Remote Electric Fence Monitoring System

Matthew H

Today I went up and back to my grandpa's farm to help with some maintenance work and got an idea for a project. Even though I will probably never build it, I thought I would bring it up to see how plausible it would be. First off, his farm is 3 hours away and we only go there for a few days at a time during the summer mainly because there is no shower or bath so we can't stand each other after a few days of hard work. And, during the winter it gets pretty cold there. He doesn't have any live stock, but he grows apples, strawberries, rhubarb, blackberries and more. His garden is surrounded by a electric fence on 3 sides and the 4th side is a 20ft fence put up by the state.

My idea is to build a system that would monitor all 5 of the fence's wires, at 2 points, one right next to the charger, and the other at the farthest point from the charger. The system would then upload this information to the internet, or just email him when the voltages drop.

The biggest problem to me would be measuring the fence's voltage at basically 10 points (2 locations x 5 wires). I believe the fence is normally about 6kv but to be safe I'll say 10kv would be the max. To me, the easiest way would be to buy 10 of these Electric Fence Testers and use photoresistors to determine which lights are on, and thus what the voltage is. But that would be a little to expensive by the time you buy ten of those.

The rest of the project should be easy. Use a few wireless modules to connect the the monitoring stations to main controller inside the house, and connect the main controller to a PINK module to get the data to the internet.

Let me know what you think, any suggestions are welcome.

Matthew

Beau Schwabe

Matthew H,
·
I'd double check the voltage just to be sure.... seems like huge effort just to monitor an electric fence... 5 wires? I'm a little confused about that.
·
Years ago, I had a friend who had a few miles worth of electric fence that required a custom design that I helped him with.· We had visual indicators at various locations along the fence using small (5 Watt) fluorescent bulbs with one side tied to the fence, and the other side left hanging down without a connection.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

kwinn

Please feel free to correct me if I am wrong, but from what little I know about electric fences they are not charged with 6KV (or 10KV) DC, but are pulsed with a high voltage somewhat like a car spark coil produces. If this is the case you could use a high voltage capacitor and resistors to couple the pulse to a 555 timer wired as a missing pulse detector. The output from the 555's could then drive an optoisolator (or 10 optoisolators) whose output is connected to your monitoring system.

Matthew H

Kwinn your right, the fence is pulsed about every 1 second, but I still think the pulse is about 6kv. I will check to confirm the fence's output when I get a chance later. I have never worked with 555 timers, but it sounds like the circuit you described would only detect whether the fence is 6kv, or something less. I would like measure the fence's voltage a little more precisely than that.

@Beau, I attached a picture to show what I mean by 5 wires, although the 5th may be a little hard to see. I know this is a big effort just to monitor an electric fence, but like I said in my first post, I will probably never build it. I just want to know how it could be done.

Matthew

Edit: I did some research and found that I was right about the fence's voltage, it's about 6kv when working properly.

Post Edited (Matthew H) : 12/31/2008 7:53:41 PM GMT

kwinn

Do you want to measure the voltage or just make sure it is above a minimum value?

Franklin

·http://www.zarebasystems.com/PRODUCTS/acc_alarm.aspx

http://www.zarebasystems.com/PRODUCTS/acc_alarm.aspx

The first might just do what you want without building anything (bummer) but with the second you could use the stamp to view which light is lit and thus know what the max voltage is. This is the one I use around here. If you tie all the hot wires together then they will all be a t the same potential and if one is shorted they all are but if they all go back to the same charger that is the case anyway.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Stephen

Matthew H

I would like to be able to measure it roughly. Basically do the same as the meter I linked to in my first post, but be able to get that reading to a micro controller. Although more precision is ok too.

Franklin, I have seen i few monitoring systems like that, but I would rather build one, after all what's the fun in just buying one. Is the second link supposed to be different, they both take me to the same page? All the hot wires are connected at the charger, so I guess I would only really need 1 meter at the charger's end. But by the time you get to the other end of the fence, if one wire is broken you might not be able to tell it with one meter connected to all of them.

Thanks, Matthew

LoopyByteloose

CANbus could easily monitor it at 10 or 20 places and feed into a microcontroller that either connects to the web [noparse][[/noparse]like the PINK] or makes a phone call. Since it only requires a twisted pair and uses RS-485 trancievers, it is very low cost for large distances involved in farm applications. And Microchip has complete small nodes with ADC input. You would just have to find a way to knock 6KV down to an appropriate digital input with resistors or Hall effect sensors.

