Newsgroups: sci.electronics.misc
Subject: Re: Pulse+scope=TDR?
References: <[email protected]>

Pulse+scope=TDR 
You are right on this.

[email protected] (Simon Lambert) writes:

> Working in the CCTV arena, a lot of 75ohm co-ax gets installed by
> engineers of varying care and competence. We can't check every yard of
> it, so use of a TDR seems attractive. Have you seen the prices on
> those things?!

Yes. They are pretty expensive.
 
> I have a 20MHz Fluke Scopemeter 123, so how do I acquire/build a
> device that delivers a pulse of nanoseconds duration into a low
> impedance cable? 

I have published one design in web at
http://www.hut.fi/Misc/Electronics/circuits/tdr.html
It can output 10 ns pulses.

> I would prefer to work on the basis of using discete
> echoes (in the manner of WW2 radar screens) rather than squint at
> superimposed reflections of simpler rising edges which make
> reflections from cable faults difficult to distinguish.
> 
> I guess a 10ns pulse would allow me to see wiggles in the otherwise
> flat scope trace (timebase at 50ns/div) which indicate that the
> installers have joined two lengths with a bit of terminal strip! A
> close-to-source dead-zone of ~10m would be acceptable, although I
> guess I could use the difference between inputs A & B to remove the
> source pulse from the trace thereby showing the dead-zone contents(?)
> Any creative thoughts (or 2nd-hand TDR's) welcome.

You could try my circuit described at
http://www.hut.fi/Misc/Electronics/circuits/tdr.html


-- 
Tomi Engdahl (http://www.iki.fi/then/)
Take a look at my electronics web links and documents at 
http://www.epanorama.net/