So let’s say you have an RG-59 cable or something similar. It’s several hundred feet long but it developed a break somewhere along it’s length. The problem is it’s not obvious where the break is. Now what do you do?
Option 1: Cut the cable in half, check each half to find which is useable without the open break. Problem with this approach is we’ve wasted at most 50% of the cable length…. What if the problem was just a few feet away from one of the cable ends?
Option 2: Geek out! Grab a Pulse Generator and Oscilloscope and find the exact location of the break!
This is what a very fast pulse looks on an oscilloscope.
Note: Rise-time of the pulse is about 8 nanoseconds! It’s very short duration pulse, and very fast. What we want to do is send this very short pulse down the cable and watch (or wait) for it’s return. If your cable is unterminated (or if it’s got an open break somewhere along it’s length, that location will look unterminated), we’re going to get a “reflection” back of the signal we sent. And the reason why we want a very fast and short pulse signal is so that it will be easier to see the “reflected” signal apart from the original signal/pulse we sent out. (As opposed to sending a continuous sine-wave signal for example.) When we get the reflection back, it will be very obvious… it will be smaller in magnitude, and it will come after our original signal. 
Next, we want to measure the time difference between the original pulse, and the reflected pulse signal. The best way to do this is using the Math functions of your scope, and the cursors. Just point the cursors to the original bump, and the reflected bump.
In this example, the difference is 29.60 nanosecond!
Now, different types of cables will have different dielectric materials and therefore, different propagation delay, or speeds.
For my commmon, cheap cable, with Solid Polyethylene dielectric, the table shows it takes 1.54 nanoseconds for the signal to travel 1 foot.
Other types of cables with different dielectric have a faster speed, almost approaching speed of light in a vacuum, 1.12, 1.13, 1.15 ns/ft.
So, now we’ve measured our time difference (29.60 ns), we can compute the distance the signal travelled. Take note, this distance is a “round-trip” distance… going out, getting to the open/break/unterminated connection, then being reflected back.
dist = 29.6 ns / 1.54 ns/ft = 19.09 feet
So if we divide this round-trip distance by half, then we get the length of the cable (or the location of the open break)!
length = 19.09 feet / 2 = 9.545 feet
Now, the moment of truth… let’s measure the actual cable I used for this experiment.
I measured 113 inches, and this includes the various extenders I used. Convert to feet.
Actual (feet) = 113 inches / 12 = 9.41 feet
Our error difference is: 9.545 – 9.41 = 0.135 feet (or 1.62 inches!) So our measured cable length, and the computed cable length are indeed very close.
