Electricity under water

Aug 16, 2007 18 Replies

I built a lake home for some clients and later installed an underground drop to a pedestal with a 240 volt disconnect (exactly like an AC disconnect). From there, the people who built the dock tied into the disconnect to run wire out to the dock.



Then the rains came. The lake rose almost 35 feet in a matter of weeks. More rain. The lake rose to flood stage and submerged the pedestal. Yet power remained to the dock. The pedestal was under about 15' of water, yet did not trip the breaker at the main service. The owner called me and asked if this was dangerous and I said yes, turn off the breaker at the breaker box, which he did.



My question is: why did the breaker not trip when the disconnect was submerged? It is not waterproof by any means.


It did not exceed the current rating, perhaps you might consider upgrading to gfci?

Sounds like the whole feed should have a 220 gfci breaker in the house. Only luck kept someone from getting jolted. Same as you would do for a hot tub.

On Thu, 16 Aug 2007 14:45:26 GMT, Robert Allison wrote Re Electricity under water:

The conductivity of the water was not sufficient to allow enough current flow to trip the breaker??

Probably the same reason that an open motor column sump pump of mine has several times been flooded totally underwater and still started and pumped out the water when power was restored. While fresh water will conduct electricity, it doesn't do it well and would prefer to travel through our bodies if they were in the water close to the source.

I total agree with you. You wonder the size of the existing breaker?

Robert Allison wrote in news:aqZwi.10204$Ns6.6028 @trnddc01:

Had a friend with a very old stone wall foundation, basement, dirt floor. Panel box in basement. Bad flooding. Basement totally filled. Power still on in house!

Check this out though. Calls power company. Fills them in. Wanna know what they told him? Are ya ready? That's right, "Go turn off the main breaker"!!!

I'm thinking the ground system wasn't a very good ground. But you would think with all the rain, that something should have shorted.

How about...break the seal and pull the meter!

Yes, grounding has nothing to do with this. Have you ever seen a simple vaporizer that you use when you have a chest cold? If you take it apart, it's nothing more than two electrodes, about 1/2" apart, connected straight to the 120V cord. That obviously doens't trip the breaker either, because the current flow is not anywhere near the limit. But it is enough flow to vaporize the water. And in the case of the lake, I'm sure some current was flowing as well. Exactly how much would depend on how much uninsulated metal was submerged, how conductive the water was, etc.

Water isn't a terribly good conductor. There was probably enough leakage to ground to trip a GFCI breaker, if you had installed one. But there wasn't the 15+ amps of leakage (either to ground or to neutral) required to trip a regular 15 A breaker.

Rain water is actually close to distilled water, which is a very poor conductor. Pure enough water can be treated as an insulator.

Dave

While true, this is not really relevant, as water is never found in the pure state in nature.

Not only that, but except for burning Hydrogen in the presence of Oxygen, all water is "used" water. What that means is that your bottle of "fresh" Norwegian Spring water was one cattle pee.

"HeyBub" wrote in news: snipped-for-privacy@news.supernews.com:

The cheap beer I drank some 20+ yrs ago was often referred to as skunk piss.

Seems we owe a lot to the animals of this planet when it comes to what we drink :-)

And some of that cattle pee was once a part of George Washington.

Not absolutely pure, but the degree of impurity determines its conductivity (more or less). The water molecules themselves don't conduct worth a darn, but other substances dissolved in it can increase conductance greatly. Well water usually has high dissolved solids, while rain water has almost none, and lake water tends to be somewhere in between. In this case, if the lake rose tens of feet in a short time due to rain, the water was likely less conductive than usual.

In addition, there was likely less than a square inch of "hot" conductor exposed to the water, and current flow across this interface will be proportional to area.

Dave

Less than usual, sure, but probably not very much less: the lake didn't rise tens of feet solely from rainwater falling directly into the lake. (That would have been one hell of a rainfall, no?) The rise had to be due almost entirely to water falling onto land in the lake's drainage basin, and hence the water entering the lake was mostly land runoff -- and couldn't possibly have been pure rainwater.

Exactly so.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required