Fence Wire Break Detector

[mrpackethead]

So, we have some very long fences, constructed of 7 high tensile wires..   Id' liek to be able to tell if the fence has been broken somwhere..      Ideas?  The wires can be as long as 3000meters, they are not electrically insulated.. 

[Rerouter]

run a sense line back to the source, and put a very small load on it, then measure the voltage, if you see a dip something has contacted the fence, if it spikes, someone has broken at least 1 of those wires.

something like a window comparator with the center reference being the feedback voltage through an RC filter, just because over the day the resistance of the fence will change, but never quickly.

[capt bullshot]

An idea comes to my mind:
Take an "impedance signature" of the know good wire as reference, then repeat taking signatures and compare them to the orignal one. Changes to the wire will result in different signature.
You could try this using TDR (time domain reflectometry) or feeding a constant wave signal into the wire, sweeping its frequency and measuring the complex response (what a impedance analyzer would do)
I don't think a lab TDR or Impedance Analyzer / VNA would be appropriate here, I'd try to build something on a simpler hardware

[Dubbie]

I suspect the fact that the wire is not insulated will make the results of TDR an unreadable mess.

[Ian.M]

What's the post material?  There's a big difference between 'not insulated' and 'low resistance short to each other and to ground'.

Any electrical break detection system will have to use AC excitation to avoid making the situation worse by inducing galvanic corrosion.

[beanflying]

A bridge of some sort is an option. Plenty of old school tech reading from 100+ years ago 

https://en.wikipedia.org/wiki/Murray_loop_bridge

[AndyC_772]

Do you need to know exactly where a break has occurred, or is a 'yes/no' output OK?

I don't see any issue with corrosion if the break detector operates with a low duty cycle; a 1ms test every couple of minutes isn't going to have any measurable effect whether it's ac, dc or some combination of the two.

Are the wires joined to anything at the far end? Each other? Ground? Could you make them into a loop, and simply pass a current from one to another? (You might wish you had an even number of wires, of course).

[danadak]

Measure the capacitance change in the line ? You would need
a reference site to eliminate environmental changes, like soil
conductivity, humidity, to correct the measured capacitance.
Create a correction table over time/weather/rainfall to apply to
measurements. Temperature and Humidity and soil conductivity
key variables to measure to build table.

Of the suggestions  posted TDR probably the best. Here again
a reference line to compare against would be useful. Line is
long enough I would think that one of the low cost ($ 20 to
$50) handheld oscilloscopes and the edge of a squarewave
might be adequate.


Regards, Dana.

[mrpackethead]

What's the post material?  There's a big difference between 'not insulated' and 'low resistance short to each other and to ground'.

Any electrical break detection system will have to use AC excitation to avoid making the situation worse by inducing galvanic corrosion.

Typically posts would be wood. But sometimes they are steel warratahs, sometimes plastic stakes, and sometimes concrete.   

[mrpackethead]

Do you need to know exactly where a break has occurred, or is a 'yes/no' output OK?

NOt really but that would be pretty cool to know. 

Are the wires joined to anything at the far end? Each other? Ground? Could you make them into a loop, and simply pass a current from one to another? (You might wish you had an even number of wires, of course).

They are typically wrapped around a big post ( called  strainer post ), 

[mrpackethead]

Interesting ideas..  No-bodys had the same idea that I have though..     

I thought about putting a strain guage inline with the wire.   One of the low cost bridge types.    Make a metal bar that the wire can attach to each side.. When the fence is strained up and tight it will read a certain value,  remove the strain ( because its now broken ) differnent value.

[Mr. Scram]

Interesting ideas..  No-bodys had the same idea that I have though..     

I thought about putting a strain guage inline with the wire.   One of the low cost bridge types.    Make a metal bar that the wire can attach to each side.. When the fence is strained up and tight it will read a certain value,  remove the strain ( because its now broken ) differnent value.

You're having mechanical ideas on an electronics forum. Do I need to say more? 

You need to be very sure strain gets transferred reliably over longer distances, or you need a lot of strain gauges. I can imagine a wire wrapped around a post or something else messes with your readings, as might corrosion or other wear.

[mrpackethead]

Its engineering. I don't distingush between displines when i'm problme solving. 

The wire is typically strained up to about 700N of force.  When it breaks theres no strain it in.   thats a wide differential to meaure between.

[Rerouter]

ok, so if you cut one of these wires, i would assume it would quite violently recoil towards the 2 ends? however what if they just cut the underlying fence and left the high tension wires, e.g. wrapped in fibreglass cloth or similar? as they would expect them to ping off quite violently.

[Mr. Scram]

Its engineering. I don't distingush between displines when i'm problme solving. 

The wire is typically strained up to about 700N of force.  When it breaks theres no strain it in.   thats a wide differential to meaure between.

You might not distinguish, but other obviously do

Is the differential reliable? So no chance of snapped wires being rusted to a fence or wrapped around a post or anything? Ideal conditions aren't quite the same as conditions after a year or so in the field, or when applied by a random outsider. Your variation in fence post construction suggests no full control over the way things are installed.

