Electric Fence Indicator/Display.

[rfranzk]

Hello all,
  I want to build a two or three light display that is driven by the fence charger and can be seen from 1/4 to 1/2 mile away.  I was thinking of using a voltage divider network and either xenon or neon tubes with reflectors and lenses. The bulbs I am considering are neon standard brightness (http://www.chml.com/products/pdf/5-4.pdf) that draw .3 ma current and breakdown at 115vdc.   I would like to have an indicator of 2kv, (red lens) 4kv (Yellow lens) and 6kv (green lens) highest kv flashing all 3 lamps.
  The questions I have are related to current draw and load on the fence and also since the input voltage is what varies, how to calculate the divider circuit.  The input could be has high as 10 kv but more likely in the 7.5 to 8 kv range. I assume the divider should supply approximately 115 volts at each resistor to light the lamp.   If I calculate for an 8 kv input will the voltage increase be too much at the display with a 10 kv input causing bulb failure?  Will a partially shorted fence, ie weeds touching fence cause the voltage to drop below levels of flashing the 2kv lamp?  The voltage is not applied constantly but in 50 ms pulses at intervals of 500 ms to 1000 ms.
  I know the bulb current draw can load the fence and cause reduced effectiveness. If two lights are better for efficient fencer output I can live with that.  Or circuit that would only flash one light depending on input.
  Any recommendations would be greatly appreciated. 

Thanks in advance. rfranzk.

[Herman the German]

Hi,

I do not recommend using a voltage divider to indicate a functioning electric fence.

Normal resistors are designed for a maximum voltage of 300V. Using those for a voltage divider you have to take care not to exceed the maximum allowable voltage drop per resistor.

If you are going to use neon lamps try get some rated 50V/2mA (normally built into switches of electronic kitchen appliances). 

Wind 4 to 6 turns of insulated wire around the HV-line and connect both ends of the "transformer" to the neon lamp.

You will probably not be able to see the neon lamp flash during day time, but at night it should not not be a problem over a distance of 8km - a distance you can see a cigarette end glowing.

To increase intensity of the flash you could use reflectors as they are often used for LEDs.

HtG

[Hero999]

Resistors with voltage ratings of 10kV and up are available.

The break down voltage is listed on the datasheet as 90V to 135V but will vary widely depending on the age of the lamp, the temperature and ambient radiation levels, yes electromagnetic radiation can actually lower the breakdown voltage.

Loading the fence will reduce the voltage, possibly to a level below the bulb's breakdown voltage.

The bulb will not fail as long as the current is kept low enough and the duty cycle is low.

[rfranzk]

Thanks for the quick reply HTG,
  Great idea of winding a few turns around the fence a sort of current transformer.  Any recommendations on gauge of wire??  What about having multiple lights for multiple voltage outputs??   Would this be accomplished with multiple xformers, or one with multiple taps?  It may be possible to drive led arrays with reflectors.   I will experiment with this idea. 

Thanks again, rfranzk.

[Hero999]

I'm not sure if the current transformer idea would work as the current will be very low.

What you'll probably find is that if you wind the turns around the fence and connect one side of the neon to it and the other to earth, it will light due to capacitive coupling, assuming the insulation on the wire is thick enough to withstand 10kV, otherwise the insulation will fail and the lamp will divert all of the charge to earth.

Capacitive coupling is probably a good idea if you just want an indication that the fence is live: connect both of the neon's leads to earth and place the bulb near the wire but not close enough for it to spark and it should light by capacitive coupling.

[rfranzk]

Thanks for your input HERO999,
 I was curious about the low current as well. With capacitive coupling would you still recommend the 50v/2ma neon bulb?? 
  All I want is a visual of fence operation from a distance of 1/4 to 1/2 mile.  I would like to have a high and low voltage indicator to inform me of potential load on fence from weeds etc.  Would be nice if it was visible in daylight but not absolutely necessary.
  I would consider building a battery powered led system if the current transformer signal could be amplified to switch a semiconductor and if power consumption could be kept low. (Could use nicad or nimh pack and solar charger also)
  With the capacitive coupling arrangement, is gauge of wire important or just connected?   I have anything from 24 gauge to 2 gauge cable available. 
  With the transformer arrangement would fewer turns and a heavier cable provide more current, low enough voltage to drive led??  I also have 16 gauge high voltage wire that I think is rated for 10 kv. I will have to look. 
Lots to think about and experiment with,  Will be a week or so before I have time to do so. 
Could use one of these led type lenses with different color leds for voltages.

http://www.goldmine-elec-products.com/prodinfo.asp?number=G16752

Thank You so much for your response.  rfranzk

[Herman the German]

Hi rfranzk,

here is a circuit with an ultrabright blue LED (KINGBRIGHT) Using an OSRAM LED reflector you should see the device flashing even at daylight.

