Time domain reflectometry, it's called, and it's been used for yonks in both electrical and optical versions.
See e.g. this nice Agilent app note:
A Tektronix 11800 series scope with an SD-24 plugin is the bee's knees for TDR.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net
There are also lots of small LCD handheld TDRs with slower risetimes and kilometers of range, for checking cable TV and phone wires and such.
Velocity should be pretty much constant for any known cable type.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
Although I have done this many times myself, I thought this guy did very well in this demo.
Years ago I had coax going up to the roof and it was in the walls to make things look pretty on the out side.. I had a DC short and I wasn't going to tear out all the walls to find it. I shot a pulse up it, did a rough calculation and found it on the second floor. It turns out the electric company had put in new wire on the outside and a strap fastener poked into the wire. I got with in foot where it was.
But over the years of doing this I've always question this practice. Can we rely on velocity being a constant? I know where I work currently we make many different communication cables and one the factors is chemistry change in the dialectic with age, especially with foams. This also effects the impedance.
I know recently I made an inquiry here to see if any actually uses that method. I was going to implement it in a cable debugger tool but then I realized the velocity isn't a constant due to inconsistent geometry.
Just something to think about I guess.
Jamie
What's really neat and I find amazing is sending an optical pulse down and back 50+ miles through a glass fiber. It is amazing how clear that glass is.
With some fiber, the twist in the bundles adds measurable length to the physical cable that one must account for when looking for a fault.
tm
We have HP TDR's at work in the lab, both optical and electrical. But at home, I do not have one, although I have been offered one that is no longer being used in the lab, maybe I should add that to my collection here at home :)
Jamie
You would think that however, we've tested products that have been in storage (samples) from years ago and found the Vfactor to change, due to the dialectic aging. It does not happen to all of the compounds used, just a couple that had been used for coax cables and such.. With foam, the Z changes and some grades will actually shrink in size..
It gets real bad when you check your cable and all you have is wire in the center with flatten braid and nothing but soup between.
Jamie
Really? I can only assume you are referring to fiber that is still sitting on the roll with no body bindings on it?
You're talking 264K feet of fiber or, half that for reflection time, but still. 132k ?
we run like 1 mile lengths bunched with a bundle of them and a TDR shows losses.
Jamie
No, I am saying it is a 50 mile long, installed cable with 216 single mode fibers. The OTDR pulse travels 50 miles out and 50 miles back and you can see every splice in the return waveform.
The fiber cable is installed in 2 inch buried conduits with access pull boxes every 1000 feet or so.
Regards, tm
This might help interpret the display. See the section "Looking at the TDR Signals" near the bottom.
My favorite screwup is using T-25 staples to secure a run of RG58/u or CAT5 cable, and puncturing the jacket at some point along the way. A visual inspection is no fun. So, I use a TDR. I can usually "nail" the length to within a foot or two.
Incidentally, I've built several TDR pulse generators along the lines of the schematic at the top of the page. No need to drag along a function generator. Incidentally, the scope calibration output is usually worthless due to high output impedance, slow risetime, and no control over frequency.
Constant from cable to cable or constant along the length of the coax cable? As far as measuring cable lengths, characteristic impedance, and locating shorts, it's quite constant along a length of cable. However, different cables have different dielectrics, properties, configurations, and therefore different velocity factors (VF). If I'm not certain about the type of coax (many are not labeled), then I take a known length, measure the propagation delay, and calculate the VF. That also gives me a clue as to what flavor of coax I dragged home from the hamfest.
Twisted pair and network cable has more variation in VF. The cheap junk varies all over the place. Capacitance can be anything between
13.5 and 17pf and be considered "good". The nominal VF of 0.7 for CAT5 will vary (not sure how much).You may be thinking of laboratory tests and phased matched cables. A TDR is not normally a precision instrument. The range of error caused by age, water absorption, contamination, and people walking on the coax, is not going to make much difference to the TDR. Two decimal place accuracy is great and even one decimal place will work.
It's in every network cable certifier and some testers on the market. It's also in every OTDR (optical TDR) for testing fiber. Anything that claims to measure cable length probably uses some flavor of TDR. If you have room for more feature bloat, by all means, please do cram in a TDR. It's very handy.
You can also do it in the frequency domain. Just add an FFT and GHz bandwidth: etc...
I'm thinking about dinner. Gone...
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
In Cat5 that doesn't have individual pair shielding, like UTP or F/UTP, each pair usually has a different twist, to reduce crosstalk. That makes each pair have a different velocity, numbers like 45 ns skew per 100 m, ballpark 10%. Sometimes even individually-shielded-pair cables have different twists.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
This one is for mere mortals such as meself.
$160 to $200. The problem with it is that you don't have the benefit of seeing what's happening which requires a scope. All you get is a cable length number. No clue if it's intermittent, leaking, shorted or open, misterminated, etc. It won't find missing T-connectors in
10base2 (cheapernet), or crushed cables.Accuracy is rather lousy: ± 2 feet Cables up to 10 ft, ± 5 feet Cables 10 to 200 ft, ± 3% - Cables longer than 200 ft I can usuallly do better than that with a decent scope.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
That sounds like Jersey Specialty Company crap wire. The worst crap wire that I've ever run into.
CATV trunk & feeder lines have a lot of expansion loops, which at the the actual VS physical length.
Nah, they're pretty cheap if you're patient--I got an 11801C with SD-24 and SD-26 plugins included for $1600 last year. That's about 3 or 4 cents on the dollar. I've been paying about $100 for SD-24s lately, which is less than 2 cents on the dollar.
I've got a metric pantload of plugins, I think 18 in all, including my latest prize, an SD-48 30 GHz OE converter that I paid $375 for. I still want one of the passthrough sampling heads, but otherwise I'm pretty well done except for replacements.
Did I mention that I love those things? ;)
I actually have a project coming up that will need most of the capability of the 11800s -- a coherent lidar system for looking at single particles down to 0.15 micron diameter, moving at up to 3 km/s. It's a follow-on to the ISICL sensor from 20 years ago. I obviously won't be able to see individual particles with a sampling scope, because they're nonrepetitive, but I'll need it to tune up the sensor and the back end. Fun stuff.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net
Oh, I certainly think you should!
I've done this sort of TDR with a scope-and-pulse-generator lashup to help characterize some of the antenna lines at our city EOC... a crude setup but it worked very nicely for the job. Then, a couple of years ago, I was fortunate enough to pick up a Tek 7S12 plugin with the necessary sampler heads for a very nice price (and won an S53 trigger recognizer head on eBay also at a nice price). Makes a very nice combination... being able to see impedance bumps a fraction of an inch apart is quite an experience!
--
Dave Platt AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
We have HV bridge fault analyzers that seem to work very well for locating shorts and opens.. But looking for a way to do it with out HV is always a plus.
In any case, it can find the defect with in 4-5 feet on lengths up in the 10's of thousands of feet! The method is archaic, but it works. There is a 10 turn pot that you set to null the meter. This meter is in the bridge circuit. You note the reading, then reverse the test leads and do this again.. This gives you the % of the real length to find it.
If you are dealing with an open, the process is the same, except now you need to apply HV to form a bridge.
Jamie
There are fiber optic perimeter detection systems using TDR. They can sense when a buried cable is driven over or a cable on a fence is jiggled.
.
I believe it read somewhere that the railways use (or work on) using a fiber optic cable along the rail to detect where trains are on the track by measuring the distance to where the fiber is bend by the weight of the train
-Lasse
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.