Sort of working through something mentally, and it has me stumped.

I was thinking of those cheap electric fence testers that light up xenon bulbs to show you how much power an electric fence has and I was curious the best way to cut down the ~15kV fence voltage to something more manageable. I don't have one currently, but I know the cheap testers are just a bunch of resistors in series/parallel and a few bulbs. That approach (a bunch of resistors) may not be the best way, but it is the cheapest I am sure to keep their costs down.

But I can't seem to wrap my head around how they do it. I immediately thought about a voltage divider, but taking 15kV down to something like 100v would require a REALLY large resistor that could handle 20+Watts (thought for a very short amount of time; a r=10mohm = 1.5mA = 22W going through that resistor).

Any clue the best way to go about it (even if it isn't the way the cheapo meters do it)? I am intrigued now.

 

Why do you want to reduce 15kV to 100V?

 

Why do you want to reduce 15kV to 100V?

The electric fences operate at 15kV, but if you wanted to do anything with that (like monitor it), it would need to be MUCH lower. I just picked 100V as a safe value.

 

The first thing to consider is the arc-over voltage of resistors. I just looked up some resistors and found 500V for 1 watt, 700V for 2 watt. The next thing to consider is the impedance of the meter. I have a 10 megohm Fluke. If I wanted to read 150 volts to represent 15,000 volts, I would need Ohm's Law. That's 15 microamps to power my meter. 14850 volts delivering 15 microamps must need 990 megohms in series with my meter. The power involved in this is .225 watts. (Watt's Law)

Apparently I think differently from MrChips, but that is neither good nor bad. It's just different.

 

100:1 and 1000:1 scope probes are readily available which would make measuring 15kV very easy.

http://www.home.agilent.com/en/pc-1659288/oscilloscope-passive-probes?&cc=US&lc=eng

 

1000:1 scope probes are readily available which would make measuring 15kV very easy.

"Only" $1533.00 from Newark.

 

100:1 and 1000:1 scope probes are readily available http://www.home.agilent.com/en/pc-1659288/oscilloscope-passive-probes?&cc=US&lc=eng

So are neon bulbs, but they are a lot cheaper.

 

If you want something in the way of a simple High impedance tester for 100v why not get the simple electricians neon indicator screwdriver/tester this just have a neon in the handle and relies on earth ground as the return path.
If it is not used on a earth ground referenced system, then you need two contacts instead of one, no resistors needed.

I brought this up in a recent posting on electric fences.
Max.

 

20kV 3W 1G ohm resistor. Ohmite
MOX-2N-131007JE

Or buy a Fluke HV probe.
http://www.fluke.com/fluke/usen/accessories/high-voltage-probes/80k-15.htm?PID=55356
http://www.fluke.com/fluke/usen/accessories/high-voltage-probes/80k-40.htm?PID=55357

 

Sort of working through something mentally, and it has me stumped.

I was thinking of those cheap electric fence testers that light up xenon bulbs to show you how much power an electric fence has and I was curious the best way to cut down the ~15kV fence voltage to something more manageable. I don't have one currently, but I know the cheap testers are just a bunch of resistors in series/parallel and a few bulbs. That approach (a bunch of resistors) may not be the best way, but it is the cheapest I am sure to keep their costs down.

But I can't seem to wrap my head around how they do it. I immediately thought about a voltage divider, but taking 15kV down to something like 100v would require a REALLY large resistor that could handle 20+Watts (thought for a very short amount of time; a r=10mohm = 1.5mA = 22W going through that resistor).

Any clue the best way to go about it (even if it isn't the way the cheapo meters do it)? I am intrigued now.

Chances are a few kV will simply crack-over around the outside of a single resistor - its customary to use a few resistors at least 10M each in series.

Unless you're tampering with the electrical supply local substation, a resistor divider that dissipates 22W will pretty much collapse your HV generator long before it gets anywhere near.

 

The problem is that a 15kV fence power supply, while delivering a lot of voltage, does not supply a lot of current.

You will have great difficulty monitoring the voltage on a voltmeter, even with a proper high voltage attenuator. You need an electrostatic meter or electrometer.

 

Could you just do the old spark-gap test? The maximum distance the spark jumps determines the voltage.

 

Fishock sells some inexpensive testers:

Blinks for under $15 (I have this one)

Digital meter good up to 19.9KV for 50 bucks.


http://www.fishock.com/store/electric-fence-accessories/other/testing

 

Or buy a Fluke HV probe.
http://www.fluke.com/fluke/usen/accessories/high-voltage-probes/80k-15.htm?PID=55356
http://www.fluke.com/fluke/usen/accessories/high-voltage-probes/80k-40.htm?PID=55357

As pricey as they are, they keep you safe, and that is worth a lot of money.

You can build one with PVC pipe and some resistors, but, in the end, you are betting your life on something you guessed at.

 

As pricey as they are, they keep you safe, and that is worth a lot of money.

You can build one with PVC pipe and some resistors, but, in the end, you are betting your life on something you guessed at.

True in most HV work, when significant energy is involved, but I think an electric fence charger is not lethal... it better not be.

 

This is more of a mental exercise than anything.

What I was envisioning was something like a voltage divider to knock the voltage down to something like 100V. Then using a cap rated for 250V to act as a "holding cap." I was basically thinking of it being a 1-bit sample-and hold.

If I had 3 or 4 of those voltage dividers at different levels, I could "bin" the voltage on the charger into a particular voltage range and read the caps off via microcontroller.

I am sure I am oversimplifying it and that the ultra fast high voltage, coupled with the low current spike isn't capture-able in this way......

 

It isn't really all that ultra-fast that it can't be done, but 10 people just wasted their time answering the wrong question.

 

10 people just wasted their time answering the wrong question.

A mea culpa is in order. I don't think people answered the wrong question, I was just trying (poorly) to clarify a little more my thoughts. I know about the cheap stuff that is out there and that I couldn't beat that price, I was just curious and wrote a poorly worded response as a followup. Sorry.