Author Topic: Locating a failed buried wire using a TDR  (Read 5025 times)

0 Members and 1 Guest are viewing this topic.

Offline cosmicrayTopic starter

  • Frequent Contributor
  • **
  • Posts: 299
  • Country: us
Locating a failed buried wire using a TDR
« on: April 01, 2019, 08:43:52 pm »
I am attempting to troubleshoot a problem, that is of my own making. About 25 years back, I buried an electric branch circuit (220V 60Hz) to supply an out building and a 1-HP well motor. The wires were direct buried in a trench, and were of insulation type USE-2. Three of the wires were 4-ga AL and one was 6-ga AL. Wires were color coded for the two hot sides, one neutral and one ground. The length of the run is 550-600 feet, buried at code required depth of 24-inches. Branch circuit breaker is a paired 50A type. In addition to the power wires, there was a telecom 3-pair tar-filled cable in the same trench, but it was never placed into service.

About 4-5 years ago, one of the two hot wires failed, and behaved as tho it had developed an open circuit. At that time I swapped wires around, and brought the 4-ga neutral over to that hot position, and moved the failed wire into the neutral bus (on both ends). A quick fix, which seemed to solve the immediate problem, but did not reveal why one wire failed.

A couple of months ago, once again, one of the hot wires failed, with similar failure symptoms. Now I have nothing to fall back on, so it's time to sort out what happened and why, and how to go about resolving this. My best hope, is that I can identify where in that long run is the failure point, dig that up, and see what happened and why, then (if possible) splice around the problem. I am trying to avoid retrenching the entire run.

My best guess as to the reason for the failed wires is either a tree root, or possibly an induced lightning surge.

The tools I have available are a very basic DVM, and a 100MHz SIGENT O-scope. Checking the wires (disconnected at both ends) have yielded some very odd resistance readings to ground, suggesting that I have open insulation (and possibly moisture) somewhere along the cable runs. At least one wire was indicating ra esistance reading that mimicked a capacitor charging up. Calling around to the usual sources, I have not been able to locate a suitable TDR, so I have ordered the TDR kit from farcircuits. Not as good as a professional unit, but at least I can gleam some information about the reflected signals.

Does anyone have good ideas about how to troubleshoot this ? I am completely unclear about what the impedance of the buried wires would be, nor the velocity of propagation.
it's only funny until someone gets hurt, then it's hilarious - R. Rabbit
 

Offline JXL

  • Regular Contributor
  • *
  • Posts: 64
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #1 on: April 01, 2019, 11:08:29 pm »
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12288
  • Country: au
Re: Locating a failed buried wire using a TDR
« Reply #2 on: April 01, 2019, 11:41:28 pm »
Not sure how good your TDR solution will be - but I'd check from both ends of each wire to improve the accuracy ... or drive you nuts.
« Last Edit: April 01, 2019, 11:51:58 pm by Brumby »
 

Offline Gregg

  • Super Contributor
  • ***
  • Posts: 1124
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #3 on: April 01, 2019, 11:56:49 pm »
My guess is that there was water ingress into the faulty direct burial wire that caused corrosion of the aluminum and eventual failure under load.  If that is the case, a TDR may give lot of false reflections and a lot of frustration as the breach may not be a clean break and the ground around it may be conductive enough to mask the reflection.  An underground conductor locator would probably work much better, but a cheap one may not be much better than the TDR.  Maybe using two methods and digging where they both indicate would help.
You might be better to employ a professional underground locator service that has the good equipment and lots of experience.  There is as much art to underground locating as there is science.  The good locating devices are over $500 from what a friend of mine who locates for a large telecom company told me.
 

Offline notsob

  • Frequent Contributor
  • **
  • Posts: 690
  • Country: au
Re: Locating a failed buried wire using a TDR
« Reply #4 on: April 01, 2019, 11:57:22 pm »
Here's an old way to find it - may or may not work depending on your situation.

Isolate the cable run.

hook it up to the spark plug on a combustion engine (motor bike or whatever).

Walk over the area with an AM radio as your detector.

Cheap method, maybe worth a try
 

Offline L_Euler

  • Regular Contributor
  • *
  • Posts: 86
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #5 on: April 02, 2019, 01:16:20 am »
I don’t think a TDR is going to be the answer. I would try searching for voltage with a multimeter that has the Negetive lead tied to a long piece of wire back to the CB box neutral and the other tied to a ground probe. Start poking in the dirt to find max AC voltage. If you don’t have a multimeter with a low impedance input mode it will help to tie a 1000 ohm or so resistor across the input jacks. Might want to wear appropriate PPE also.
« Last Edit: April 02, 2019, 01:21:53 am by L_Euler »
There's no point to getting old if you don't have stories.
 

