Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
I'm not familiar with "Sitemaster" products but the usual principle is to use TDR (Time Domain Reflectometry). Basically, you shoot a short pulse of voltage from one end of the cable and time how long it takes to bounce back. Then, knowing the velocity factor of the cable, you can work out the distance the pulse traveled and as it is a round trip, half the calculated distance is where the fault is.
Some instruments use FFT from network analyzer frequency sweep instead of actually doing time domain pulses.
What resolution/accuracy for location do you need, and what fault to do need to measure? Little difference in line impedance or massive fault like open/short?
I need distance in 5 to10 meter accuracy. Once i opened SITE Master and observed that at the input stage there was a VSWR bridge built around a ferrite-core.
- - - Updated - - -
The fault type is little difference in impedance.
- - - Updated - - -
Is it possible to do this using HF 40 MHZ frequency?
I never tried, but here are numbers for reference: My KC901S+ analyzer can do cable fault location to ~5cm resolution using frequency sweep up to 4GHz.
If we scale that by factor 100 then 5m resolution from a DC to 40MHz sweep might work.
Yes, the input data is from return loss measurement (complex S11).
For a simple bridge that works down to DC, maybe you can check out the presentation of the VNWA3 low cost network analyzer here: https://www.youtube.com/watch?v=C94J4AutCRc#action=share
Hardware block diagram including resistive bridge at 08:00
Either a time domain pulse is sent down the cable, and a return pulse is detected, OR a swept refelction coefficient test is done on the cable end and a fast forrier transform is performed to figure out its physical distance away
I'm familiar with the sampling scope method from long ago when I used TEK sampling scope for TDR, but that also required very fast analog front end for the sampler. I'm curious to see how this is solved now, 30 years later.
the use sampling diodes with ps switching speed, a hold capacitor, an SLOW analog ramp generator, and an op amp or two.
Attached is one paper...
I will try to find some better papers tonight, but they were used starting in the 1950's. Impulse radars today use them commonly, especially the commercial cheap ones. With a nonlinear transmission line generating the pulse to drive the sampling diodes, you can catch >100 GHz signals.
- - - Updated - - -
if you want to get fancy, you can use an IC to do the sampling:
With all these samplers...you have to conquer the spike noise from the sampling pulse leaking into the analog sampled output...usually done by some sort of balanced pulse driver and a ring of diodes
Thanks for reply.
For swept frequency approach what bandwidth and sweep rate is required? is it possible to use few MHZ bandwidth and very fast sweep using DDS?
Accuracy of locating the fault/reflection depends on bandwidth - more is better. "A few MHz" sounds not useful to me, as discussed above. Sweep rate doesn't matter.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
