Strange Stray Voltage Problem

[OldReefer]

I am not a stranger to stray voltage problems in sumps. Normally it is pretty easy to troubleshoot with a voltmeter, but I have had an intermittent problem that has been making me crazy for months.

 

I have a big power strip connected to a GFCI outlet about 3 feet above my sump.

 

Into that outlet, I plug in two EB8s, and a couple of powerheads and my grounding probe.

 

I was getting getting 16 to 30 VAC in the sump, but when I upplugged the individual components from the EB8 it would not go away. But when I uplugged the EB8, it went down to 0.1V. Then when put my grounding probe back in the sump, the voltage went back up to 6 !?!

 

I plugged the EB8 and the gounding probe into an extension cord on another circuit. The stray voltage stayed at 0.1.

 

So apparently it has something to do with the powerstrip, but I am at a loss to understand how that could happen. I have another EB8 plugged into that strip but I do not get the same problem. I don't want to replace the strip until I can understand what it is I might be fixing.

 

Anybody got a clue?

[mr.x-ray]

Did you unplug both the #4 and 8 switch on the DC8 at the same time?

[OldReefer]

I unplugged the whole EB8 at the source. The 4 and 8 outlets run dosers. They were never on during the testing. I had them both unplugged anyway.

[Coral Hind]

What brand strip do you have?

 

Most ground probe packaging states not to plug them into power strips. They should go directly to the wall outlet. If it is a fancy strip it might have MOV's in it which can leak voltage. You might have a loose connection to ground somewhere which is not giving you an active ground. Water might have condensated in the strip and is causing the leakage. Try putting the ground probe directly into a wall and check again.

[Coral Hind]

I also wouldn't worry too much about such a small voltage leak. It could be induced voltage which doesn't have much amperage or pressure behind it, if it did the GFCI would have tripped.

[Origami]

Where is the EB8 mounted? Could the case be picking up a charge (radiated or conducted), delivering it to the ground wire in the strip, and carrying it into the sump via the ground probe? Coral Hind, have you seen power strips with isolated (or filtered) earth grounds?

[OldReefer]

It is the most basic powerstrip from home depot. It is just a long metal strip with outlets that you pick up next to the surface-mounted wiring parts (the sort of thing you plug in about a workbench). I am thinking something has gone south with it that is causing voltage to feedback through the ground circuit. The other EB8 and the other powerheads I have plugged into it do not seem to be a problem, but that may be because none of those devices have three-pronged plugs that enter the water.

 

Also, it is possible that something could be wrong with that circuit that could cause voltage to be sent into the ground circuit?

[Coral Hind]

Oldreefer, Did you ever get this issue fixed?

[linkfalcon]

Move the ground probe

[OldReefer]

I do not seem to have a significant stray voltage anymore (noticeable by touch). But my pH probe is now pretty useless. It makes a couple of big excursions every day. I am trying to correlate outlet changes to the wild swings, but so far no luck.

[paul b]

As was stated, that amount of induced voltage is normal and you should get that in almost any body of water that is near anything elsetric. Don't worry about it.

[Matt LeBaron]

This thread got me thinking so I grabbed my multimeter, checked and found about 35V in my tank, mostly it's 1-3 from a bunch of different stuff but my PH probe is bleeding in 10V, I can't imagine this is normal. Any thoughts on if that seems right? I think I'll be grabbing a grounding probe tomorrow.

[Coral Hind]

It is pretty normal for our tanks to have induced voltage. You are only seeing half of the picture reading the volts. The induced voltage has basically no amerage behind it which is why it doesn't hurt you. I do recommend getting the ground probe to bleed off that induced charge.

[paul b]

If you stuck a probe in you while you were standing near some transformers you would have that much voltage generated in yourself. It is normal. Anything near ac current, which is everything in your tank will have some voltage induced in it.

[OldReefer]

Whatever was zapping me I seem to have fixed. I just have the normal residual voltage now. Now I just wish I could fix my erratic pH probe. It is all over the place lately.

[Origami]

The probe or the wire doesn't run near a ballast or other high current or frequency device, does it? These are an induce electrical noise that contaminate the input of sensitive sense amplifiers iat the probe input circuit.

[OldReefer]

I am thinking he same thing Tom. I just re-routed the cable for the pH probe. I am still getting occasional spikes that shoot it up from 8.1 to about 8.5. I am going to dump the Apex log and see if I can find some sort of correlation.

Edited March 11, 2013 by OldReefer

[Coral Hind]

How old is the probe? Have you tried replacing that? Is the pH probe in a section of the sump where bubbles can collect under it as that throws off the readings? Is the probe near a heater or output from reactor? Is the probe too deep under the water? They seem to have issues when placed too deep.