Offline TERRA Operative

  • Super Contributor
  • ***
  • Posts: 2871
  • Country: jp
  • Voider of warranties
    • Near Far Media Youtube
Re: Locating a failed buried wire using a TDR
« Reply #6 on: April 02, 2019, 02:36:28 am »
You could always hook up an arc welder to the ends of the cable then look for the steam... :D

Being direct buried, you'll probably be looking at digging a new trench to replace the cable anyway, I would suggest using a largeish conduit this time to help protect the cable and make future replacement much easier.
Where does all this test equipment keep coming from?!?

https://www.youtube.com/NearFarMedia/
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 9234
  • Country: us
  • $
Re: Locating a failed buried wire using a TDR
« Reply #7 on: April 02, 2019, 03:17:53 am »
im very curious to know what kind of results a TDR would deliver in this situation
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28136
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Locating a failed buried wire using a TDR
« Reply #8 on: April 02, 2019, 04:32:57 am »
im very curious to know what kind of results a TDR would deliver in this situation
Precise with the right tool.

Google: DTF measurement

Still, my first port of call would be a sparky with a mains cable locator and follow the line of the hit until it stops.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 6733
  • Country: pl
Re: Locating a failed buried wire using a TDR
« Reply #9 on: April 02, 2019, 07:54:57 am »
I have ordered the TDR kit from farcircuits. Not as good as a professional unit, but at least I can gleam some information about the reflected signals.

Does anyone have good ideas about how to troubleshoot this ? I am completely unclear about what the impedance of the buried wires would be, nor the velocity of propagation.
Might work, I have used similar circuit to locate faults in headphone cables.
Impedance doesn't matter, the initial level will just be different than 2.5V, no big deal.
If the cable is lossy at RF (I think, or maybe there's some other reason for that?) you may see a slowly rising ramp rather than flat top, ignore ;)
To determine velocity factor, send a pulse down the remaining good conductor pair.
 

Offline Muttley Snickers

  • Supporter
  • ****
  • Posts: 2333
  • Country: au
  • Cursed: 679 times
Re: Locating a failed buried wire using a TDR
« Reply #10 on: April 02, 2019, 09:35:28 am »
I have both the Tektronix 1503C and a TS100 TDR here along with a variety of other testers and tracers which you would be welcome to borrow all but for the distance. Using a TDR an experienced operator will be able to detect and determine the type of fault or faults present and the distance to such faults provided they know the cables velocity of propagation or can calculate it by means of the existing cables physical length.

A TDR cannot display where a cable fault or junction is located only the distance to it based on certain parameters and as previously mentioned where a cable incorporates junctions or is saturated at multiple points determining the exact location of faults will be a tricky process. Half the battle for these types of jobs is determining the location and direction of the original cable trench and any branch feeds.

Based on the information provided in the first post I strongly suspect that these cables are now saturated and in need of replacement, even if the current faults were located and repaired sooner or later similar problems are likely to return elsewhere, also If you do need to dig a new trench and replace these cables be sure to place them in proper watertight conduits.
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 4136
  • Country: gb
Re: Locating a failed buried wire using a TDR
« Reply #11 on: April 02, 2019, 10:03:28 am »
A TDR cannot display where a cable fault or junction is located only the distance to it based on certain parameters and as previously mentioned where a cable incorporates junctions or is saturated at multiple points determining the exact location of faults will be a tricky process. Half the battle for these types of jobs is determining the location and direction of the original cable trench and any branch feeds.

What Muttley said, it's not likely to be an easy job for a TDR because your cable has failed once already and it's possibly damaged in several places as there's no guarantee it failed in the same place, it's going to be further complicated due to water ingress and mechanical factors, sharp bends, direction changes, even the moisture level in the ground can make a difference to how the TDR 'sees' the cable if the insulation is compromised.

Personally, I think I'd bite the bullet and replace the lot, if I could afford to I'd also run a conduit so the cable can be pulled and replaced with relative ease next time.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: Locating a failed buried wire using a TDR
« Reply #12 on: April 02, 2019, 10:21:00 am »
With a 100 MHz oscilloscope you have almost enough for a basic TDR.  Connect a pulse generator to the pair you want to measure through a resistor of about the estimated impedance and connect the high impedance x10 oscilloscope probe in parallel to the pair.  This will not be calibrated but the reflections and approximate relative impedance will be apparent.  A small 1 kilohm variable resistor will allow adjustment but is not really necessary.

The pulse generator needs to have a fast edge to deliver good resolution.  The propagation delay will be about 1.5 feet per nanosecond.  Any logic output will be sufficient like the sync output on a function generator.  The linear output of the function generator will likely be too slow.

A low frequency square wave makes the least confusing source.  The half period just needs to be longer than the propagation delay which will not be a problem.

