# High-Speed Digitizers

cancel
Showing results for
Did you mean:

## Measuring time delay caused by the cable

Hi everyone,

We have two parallel copper cable (a very standard, un-shielded,  with 1.5 mm diameter) lines with lengths of 2500 meters (about 1500 miles). Normally these lines are not connected to each other and suppose that there is no load at each end of these lines (open circuit conditions). What we want to measure is the distance where a short-circuit condition occurs or in other words an approximate localization of point where these lines touch each other.

In the case of short-circuit, these 2 parallel cables form 2 different loops. Suppose we have a measurement system at one end. We may send any signal from one cable and recieve it (which travels all the way to the short-circuit point on the first cable and enters at that point to the second cable) from the other cable. Based on the information that I have read from the Modular Instruments Course Manual, there will be an un-avoidable delay with the transmission of the electricity signals :

"...Electricity travels through a wire at approximately two-thirds the speed of light (200000 km/s - 0.2 m/nanoseconds), which means that it travels at approximately two-thirds a foot per nanosecond... "

Here comes my question : Is it possible to measure the delay which is introduced by the copper line to the signal transmission ? If it is so, we may approximate the length of the cable on which the transmission occurs (based on the measured delay) and simply find the point where the short-circuit has occured (by diving the distance by 2).

I know that HS Digitizers (such as PXI-5122) are capable of timestamping to nanosecond time precisions. Definition of TDC (time-to-digital converter) also implies that :

"...A digitizer with a TDC uses the TDC to measure the time between the trigger and the first sample. This allows for extremely accurate trigger measurement. For example, without a TDC, when using a 100-MHz sample clock, the trigger time would have a maximum error of one sample clock cycle or 10 nanoseconds. Some NI digitizers with a TDC can reduce
this error to 100 picoseconds or better..."

And we are also planning to use an Arbitrary Waveform generator (AWG) for tight synchronization of the signal generation and recieving. In that regard, TClk technique of the Synchronization and Memory Core (SMC) architecture is a suitable candidate.

However I haven't used those techniques before and I am not sure about the synchronization of these devices. Nevertheless, tenths of nanoseconds of synchronization error is acceptable which would yield  2 - 20 meters of localization error (less than %1 localization error for the overall line).

I need an approval of the usage for an HS-Digitizer and an AWG side-by-side in this application. This kind of application seems very rare and the physics of the system should be considered thoroughly.

All the valuable ideas are highly appreciated.

P.S. We are also thinking about to use a high precision DMM such like PXI-4071 for impedance or resistance measurement however the copper lines won't be in an isothermal enviroment, there may exist temperature differences up to 20oC and field conditions may also affect the impedance of the lines.

Best Regards,

Candan Caner

candancaner@yahoo.com

Message 1 of 3
(8,530 Views)

## Re: Measuring time delay caused by the cable

Candan,

Can you clarify whether your cable length is 2500 meters (1.55342798 miles) or 2,414,016 meters (1500 miles)? The length that you posted was inconsistent.

The technique that you are describing is called Time-Domain Reflectometry (TDR). Here are some resources that may prove useful in developing your system:

How do I Perform Time Domain Reflectometry (TDR) Analysis Using NI Products?
http://digital.ni.com/public.nsf/allkb/505248D86985B16486256BBA0075ECCC?OpenDocument

NI Developer Zone: Time-Domain Reflectometer (TDR)
http://sine.ni.com/devzone/cda/epd/p/id/3505

NI Developer Zone: Applying Limit Masks to Time-Domain Reflectometry (TDR) Measurements
http://zone.ni.com/devzone/cda/tut/p/id/3888

Wikipedia: Time-domain reflectometer
http://en.wikipedia.org/wiki/Time-domain_reflectometer

The second link above includes an example that uses a PXI-5122 and a PXI-5404 to create the Time-Domain Reflectomer. Please note that this example was designed for transmission line lengths of 3 feet - 180 feet and has not been tested with longer lengths.
Regards,

Chris Delvizis
National Instruments
Message 2 of 3
(8,513 Views)

Hi Chris,