The whole network would be merely a pair of 22 gauge wire and you might use small solar panels to trickle charge NiMH AA cells for each node's power. After all, I don't think any node requires more that 100ma at peak usage. And the nodes may be off 95% of the time.

Five wires could compose 5 bits of one status byte. You would have another 3 bytes to monitor the battery status.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
How do you like my name change?

aka G. Herzog [noparse][[/noparse] 黃鶴 ] in Taiwan

erco

Hope you don't·monitor one of THESE in you fence. Happy New Year!

erco

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

kwinn

Attached is a rough schematic of a monitor circuit for each end of the fence. Depending on the pulse width you may have to remove the 10K resistor and put a 5V zener in parallel with capacitor. Any 8 channel ADC or micro with 5 or more analog inputs should work. I would recommend a serial ADC since it is easier to wire.

Matthew H

Thanks a lot kwinn, that's what I was hoping someone would come up with. What input voltage should the ADC have?

Erco, I don't think I'll have to worry about any of those getting caught in the fence.

Loopy, actually I was thinking of a wireless solution like this transmitter and receiver from Spark Fun. They have a 500ft range at 12V but I only need about 250ft.

Thanks for all the help,
Matthew

kwinn

In theory the capacitor should get up to 6V for a 6KV pulse so you would need an ADC to measure 10 V, or reduce the 10K resistor to produce a voltage within the range of the ADC.

mahesh_s_patel

kwinn wrote: »

Attached is a rough schematic of a monitor circuit for each end of the fence. Depending on the pulse width you may have to remove the 10K resistor and put a 5V zener in parallel with capacitor. Any 8 channel ADC or micro with 5 or more analog inputs should work. I would recommend a serial ADC since it is easier to wire.

Hi can you please send schematic to me on mahesh_s_patel@yahoo.com

Thanks in advance.

LoopyByteloose

Rather than mess around with wiring directly into 6KV or more of electricity, it seems that the best solution would be hall-effect sensors measuring the magentic effect of the current through the 5 wires at each sampling location. No change of failure due to the add-on sensor devices.

While there are industrial hall effect current sensors that read up into the hundreds of amp, I suspect that a convention hallf-effect magentic proximity sensor might work quite well for the lower currents this is putting out. After all, you don't want an actual number... just an on or off condition indicator.

One Station can use on Propeller to monitor 5 wires, or more. You seem to have two stations in mind.

You can go wireless if you want, but RS422 will send and receive serial over 4000 feet and may be cheaper to deploy. You don't really need CANbus unless you want the added assurances that the network is working properly. CANbus would allow another message to indicate the communications wire had been broken. I still like hard-wire befter that wireless for reliablity.

And of course, you and send email to a modem via a Propeller Spiinerette to complete the project.

http://www.ebay.com/itm/Hall-Effect-Current-Sensor-For-Hobbyist-Arduino-Compatible-AC-DC-10A-75A/230991057211?rt=nc&_trksid=p2047675.m1851&_trkparms=aid%3D222002%26algo%3DSIC.FIT%26ao%3D1%26asc%3D163%26meid%3D77056418266994332%26pid%3D100005%26prg%3D1088%26rk%3D1%26rkt%3D4%26sd%3D230955026196%26

http://www.ebay.com/itm/Hall-Effect-Hall-sensor-Proximity-Switch-NPN-3-wires-normally-open-magnet-NEW/230979790688?rt=nc&_trksid=p2047675.m1851&_trkparms=aid%3D222002%26algo%3DSIC.FIT%26ao%3D1%26asc%3D163%26meid%3D77056418266994332%26pid%3D100005%26prg%3D1088%26rk%3D2%26rkt%3D4%26sd%3D230955026196%26

http://www.ebay.com/itm/Updated-Version-Hall-effect-Magnetic-detecting-sensor-module-detector/230955026196?rt=nc&_trksid=p2047675.m1851&_trkparms=aid%3D222002%26algo%3DSIC.FIT%26ao%3D1%26asc%3D163%26meid%3D77037616509200018%26pid%3D100005%26prg%3D1088%26rk%3D2%26rkt%3D4%26sd%3D221114239393%26

skylight

Quite an old thread, Just wondering if anyone has worked out how many pulses (once a second?) there have been since the original post?
Answers on a postcard.......