I looked at the referenced research paper. It is a point in time of a research project. Probably from a PHD thesis application where the applicant only has to defend the science. Not necessarily demonstrate a working model. (Though that would help). It is not a DIY project and the Arduino and PC code would not be germane to the purpose of the paper.
As an engineer, I don't see how the circuit and explanation in the paper would provide a fault location. Just a fault somewhere.
A TDR is basically inserting a signal pulse, then waiting for a reflection. The time between the start and reflected pulse multiplied by the velocity factor of the cable under test gives the distance to the reflection point.
I think the same principle would apply to an optical TDR, but unlike Coax cable, I don't think that fiber anomalies reflect light like wire cables reflect RF.
This is why optical TDR equipment is really expensive.
No, you first need to define what problems are possible and how to discern which of the many do you have right now.
The general manager of my old company was technician for Tektronics division that designed and built optical time domain reflectometers. He had many stories to tell of how the designs were developed and the devices were built and how they were tested.
Do you know there are many different types of optical fiber? Do you know how to test each one? And what will you do when you discover one has a failure? Are you skilled at cleaving the ends and properly applying and testing the patches?
Good luck!
Paul
Steve,
OTDRs work on exactly that principal and have been around for a long time; I was using them in the 1980s and 1990s. You get a really nice trace of distance Vs attenuation and the discontinuities are obvious, such as where there is a splice or a break. The circuit proposed in this discussion is well short of what is needed.
The code should be fairly trivial. All you are doing is reading an analog port, and then testing the value read to see if it fits your definition of a fault (I'm guessing either a major deviation from a previous reading, or simply above or below a certain value). If there is a fault, you display the result on the LCD and send an appropriate message on the serial line. If that level of coding is beyond your abilities, this project seems much too advanced.
This does not tell you if the fiber optic cable is bad, only that the light level has changed (and it also presumes that the normal operation produces a fairly consistent light level, which might or might not be a reasonable assumption).
I used to work on trunk (as in between towns and cities) fibre, an OTDR was essential. An OTDR will tell you to an accuracy of a meter or 2 where a break is (well, they did then, I can only imagine they are cheaper and better now).
@guber , what kind of fibre are you working with, how long and what exactly are you expecting from the tester; your descriptions and ambiguous.