Distance calibration might be possible using one of the good pairs.  Or make the measurement from both ends.

I agree with Muttley that simply finding the fault will not be helpful because the damage is likely to require complete replacement.
 

Offline cosmicrayTopic starter

  • Frequent Contributor
  • **
  • Posts: 299
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #13 on: April 22, 2019, 03:45:59 pm »
My current attempt to solve this issue goes like this ... I know the open wires show resistance between one another, both from the CD box and at the load end. The resistance is high enough to tell me it's not a dead short, but low enough to suggest a breach of the insulation somewhere. My recollection is that I was seeing a 10k-15k resistance between the wires. The calculated current flow is going to be way below what it would take to trip the CBs, but there should should be some flow.

Enter one other piece of test gear I had lying around, a Simpson Electric Amp Clap. The lowest scale on the Amp Clamp will read up to 5 amps AC current on the 2.5 volt AC scale on a Simpson 260. As a test (at a different location), I tried using jumper clips between the amp clamp output and the O-scope. Sure enough, the scope is sensitive enough to see reasonably low levels of AC voltage (which the clamp is capturing from current flow).

I have dug several inspection holes over the buried wire path. The plan is to measure the leakage current at the feed FB box, then move down the line and see if similar values can be seen at each inspection hole. The Amp Clamp should save me from having to breach the wire's insulation to look at resistance and/or voltage. If I hit a wire location that shows no leakage current flow, then I need to look back up the line for a possible reason for the wires to have failed. At least this is the plan. I'll check back in a few days with progress.
it's only funny until someone gets hurt, then it's hilarious - R. Rabbit
 

Offline Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 3321
  • Country: nl
Re: Locating a failed buried wire using a TDR
« Reply #14 on: April 22, 2019, 03:57:42 pm »
 I like the idea of the spark plug with AM radio and it just might work.

I also have an UT210E current clamp which has a resolution of 1mA and both AC and DC ranges.
It also has a NCV measurement (as lots of other DMM's have) and you can also try using that for locating the open connection.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #15 on: April 22, 2019, 03:59:17 pm »
While locating the fault sounds like a fun excercise, I don't see how you will ultimately repair faulty direct burial cable without replacing the whole run anyway. I would run new wire in conduit if faced with this situation.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Locating a failed buried wire using a TDR
« Reply #16 on: April 22, 2019, 04:26:45 pm »
I agree, the cable is already buried, it just doesn't know that it's dead yet!   :D

However, one very basic idea that might yield some sort of result...

On the assumption that moisture has got into the cable at some point along its length, it must have got in though some defect in the outer covering. Assuming that the defect is still damp, there should be some sort of conductive path to ground at that point.

Try isolating and commoning the conductors at both ends and taking resistance measurements between them and a ground spike, stuck into the ground at intervals along the path of the cable. Assuming that you can get some sort of resistance reading, the lowest resistance point should be in the vicinity of the damage. [EDIT: It could equally indicate a damper area of ground of course!]

It's only going to be helpful in salvaging the cable if the problem is actual mechanical damage at a particular point. If so, you'll need to take a decent chunk out to eliminate moisture creep and make really good waterproof splices.

If the ground is dry, you may need to use more voltage than the DMM resistance range provides and use a low current range but it could be worth a try. All you've got to loose at this stage is a 300ft reel of thin insulated wire (test from both ends).
« Last Edit: April 22, 2019, 04:30:27 pm by Gyro »
Best Regards, Chris
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Locating a failed buried wire using a TDR
« Reply #17 on: April 22, 2019, 04:36:03 pm »
Why would you even want to search where the problem is, after 25 years of rot?  Replace the POS cable right away. Otherwise you will be looking for short/open next year again.
 

Offline tecman

  • Frequent Contributor
  • **
  • Posts: 444
  • Country: us
Re: Locating a failed buried wire using a TDR
« Reply #18 on: April 22, 2019, 05:12:17 pm »
A TDR is the right tool.  Even if you cannot determine the velocity factor, you should be able to determine the fault distance as a percentage of the overall length.  Sometimes this means measuring separately from each end, and determining the ratio of the fault to the overall distance.  Since you know (I assume) the overall length and location, knowing the percent of the distance to the fault will get you quite close. 

My guess is either lightning or water intrusion has caused your fault.

Paul
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Locating a failed buried wire using a TDR
« Reply #19 on: April 22, 2019, 05:28:33 pm »
im very curious to know what kind of results a TDR would deliver in this situation

Me too, presumably with a short impedance of several k \$\Omega\$ the reflection would be very low-Q?
Best Regards, Chris
 

Offline Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 3321
  • Country: nl
Re: Locating a failed buried wire using a TDR
« Reply #20 on: April 22, 2019, 05:33:45 pm »
It's common to have splits in underground cables.
They don't run separate cables from each house to the power company.
Usually these splits (or repairs) involve epoxy.

