I would like to build a coax cable tester and would like to use an Arduino with an LCD display. I am often going through old sections of coax cable, mostly 50 ohm terminated with PL239 connecters, but there could be others.

I would like to be able to:

* Measure the characteristic impedance of the coax cable under test
* Determine the length of the coax cable under test
* Simple connectivity test from one end to the other. It would be cool to send rf data from one end of the cable to the other end, and determine if the cable is functional, but I suspect this would be harder to do. Not necessary in the first version.

I'm not an expert with the arduino or electronics for that mater but if someone can get me going in the right direction I should be able to figure it out.

I found this page http://www.irb.hr/users/stipcevi/ele/coaxweb/index.html but I'm not sure if this is adaptable for use with the arduino.

this page is a discussion of characteristic impedance of cable http://home.mira.net/~marcop/ciocahalf.htm

Thanks... jrsphoto

Awesome idea! I've seen arduino projects that implement AF scope capabilities. The limitation I see is the speed of the Arduino; the design calls for time resolution in the nanosecond range, which, iirc, the Arduino runs at 20 MHz (too small for ns accuracy.) I've also seen Arduino projects that will calculate capacitance. In any case, I'd like to work with you on implimenting this as it would be a good tool to have at the HF and would help me understand capacitive and inductive networks better. One thing I'm really interested in is impedance matching. E.g. iPod to audio receiver; I don't have an iPod, but my pilot brother recently tasked me to create a coupling device between the iPod and the aux input to the cockpit of his plane because the audio level was too low due to the impedance mismatch.

Also, it would be neat to calculate the Q of such conductors, both loaded and unloaded...

as long as we are at it how about a network tester ethernet.

That can be done quite easily.

not 100% sure but my initial thoughts are:

at 10nS you need to be sampling at a minimum of (1/10e-9)*2 = 200Mhz which would be a bit out of the range of the 8Mhz arduino.

How about a PLL subcircuit that communicates w/ arduino? Would 7400 or 4000 series logic chips be fast enough? There's a competition from dangerous prototypes and adafruit for making something interesting using logic chips...

Also, would an fpga be more appropriate for this? I've been dying to do something with them, as many of you may know.

Let me start off by saying that I am in now way an expert. Now for the op; I would assume that your impedance is a result of the electromotive action of putting a frequency through the coil. I would treat your wire as an inductor. The electromotive action of current running through the line is probably reflected on it self?? So I rummaged through my school worksheets (because I kept them) and found my crib sheet for series rc and rl formulas.

Here is what I'm thinking for a simple set up. Set up a generator for say 1khz sine wave (don't know if an arduino can do that with just pwm pins in a certain formula??) 5v would be your ea?? I don't remmber how I did it on the generator in the class but I got the impedence of 40 ohms. Soo I have a known resistor, a known rg (resistance of the generator); a known freq; a known ea; and an unknown inductor. You measure the voltage drop on the known resistor to get eR1. Divide eR1 by R1 to get iT (total current in series circut). Multiply the rg by iT to get eg. Add eg with eR1 to get eR. The according to my crib sheet (I think) it is el=√(ea²-eR²) Then el/iT gets you xl. l=(xl/2(pi)F).

You might want to test that on a known inductor to see if that is right. I think that is how we did it in class.

If the arduino can do a sine wave (pwm on two seperate transistors to change direction of flow??) and you program a known resistor with the rg factored in I don't see why you couldn't just do an analog read in to get the voltage drop on the inductor. Sorry for spelling errors.

p.s. I wrote that at like midnight on sleep deprivation. You can also just use a dc rl time constraints. http://www.greenandwhite.net/~chbut/new_page_28.htm Seems that would be alot simpler. IMO

The issue here is that the coax has very low inductance (Pico henries) and it also exhibits a little capacitance. I think 1 kHz would be too little to "tease" out the characteristics. But, your idea is correct. I'm wondering if there exist digitally controlled capacitors or inductors to sweep impedances and frequencies; that would be neat.

I would say, keep the testing circuit as analog as you can, with the Arduino there just to manage the test. In theory, you are just generating a signal on the one end, and verifying it on the other, correct? If you setup a frequency generator, and then your test on the other end, you could go as low tech as using a motor to turn a knob to do the testing. A 1GHz processor is doing one *instruction* in a nanosecond, so you are not likely going to be doing anything digital there for a while.