Intermittant Buried Cable Problem

High-current. Maybe you'll even see the smoke. ;-)

Time-domain-reflectometer.

A used TDR can be had for much less, and is a cool toy too. ;-)

A Few Legitimate Suggestions Offered.

--
  Keith

The installer doesn't guarantee his work, or did YOU install it ?:-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

Did the installer, or you put in a cable that was rated for underground use? Also, was the cable put in to a non-corrosive type of piping that is rated for underground cable?

There are standards for these cables. If they are not met, the settling of the ground will eventually tear or damage the cable. There is movement of rocks, and the earth itself that has a lot of force against things. There is also plant growth that can root around and tear a cable, and small animals that may borough through the ground, and chew on a cable.

There are really no easy shortcuts for this. If you have an opened connection, there are some sophisticated instruments, that can approximate where the break is. These use the technique of sending a pulse down the cable, and measuring the reflected open end pulse coming back. Then with knowing the exact characteristics of the cable, some calculations can be made to determine the approximate location of the open or break. With an underground cable, the result would be undeterminable, because it would be hard to predict the exact characteristics effect of the ground on the cable. Even the moisture in the ground at the time would effect the readings.

If the cable was in a pipe, then this may be more exact and give a better chance to analyse, because the characteristics of the pipe on the cable would be a fixed value to work with.

I have a strong feeling you are going to have to lay in a new cable. Also, you would be very wise to do the job properly using the proper rated piping that is able to withstand ground and rock movement, and earth pressure.

--

Jerry G.
=====


"Gary Lecomte"  wrote in message
news:ca76d1a9.0405231739.69b532b8@posting.google.com...
i have a Buried cable that is about 1/2 mile long. It contains 6 pairs
of #22 AWG wires wrapped in a shield and filled with gel. But some
wires are shorting out with others on occassion. And not necessarly
just pairs.

I believe I have Possible solutions to the problems below, but I would
like other opinions before giving mine.

1) How do you make these intermittant shorts, Permanent so they can be
more easily found?
2) How do you determine their location?

Digging up and/or replacing this cable is Not an option. It cost
$10,000 and is only two years old.

Any Legitmate Suggestions Appreciated.......Gary

...and use a pipe larger than the cable, all bend radii about 10 times cable diameter (or more), to make it easy to pull cable for replacement.

Perhaps you could fit some kind of data multiplexer at each end and use just a few pairs - swap them around until you find a reliable set.

At that cost I assume it is proper stuff so my first thought would be where did they joint it at install and dig there.

If there are no joints then you have to wonder if fixing it is sensible becasue it obvioulsy isn't up to the job so even if you isolate the problem and cure it I would have thought that there is a fair possibility of it going wrong again.

Out and return on half a mile will be about 5 us so put a pulse into a good pair that is shorted at the far end, put the same pulse into a shorted pair and the external trigger of an oscilloscope put the two traces of the oscilloscope on the two return wires you will receive two pulses that are time seperated from the trigger and each other in proportion to the distance from the source to the short. In theory :).

Before getting too carried away with fancy test equipment or not so fancy shovels, keep in mind that the vast majority of cabling problems occur at the very ends or someplace where extra hardware is attached (such as T's and the like). Typically the ends of the cable are easy to access, so before doing anything else I would (at the very least) inspect the end connections to make sure they are not the cause of your problems.

Thanks to the Honest Answers so far. To Further Answer your questions, as well as my idea:

1) This was a Direct Bury, Except where it comes out of the ground, where they used a 2" conduit at the ends to better protect it. Not Sure how long these are, going into the ground.

2)It was installed by a Professional Electrical Contractor and a good layer of sand was laid down first, followed with the cable and than a Foot more sand on top. I Extremely Doubt any rocks have damaged the cable.

3)There are no trees close to this cable, and this is in a Very Dry part of the Country. This is a Mountain Area with Very hard, Stable base under this cable.

4) I have a "Sencore", LC102 capacitor Inductor Analyzer. And according to the manual, if these shorts were not intermittant, I could determine the location of these shorts with an Inductance test. Unfortunately you need a 25 foot piece of cable to get a reference Inductance Value. And the town sent back all the Extra cable. Damm.

In my inital tests, I used an Ohm Meter, but was coming up with Varying Readings because of probable Contact Resistance at the point of the short. So I than introduced a 1 Amp, Constant Current Source. Measuring the voltage at the entry point and calculating the resistance verses the resistance of the wire, divided by two Because its a Round Trip, I determined one short to be 7 feet from the end. Repeated tests that day concured this. However a month later, that short No Longer Existed! I believe this must Temperature Related, causing Small Amounts of Contraction and Expansion. In all Probability I would think all these shorts are at the same point, But I need to Confirm it before digging this up.

My New Plan is to use a 150 volts or 300 volt @ 100 mA Current Limited supply. In hopes of Arcing and Fusing these bad spots. Maybe also a Cap to Give a High Current Burst of power but Hopefully, not enough to damage the other wires.

Does This sound Reasonable?

Thanks.....Gary

First off, you should be posting this to comp.dcom.cabling. You should get a cable tester that can do TDR and find out how far the intermittents are from the ends. This might give you a good idea of what the problem is. If you have it buried, I would guess that you might have a rodent problem. But it could be something else such as tree roots. In any case, if you have spare pairs, you're in luck. Just hope that the spares don't also go bad.

Jerry G. wrote: [snip]

Depends. It it's in a conduit, then pulling a new cable may be the cheaper option. If it's direct buried, then finding the breaks and repairing them will probably be cheaper.

Gary Lecomte wrote: (snip)

If this test result is at all valid, I would bet a dollar that the cable has been cut by the bottom end of the conduit. After the next time you have made a test, try pulling up, hard (or push down), on the conduit, and see if the resistance changes.

--
John Popelish

Go to a telephone supply house and get a cable fault tracer. Put the signal generator on the bad pair and follow the cable until the level changes - that's where the fault is.

You can also configure a good oscilloscope to make TDR measurements. Not as convenient as a des-gned-for-the-purpose TDR, but it works. I've used it to measure the length of coax cable run between floors of a building - impossible to measure otherwise.

More about me:

VB3/VB6/C/PowerBasic source code: Freeware for the Palm with NS Basic source code: Drivers for Pablo graphics tablet and JamCam cameras: johnecarter at@at mindspring dot.dot com. Fix the obvious to reply by email.

I agree with the principles already mentioned:

1. Check the obvious
2. Check the easiest (cable ends, conduit)
3. Time domain reflectometry will reflect a small pulse when the cable impedance changes (i.e. the cores get closer, or resistance changes). Hire one.
4. I don't recommend using high power to burn the bad joints - it may damage the rest of the pairs. Unless of course, you don't mind repairing the complete set of pairs. What if the complete cable is overheated as a result of this?
5. If the repair is 'mid cable run', then you will need some spare cable to make a repair, becuase you might not have any slack.
6. Determine the cause of failure (you don't mention if you are overloading the conductor and heating the insulation).

I hope that's enough, Best regards, Dave

when i worked on a TDR high voltage power line fualt detector equipment i heard about the traditional way of determining the position of fualts in long hi power cables. (for shorts anyway)

the cable is buried in sand and an extremely high current is put through the cable for a very short period.

the magnetic field produced by the current makes the cable jump and the sand is disturbed but only for the lenght of the cable up to the short.

Colin =^.^=

I remember seeing this conversation about a month ago. I would like to know how you resolved this, if you did by now...

Jerry G. snipped-for-privacy@hotmail.com

--