and measure it. Say it's 1.05m long, or 5% (made-up number - I've no idea whether it's realistic) longer than the cable it came out of. Apply that extra 5% to the length of your cable to get the length of the conductors inside.

JRobert
  • 3,192
  • 14
  • 27
2

One easy and obvious way is to connect the cable at the end and measure the loop resistance with an ohmmeter. Of course you need to know the resistance of the cable.

The better and very precise way is to use a time-domain reflectometer (TDR). This device sends an impulse into the cable which is reflected at the (open) end. The time of the reflected signal is measured and because of constant wave propagation, the length of the cable is calculated.

Stefan Wyss
  • 7,396
  • 1
  • 12
  • 24
  • 4
    One limitation of using TDR is that you need to know the cable's relative permittivity in order to use the proper propagation speed used in the length calculation. Any inaccuracy on the permittivity will result in an inaccuracy in the length estimation, albeit to a smaller degree due to the fact that the propagation speed is inversely proportional to the the square root of the permittivity. – joribama Nov 12 '19 at 06:07
  • "Of course you need to know the resistance of the cable." And you also need a very good meter. Cable conductors are usually rather low in resistance, so measuring takes considerable finesse. A Kelvin connection is usually called for. – WhatRoughBeast Nov 12 '19 at 18:49
1

I used the 2 probes of an oscilloscope to measure the propagation delay and found 16ns for a 1.8m long (external) Ethernet cable (and 26ns for 3m). Either vp=1.1e8 m/s (not the expected 2/3 co=2e8 m/s https://en.wikipedia.org/wiki/Velocity_factor) or the cable are (2/1.1-1)*e100 = 80% longer. I was not expecting such a large difference.

Voltage Spike
  • 82,181
  • 41
  • 84
  • 220
FredW
  • 11
  • 1