Whether a 25 year old cable should be replaced is another question.
I'm certain though that undergound cables in the city where I live do not all get replaced every 25 years.

Locating the fault seems a logic first step. If the fault is caused by some external event or by simple rot of the cable could make clear what the next step should be.

Just curious: Does anybody know what the life expectancy is of underground power cables?
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: Locating a failed buried wire using a TDR
« Reply #21 on: April 22, 2019, 05:46:47 pm »
Quote
Just curious: Does anybody know what the life expectancy is of underground power cables?

In our street, about 15 years!  ::)

The cable is plastic outer coated, with Aluminium sheath inside and 3 phase Paper insulated cores inside that. The house builders weren't careful about handling and bend radius when they installed it. This lead to cracking and water ingress. We've been suffering power outages once or twice a year since the first failures (at about 15 years).

The electricity company comes out each time, replaces the substation fuse, then, when the fault gets more permanent, fit an automatic re-closer/recorder until it becomes a hard fault. Then they use a combination of TDR and sniff test to locate the position (our pavements now have rows of little holes in the tarmac). Then they dig down, splice in a new length of a few feet with epoxy jointing boxes at each end, leave it for a week, and then refill to hole and re-tarmac.

The whole thing should have been replaced years ago but that would involve massive disruption and reconnection of the tails (presumably also paper insulated) to each house!

P.S. We had two faults outside our house alone - they flexed one of the ends of the old cable a bit too much when splicing and it failed a metre further along about a year later. At least we have a very neat section of tarmac pavement outside.
« Last Edit: April 22, 2019, 05:49:37 pm by Gyro »
Best Regards, Chris
 

Offline soldar

  • Super Contributor
  • ***
  • Posts: 3128
  • Country: es
Re: Locating a failed buried wire using a TDR
« Reply #22 on: April 22, 2019, 05:51:02 pm »
You might spend a lot of time and effort to patch here and a short time later it fails again elsewhere. With a little labor and not much money you can do a really good job which will allow repair and upgrades.

What I would do is bury conduit which is larger than needed. That allows for additional cables, repairs etc.

No bends. Straight and ending in a register well which can be brick, masonry or just a plastic box like used in sprinkler systems. From there another conduit to wherever. That way you will never have a cable stuck. And I speak from experience. In Spain they always put conduit which is too narrow and has too many curves and most of the time you cannot introduce another cable or even get out the original one. It is really stupid.

If you bury conduit with upwards bends at the ends chances are water will accumulate in there.

While you are at it you might want to install two conduits so you can separate power from data. This might be required by code but it is a good idea anyway.

It is not complicated and if you do it right you will save yourself a lot of aggravation in the future.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline chris_leyson

  • Super Contributor
  • ***
  • Posts: 1541
  • Country: wales
Re: Locating a failed buried wire using a TDR
« Reply #23 on: April 22, 2019, 08:09:22 pm »
Over a few weeks some gas contractors have been digging up a street near me. All over the pavement on both sides of the street they've marked LV and HV. LV markings are for the most part along the pavement within a few feet of property boundaries, whereas HV markings are all over the shop, they're under the pavement or sometimes in the street. They're just saying to the contractors "don't dig here" maybe they got that from local council plans or maybe they got it from a guy sweeping the area with a detector.
 

Offline Stray Electron

  • Super Contributor
  • ***
  • Posts: 2039
Re: Locating a failed buried wire using a TDR
« Reply #24 on: April 22, 2019, 11:26:14 pm »


I agree with Muttley that simply finding the fault will not be helpful because the damage is likely to require complete replacement.

   I agree. If the circuit has failed twice already, the rest wouldn't be far behind.  In addition, code does not allow you to splice wires outside of a junction box, it has to be one continues cable.  And trust me, I've had to replace underground wiring that was spliced underground without a sealed box, it doesn't last long!

   I know it's painful but I think your best bet is to just replace the entire cable.  And I wouldn't use aluminium wire underground.   

   Youtube has a couple of good videos on using a TDR (a simple scope and a home made pulse generator) and you can find the electrical velocity by measuring the remaining good conductors. You can also measure the broken conductors from each end to get a more accurate measurement of where the open circuits are.  It might not give you the exact distance but it will be a lot closer than just having to dig up the entire cable.

   FYI one of my buddies uses an AM radio to locate broken underground cables and it works for him.  He doesn't even connect any kind of source to the cable, he just listens to the static as he walks the length of the cable and he can hear the difference when he gets to the break.  He tells me that he's even located broken water pipes that way.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf