Ground fault

There are really only three reasons this breaker will trip.

1. You are overloading the breaker.
2. You have uneven hot/neutral current flow. (ground fault condition.)
3. Its defective or miswired.

Tom

I assume you're in the US.

Overload is unlikely if the only load is a 1 hp motor. The breaker is at least 15 A?

Defective is always a possibility with GFCIs, but the circumstances under which it is tripping don't seem to match the time constants likely to be found in the device.

The regularity of the tripping and the fact that it trips after being under load for 3 hours strongly suggests heat as a factor.

Heat inside the GFCI is possible if the connecting wires don't make good electrical contact. I would not expect GFCIs to be very sensitive to heat.

Heat inside the motor could easily cause expansion of a conductor with consequent shorting to the motor's case. Measure resistance from hot wire to case with the motor cold and it may be hundreds of megohms. When hot (immediately after removing ALL power after a trip) it may drop to 25,000 ohms or less. If so, it will cause at least 5 mA to be returned through the ground wire, thus causing the GFCI to trip. That's what I would look at first. Let it run till it trips and then remove all power and measure leakage resistance.

Good luck.

Chuck

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Ok thanks Lets say it was an over load if that was the case would the pump run for about 3 hours before it triped If it was a ground fault condition would I not see some kind of ground on the hot or neutral side ( I see nothing) If it was misswired would it not trip right away Thanks

The overload could take place when the motor is hot. It would be hard to measure the ground fault since trip will take place with very little flow. The miswire probably would trip right away unless you did something weird.

When you say it runs for 3-4 hours, is that after you have been splashing around in the pool?

Tom

Did you use an insulation tester? a cheap VOM won't find all faults.

unlikely.

cable insulation not waterproof?

Bye. Jasen

News==----

Thanks guys for the input I will check for grounds today when it trips and let you know what I find

Kill power and look for crud/dirt in all involved junction boxes.

Ed

Ok I waited around for a couple of hours until the pump tripped the breaker (15 amp ground fault) . I checked both the hot and neutral side of the plug for a ground and nothing (my meter goes to 3 million ohms). There are no junctions between the breaker and the pump. This is a new pump only 3 weeks old (1 hp). The one from last year sized up (bearings got wet) I could not get it going this season and even the old pump did it last year about half way through the season. Any other suggestions guys Thanks

Yes.

How much current does it draw ?

Not watts, not hp but AMPS.

Graham

Additionally, the startup surge can be enough to trip a GFI. Another typical cause is if there is a 3-way on the same circuit.

What does your meter indicate when you probe ground to neutral?

It' cant be the startup as I said in the original post the pump runs between 2 and 3 hours before i trips. This is the only thing that is on the GFI not even a switch. When I check ground to neutral its over 3 million ohms Rick

million

Read the AC volts between the ground and neutral. It should be very nearly zero however, I have measured 70 volts here. A new ground solved the problem.

The 3-way can be ahead of the GFI rather than behind it and cause trouble.

Something is causing an imbalance in the current between the hot and the neutral. That is what trips the GFI. Assuming the GFI isn't defective, there must be a path to ground somewhere.

Can you PLEASE check the current draw when it's contuously running ?

Graham

Think of the pool as one plate of a capacitor to ground - may be no DC path to ground but that doesn't mean there isn't some leakage from the pump to the water - through the water to the pool - across the pool membrane to ground. AC only circuit.

Check the resistance between hot and neutral to the water.

Another thought on the idea of current flow through the water being pumped - there may be air voids in the piping that occur when the pump is off preventing an honest ohm meter reading.

Remember that three things can trip a GFCI:

leakage currents that bypass the neutral defective GFCI excessive current drawn

Too bad the resistance test was negative. Doesn't necessarily mean there was no leakage. It gets more complicated from this point on. I still suspect heat because of the time constant. Could be a plugged filter that reduces water flow thru the pump and causes the motor to overheat.

You can measure the AC current drawn by the motor, by please do be careful since your test leads will be hot.

You can also measure the AC leakage current in the ground wire. But be extra careful if you do that, especially if leakage is a real possiblity!

It might be easiest to start by either replacing the GFCI with a known good one, or routing your pump circuit to a different one.

Use extra caution in all of this.

BTW, did you say if you reset the GFCI it will run for another 2-3 hours?

Chuck

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Let's clarify, shall we?

Overcurrent will trip a GFCI _circuit_breaker_ -- but that's because it's a circuit breaker, not because it's a GFCI.

If it runs for say another 15 mins and trips again, then the fault is likely to be a small overcurrent on the circuit.

Graham

Well...... if the bearing getting wet caused the bearing to seize, it's reasonable to assume that before they seized they became 'tight'.

If they did that, the motor would have to work harder (drawing more current) and that could cause an overcurrent trip of the breaker, i.e. nothing to do with a ground fault at all. I see that 1 hp pumps can draw ~ 11A so a modest 'overload' could indeed do it.

Graham