[johnnybv]

Always been curious of the science behind the grounding probe. I have always used them, but my thought being if there is a current being introduced into the water wouldnt adding a gounding probe just complete the circuit and let the voltage flow? I ask this because I too have a stray voltage that zaps my team but doesnt zap me, I have even put my tounge in the tank to prove that there was no shock. What I did notice was that I had on thick sole boots where are the other subjects had on much less insulating shoes. There is a grounding probe in the tank that is properly connected to the grounded outlet. Just something to ponder.

[Scott_LM]

The grounding probe does provide the difference of potential that is needed for electricity to flow. So does placing your hand in the tank. Without a path to ground your tank is the proverbial "bird on a wire". Wearing thick soled boots or rubber gloves will insulate thus not providing the ground(increase in resistance). Voltage also flows through people differently, ie if you have a cut on your finger you will feel the voltage more than if you didn't(the cut actually reduces your skins natural resistance).

 

There are lots of arguments out there about the use of ground probes. I suggest using GFCI and ground probes, but that is just me

 

David,

Did I get that right? It has been a long time since I was a practicing electrician and the CDCs

[johnnybv]

Im talking more of the effect on the livestock, if there is no ground then the voltage can not shock anything including the livestock. (assuming they are not wearing ruber soled shoes)

john

[Origami]

The induced voltage in a tank typically, that comes from, say the rotation of pump parts is typically induced static charge with little ability to sustain dangerous currents. In this case, a grounding probe simply dissipates the charge bringing the voltage potential of the tank to the ground probe reference.

 

If there is a voltage that is leaked into the tank, such as from an open wire, then the ground probe can carry current between the two. In this case, the ground probe tries to hold the tank water at the near-ground potential (although there can be a gradient). If the wire is on a GFCI, then the GFCI should trip because current is being leaked outside the GFCI-protected circuit. That's why the two are recommended to be used together. In this situation, if the situation continues, you can actually upset the chemistry of the tank because electrolysis takes place.

 

If the voltage is left floating, then there shouldn't be an effect on the livestock. However, I can't say this for sure. Fish have different systems than we do including their lateral line systems which may be affected.

[Coral Hind]

David,

Did I get that right? It has been a long time since I was a practicing electrician and the CDCs

Yes, perfectly!

[Coral Hind]

The induced voltage in a tank typically, that comes from, say the rotation of pump parts is typically induced static charge with little ability to sustain dangerous currents. In this case, a grounding probe simply dissipates the charge bringing the voltage potential of the tank to the ground probe reference.

I agree.

 

If there is a voltage that is leaked into the tank, such as from an open wire, then the ground probe can carry current between the two. In this case, the ground probe tries to hold the tank water at the near-ground potential (although there can be a gradient). If the wire is on a GFCI, then the GFCI should trip because current is being leaked outside the GFCI-protected circuit. That's why the two are recommended to be used together. In this situation, if the situation continues, you can actually upset the chemistry of the tank because electrolysis takes place.

Just to elaborate so there is no confusion and people think you need a ground probe for each GFCI. The ground probe wire doesn't need to be on the GFCI circuit. The GFCI is not looking at the ground wire but comparing the hot leg and neutral leg only. When it sees that power going out is not near the same as what is coming back on the neutral then it trips. So the ground probe can be installed on a different circuit or even bonded to a pipe and the GFCI will still work. So only one ground probe is needed even if you have multiple GFCIs on a tank.

 

Even if power is elevated to something dangerous and a grounding probe is used the flowing of voltage through the system would still be fine for the fish because of equipotential. The fish are surrounded by water and the voltage will flow around them instead of through them because their body is slightly more resistant. The water acts as a faraday cage. When a heater breaks in the tank it is not usually the power that kills the tank but the electrolysis and rapid decay of the heating element.

 

This is my favorite video of equipotential:

[Origami]

Just to elaborate so there is no confusion and people think you need a ground probe for each GFCI. The ground probe wire doesn't need to be on the GFCI circuit. The GFCI is not looking at the ground wire but comparing the hot leg and neutral leg only. When it sees that power going out is not near the same as what is coming back on the neutral then it trips. So the ground probe can be installed on a different circuit or even bonded to a pipe and the GFCI will still work. So only one ground probe is needed even if you have multiple GFCIs on a tank.

Thanks for clarifying that, David. The idea of current being "in circuit" for me is restricted to the flow through the hot and neutral legs. Thus, current lost though other channels, whether it's the ground wire, grounding probe, or your body, results in an imbalance between the hot and neutral legs of the GFCI-protected circuit, triggering the device. It's by this means that a GFCI protects you. In a way, the grounding probe connection takes the place of your body, triggering the circuit so you don't have to.

 

Great video. Like a bird on a wire.

 

Did you ever see fluorescent tubes glow under the field induced by high tension wires? Kind of cool.

Skip to about 5:20 into the video (as it's unbearably long, otherwise) to see the effect in action.