[swimmer]

I was working on a fish pond where something was tripping the GFI. This GFI was protecting the underwater lights, a fountain pump and some landscape lights around the pond. I found the problem to be the fountain pump.

My first question was: Is this a 10uA problem or a 20A problem? I used my ohm meter and got a very low reading from the 3 pond lights in parallel.
I wanted to simply bypass the GFI to see if the 20A breaker would trip but feared I'd electrocute thousands of dollars worth of fish if there was a flooded light.

It seems that a 60Hz ohm meter (if there is such a thing) would quickly tell you whether you are looking at a light filament, motor winding, or a real short.

Anybody know of a device or technique to deal with these questions?

 

[HARRY304E]

I was working on a fish pond where something was tripping the GFI. This GFI was protecting the underwater lights, a fountain pump and some landscape lights around the pond. I found the problem to be the fountain pump.

My first question was: Is this a 10uA problem or a 20A problem? I used my ohm meter and got a very low reading from the 3 pond lights in parallel.
I wanted to simply bypass the GFI to see if the 20A breaker would trip but feared I'd electrocute thousands of dollars worth of fish if there was a flooded light.

It seems that a 60Hz ohm meter (if there is such a thing) would quickly tell you whether you are looking at a light filament, motor winding, or a real short.

Anybody know of a device or technique to deal with these questions?

Are you sure that those lights are line voltage?

Normally those are low voltage look for a transformer that may be feeding those lights.

 

[swimmer]

Are you sure that those lights are line voltage?

Normally those are low voltage look for a transformer that may be feeding those lights.

Good point. It seems it was a 3-light assembly so there may be a junction box or transformer hidden in the landscape somewhere. Line voltage may be isolated from the water by a transformer and I was ohming the primary of the transformer. But it's still scary to bypass the GFI with living things in the water.

 

[HARRY304E]

Good point. It seems it was a 3-light assembly so there may be a junction box or transformer hidden in the landscape somewhere. Line voltage may be isolated from the water by a transformer and I was ohming the primary of the transformer. But it's still scary to bypass the GFI with living things in the water.

Yup Do not bypass the GFCI.

Also you can install a new GFCI that could be the problem as well.

Disconnect each light one at a time it could be that just one is tripping the GFCI ,.

Also if they are low voltage it could be just a bad transformer.

 

[CraigV]

If it's low voltage, will a fault to ground on the secondary side trip a GFCI on the primary? I'm thinking "no", but I'm curious to see reasons both ways.

 

[HARRY304E]

If it's low voltage, will a fault to ground on the secondary side trip a GFCI on the primary? I'm thinking "no", but I'm curious to see reasons both ways.

I would say no as well if they are low voltage then you are right the primary side of the transformer would be tripping the GFCI and there may be no problem with the lights them selves...

 

[dronai]

Most likely the pump. I had the same thing before.

 

[dronai]

Most likely the pump. I had the same thing before.

Didn't see this part

I was working on a fish pond where something was tripping the GFI. This GFI was protecting the underwater lights, a fountain pump and some landscape lights around the pond. I found the problem to be the fountain pump.

My first question was: Is this a 10uA problem or a 20A problem? I used my ohm meter and got a very low reading from the 3 pond lights in parallel.
I wanted to simply bypass the GFI to see if the 20A breaker would trip but feared I'd electrocute thousands of dollars worth of fish if there was a flooded light.

It seems that a 60Hz ohm meter (if there is such a thing) would quickly tell you whether you are looking at a light filament, motor winding, or a real short.

Anybody know of a device or technique to deal with these questions?

 

[swimmer]

My real concern is this:
How do I tell the difference between a short and a light filament or motor winding?

I know one will be 0 ohms and the other will be 3.5 ohms but I don't want to bet the fish's lives on what could be meter calibration issues or wire resistance.

 

[dronai]

I would not do this with a meter. Disconnect the lighting, and see if you get the GFi to reset. I've had this problem many times, and for me, it's always been the pump.

Harry's right, (my experience) the lights are always low voltage, so locate the transformer, and disconnect there.

If the lights are shorted somewhere it will take out the transformer, and the thermal protection will be tripped.

 

[McClary’s Electrical]

My first question was: Is this a 10uA problem or a 20A problem?

A gfi will not trip on overcurrent.

 

[McClary’s Electrical]

I wanted to simply bypass the GFI to see if the 20A breaker would trip but feared I'd electrocute thousands of dollars worth of fish if there was a flooded light.

?

The fish would have been fine. Ever heard of a farraday cage?

 

[Dennis Alwon]

Why do you want to see if the 20 amp breaker is going to trip? There is some leakage in the pump that is causing the gfci to trip. If you determined it is the pump then get a new one and be done with it.

 

[Zog]

Still trying to wrap my brain around 60hZ ohmeter

 

[HARRY304E]

Still trying to wrap my brain around 60hZ ohmeter

I think that could be just bad Electrical jargon that he learned from someone else.

 

[dronai]

Still trying to wrap my brain around 60hZ ohmeter

:laughing:

 

[CraigV]

The OP's first post confused me as well. I don't think the OP is referring to a "20a problem" as an overload; I think he means, is this a huge fault to ground, or a tiny one? The huge fault would very likely kill fish (and people), while the small one may or may not be deadly. And I think he's only posing this question for future issues, as he did state that the pump was the cause of his last problem, and got him thinking. Please correct me if I'm wrong here.

I also think dronai's advice is the way to go for troubleshooting; remove all possibilities but one, and test. Keeping adding loads until you get a trip, and that's most likely the problem. There could be instances where two or more extremely small faults exist in different loads, and only having both loads appplied caused the GFCI to trip, but that's not typical IME.

 

[swimmer]

Why do you want to see if the 20 amp breaker is going to trip? There is some leakage in the pump that is causing the gfci to trip. If you determined it is the pump then get a new one and be done with it.

I'm done with it but I'd sure like to know whether I'm looking at shorts, coils or filaments. It would have saved me some time on this job as well as others.

Locating and disconnecting loads is not always a trivial matter.

Anyway, I guess the answer is no. There is not a meter available at a reasonable price that will tell me that a 100W incandescent is 144 ohms at 60Hz.

 

[Zog]

Resistance does not vary with frequency. Now if you want to accuratly measure impedance then you would want a Z meter. But I don't see the application for this senario. http://www.teknetelectronics.com/DataSheet/SENCORE/SENCOLC103.pdf

 

[Mshea]

Resistance meters only work on dead circuits as they supply the voltage and current to do the resistance check. Generally they supply DC and I have never seen an AC resistance meter.
A broken filament cannot contact grounded metal unless the glass is also broken so as long as the bulb is in tact the ground fault is external.
You want your resistance to ground to be around 20,000 ohms to limit current to below 6ma if I put the decimal in the correct place.
A megohm meter is probably the best tool for finding ground faults as the voltage output is high enough to overcome high resistance or intermittent ground faults. The typical 9 volt battery in a digital meter may not overcome some higer resistance faults.
To test the lamps may be a real challenge if a transformer is used as you need to check the output and input circuits separately and primary to secondary windings checks too.
If you changed the pump and turned the circuit back on and the breaker holds then the leakage is below the 6ma, assuming the breaker is good. a 20 k resistor from either feed wire to ground should trip the gfci protection circuit.