DC cable puzzle

If a wire is close to another wire with voltagae on it, there is a coupling effect, either capacitance or inductive like a transformer.

I doubt that 21 volts would shock you unless your skin is wet.Usually anything under 1 or 2 miliamps will not be very noticable as far as the shock.

You aren't seeing any capacitive effect there. That's got to be leakage through the soil. You might try measuring it then sprinkling it with water (assuming you know where the actual wire run is) - one would expect the leakage current to increase.

Use test instruments, you never know when the current may increase dramatically.

You can feel 12V DC with salt water and a few minor cuts.. 21 VAC is ~30 volts peak and it is probably enough to feel.

I'd categorically rule out coupling, it may be enough to cause audio hum in signal wire but 21VAC sounds unbelievable to me. I've wound my own induction coil and transformers, built more than a few capacitors for Tesla coils. It takes some serious magnetic fields (which fall off rapidly with separation in air - I assume soil would be similar) and a fair amount of surface area and very tiny space between plates, to make a small cap.

Damp soil and exposed conductors is a different story and more believable.

Right. This should be investigated. Sounds like bad insulation on the mains cable. Measure its current.

Thanks all, mains leakage sounds the culprit. I?ll get on it today. It?s a long run, with an intermediate outdooor mains socket, then a (supposedly) waterproof case with the relay.

Terry, East Grinstead, UK

Maybe something underground is full of water or ants or something. Possibly the wire insulation got nicked.

There are some fun instrumentation possibilities, like probing the soil at the surface and measuring/listening for AC potentials.

A map of the surface potentials of some chunk of land, color coded by frequency, would be fun.

We were on the roof yesterday and speculated how cool it would be if we could see RF, all the microwaves and cell phones and wifis and transmitters. That's not possible. But a surface potential map is.

I've often thought of that. You'd need eyes bigger than Zoey Daschanel and Irene Bordoni combined.

You know that famous chaotic function that can be modelled with a circuit of 2 multipliers and fed into a scope in XY mode to produce a pair of spirals? There's a complete circuit on Paul Horowitz's website. I might have made one for fun if the multipliers weren't $20 each.

Well, it would also be neat to put one in the cabinet under that old scope in your lobby.

To my surprise the mains cabling seemed OK, giving readings outside my DMM's 200Mohm at various points.

A side issue arose has me puzzled. I found about 1V DC across a twisted pair cable, disconnected at both ends, buried in several places. How does that come about? Even the short section I've cut out, with gaffer tape around connector blocks, has about 0.5V DC.

Terry, East Grinstead, UK

What are you using to measure the potential difference? A high input impedance meter?

What's the reading if you have a, say, 10kohm resistor between the conductors?

You can display anything you want on a XYZ scope:

DC voltages below 1 V could indicate some electrochemical pair between dissimilar metals, such as the multimeter probes and the actual wire. If the conductors were supposed to be separated, a well conducting fluid could cause such phenomenon.

Did the multimeter show anything on the AC range ? Some multimeters show some DC reading even with a symmetric AC voltage. Asymmetric AC (even harmonics) could also cause a DC indication.

When I worked at a Navy DF/listening site we had "Time domain reflectometers" that could locate a buried cable malfunction to within inches. I wonder if something similar couldn't work on buried mains cables?

We had an antenna array with 150 elements whose cable length had to be accurate to an inch or less. The reflectometer was a scope where a sig-gen sends a step waveform down the line then measures and displays the amplitude of the reflected signal. Glitches in the waveform were displayed against a horizontal time base. One of those "A intensified by B" type textronics gizmo let you dial in the exact distance.

Don't trust high impedance meters on long wire runs. Some devices respond fast enough to "see" radio waves and background noise as voltages. A high value load resistor should nip it if that's what it is.

I built this differential thermometer and rigged a DC motor to display micro differences in temperature between two thermistors in a bridge circuit. Worked like a champ (would sense radiant heat from my hand a few inches from one sensor) The geared DC motor was just a meter movement to drive a long pointer.

The heatsink on the motor was always way hotter than it should have been. (especially with the bridge in balance) I tracked it down to radio waves that were picked up by the (short) wires to the 10K sensors. They were flying between the power op amps I was using for a motor driver and saw the .01uf cap across the motor brushes as a dead short. The motor wouldn't even jitter a little bit so I had no clue to the heat problem until I got a scope on it.

Measurement technique and equipment means a lot when you want to really know what is going on.

That's great but for an old scope on display it might be more appropriate to have a presentation that is more vector and less raster.

Of course you don't want to burn it in so have it come on when a motion sensor detects someone.

Jou just re-invented the battery, two different metals with a bit of groundwater.

Ask Geordi La Forge...

-- A host is a host from coast to snipped-for-privacy@nrk.com & no one will talk to a host that's close.......................... Unless the host (that isn't close).........................pob 1433 is busy, hung or dead....................................20915-1433