[tautech]

Its engineering. I don't distingush between displines when i'm problme solving. 

The wire is typically strained up to about 700N of force.  When it breaks theres no strain it in.   thats a wide differential to meaure between.

This will only work with short strains.

So, we have some very long fences, constructed of 7 high tensile wires.. ............  The wires can be as long as 3000meters, they are not electrically insulated.. 

7 wires indicates a battened fence that will continue to hold some strain even if all wires are cut in the middle.

Breaks indicate overstraining or some other fault in construction.
Recently I saw a professionally built fence that had had multiple repairs due to an incorrectly adjusted batten staple gun over punching staples and damaging the HT wire.

Yeah, I've done this stuff too for decades. 

[beanflying]

Drone with a suspended camera and or inductive loop 

[richard.cs]

Measuring resistance may still work despite the lack of insulation, the wooden posts are orders of magnitude higher impedance than the wire and you could possibly measure at a low enough voltage that corrosion and crud on the metal posts are sufficient to provide insulation - often that kind of rusty contact is high impedance until you get to a few hundred millivolts and then breaks down to just a few ohms (look up wetting current, etc.). If that doesn't work then you'd need a different approach on any fence using metal posts which is a pain.

[Ian.M]

Even a concrete post can be problematic if the reinforcement is exposed in more than one wire hole.

[BradC]

Interesting ideas..  No-bodys had the same idea that I have though..     

I thought about putting a strain guage inline with the wire.

You don't need to put it in line with the wire. Put a post up somewhere near the middle and put a strain gauge between each wire and the post (perpendicular to the wire, like a lever). Any significant change in wire tension either way will cause the gauge to deflect. Much more sensitive and also resilient.

This problem was solved years ago in the security industry. Commonly known as a taut wire fence. Designed to detect the minute deflection caused when trying to slip between the wires, but it'll certainly detect a snapped one.

[mrpackethead]

This will only work with short strains.

Because?  Its one continuous length of wire,  If it broken / cut somewhere down the length surely the tension cant' remain in it. ( granted i have not gone and practically tested this yet, its just been a mental experiement so far.    I'm curious as to what i've missed.  ( hence why i posted this in the first place! )

7 wires indicates a battened fence that will continue to hold some strain even if all wires are cut in the middle.

in deed, typical 7 wire farm fence, timber posts and battens.

Breaks indicate overstraining or some other fault in construction.

or it being deliberately cut, a car crashing into it, a tree falling on it.. and yes, failure due to construction.. ( i've spent several days fixing my own stuff ups, because i did'nt construct my strainer posts the right way.  ( they call us sunday farmers for a reason, lol.  )..   Nothign like a bit of experience to hone your fencing skills!

[Dubbie]

Like tautech, I have built a lot of fences in the past and the problem is the staples in the battens quite happily hold the tension on the wire. One of two don’t, but a couple of dozen do if they are overdriven. This is very easy to do. If you were building the fence new you might have a chance but an old fence seems like wishful thinking.

[tautech]

This will only work with short strains.

Because?  Its one continuous length of wire,  If it broken / cut somewhere down the length surely the tension cant' remain in it. ( granted i have not gone and practically tested this yet, its just been a mental experiement so far.    I'm curious as to what i've missed.  ( hence why i posted this in the first place! )

7 wires indicates a battened fence that will continue to hold some strain even if all wires are cut in the middle.

in deed, typical 7 wire farm fence, timber posts and battens.

Breaks indicate overstraining or some other fault in construction.

or it being deliberately cut, a car crashing into it, a tree falling on it.. and yes, failure due to construction.. ( i've spent several days fixing my own stuff ups, because i did'nt construct my strainer posts the right way.  ( they call us sunday farmers for a reason, lol.  )..   Nothign like a bit of experience to hone your fencing skills!

Sound fences are our only security and like you I have concern for their integrity. The route they take can have a lot to do with their general robustness from forces of nature or stock. Knowing stock habits and typical behaviour, even when stressed/driven has one hell of a lot to do with the longevity of a traditional NZ farm fence.
Present day materials, especially CCA treated posts and battens have an effect too, both galv wire and staples suffer from the copper in CCA treatment. For a while I used hot dipped post staples and they will typically last double the time of plain galv staples.

I've taken to adding a further wire to fences, a hot one from a real beefy mains unit. This keeps the stock from putting pressure on fences and discourages persons from unauthorised entry.

Still, you can't get away from occasional physical checks like they used to do in the old days. Hammer, strainers, wire and tools and ride the boundaries in particular. The best fencing advancement in recent times have been wire crimping systems like the Strainrite or Hayes products.
http://www.ezepullfencetools.co.nz/
http://www.hayesfencing.com/product.asp?categoryID=4

Getting back to some loop sense....there was a single staple insulator, its plastic part was within the staple and while today it's totally useless for high power energisers it would be OK to provide insulation for a return system.
Can't remember who made them.............