The PCB design is purely single sided.

The amplifier has a total gain of up to 1,104 and should suffice to amplify any voltage picked up by the "transformer". It is dimensioned to amplify 0.1mV from the transformer.

For the transformer use only cable with the appropriate voltage rating.

If you are interested to obtain the Eagle files please PM me your email address.

Regards

HtG

[Hero999]

I was curious about the low current as well.

If the fence is unloaded the current will be tiny. The only current flowing will be due to the leakage resistance, corona discharge and charging/discharging the fence's capacitance. The current should be much higher near the energiser than further away from it and will actually increase when a load is applied.

With capacitive coupling would you still recommend the 50v/2ma neon bulb??  

The neons you've linked to don't strike until the voltage is at least 90V.

 With the capacitive coupling arrangement, is gauge of wire important or just connected?   I have anything from 24 gauge to 2 gauge cable available.

The gauge shouldn't matter as the current is very low. You could also try just connecting both leads of the neon to the fence and it may even light due to the corona discharge. If it doesn't light put a grounded metallic object (wire or even the fence pole) near the neon but too far away for an arc to from and it should light.

 With the transformer arrangement would fewer turns and a heavier cable provide more current, low enough voltage to drive led??  I also have 16 gauge high voltage wire that I think is rated for 10 kv. I will have to look.  

That's true but too fewer turns and the voltage will be too low for the transformer to act like a constant current source and drive the required burden resistor or LED.

Current transformers are used for monitoring high current sources but an electric fence is a low current source, try it if you like but I don't think it'll work, certainly not by inductive coupling, capacitive maybe. If you wrap several turns of HT cable around the fence wire and connect both ends to a neon and it lights then I think it'll be down  capacitive coupling not inductive but if it works then so be it.

[rfranzk]

Hello and Thanks!!!
  Thanks HTG for the schematic and eagle files. Just had a chance to look at them this p.m.  A couple of questions.  On the resistors you don't have specified wattage.  I am assuming 1/4 watt is good for this??
On the caps, just a ceramic?? or is tantalum or some other type a better choice??  And last the schematic notes say to increase value of R4 to increase voltage. I assume this is the voltage of the trigger output signal??  And one more, would not adjustment of P1 change this setting also?? 
  I attempted to wrap 5 to 6 loops around fence and did not measure any voltage with a cheapo multimeter.  Will try it again with my good meter.  I see the design is scaled for .1mv input and my cheapo meter may not be sensitive enough to read this. 
  My plan is to breadboard this first and try it before I etch a board.  I will keep you posted on results.  Will probably be a couple of weeks. 

Many Thanks again and again.  rfranzk.

[Herman the German]

Hi rfranzk,

all resistors are standard 1/4W.

If you have the schematic loaded in Eagle you'll get info about every part.

Use the "I" symbol (I=info) and click the part you want to know more about.

If you click on C1 you'll get the info that it's an MKS-02 (made by WIMA). Caps are not critical in that application and you might use ceramic or polyester caps.

Tantalum and electrolytic caps are polarized and not to be used for AC which will be present at the circuit input with the shape of a pulse.

With a cheap DMM you won't be able to measure uV (0.1mV=100uV). You won't be able to measure that voltage even with a mid class DMM since it won't zero accurately.

Even with an oscilloscope you won't see any trace change in the Y-axis since most of the scopes require at least 2mV input.

Just amplify the signal and check the circuit output voltage. If it is not sufficient you might use 470K for R3 to obtain a total amplification of 1,849 (presently 1,104)

If all the changes are not successful you might increase the number of turns of the "transformer".

I haven't tested the circuit, but it should do well.

Regards

HtG