GFCI weirdness

Originally Posted By: westom
Originally Posted By: bmwpowere36m3
Same happened to me.... switched the circuit over to a regular one and haven't had a problem since.

A classic and irresponsible example of curing symptoms. That GFCI was reporting a serious human safety issue. So we killed the messenger - removed a GFCI. Left the threat to human life in place. And then played mind games to justify that change.


Don't stick your finger in the sump pit with the power on...

I'm all for GFCI and have them as code requires, but for a sump pump I don't see the benefit and/or risk. I have a higher chance getting zapped on a light switch which is not GFCI protected.
 
That is just one example of many of how a GFCI can trip with no identifiable cause (you cannot, after all, see RFI, and if the RFI is a transient source, such as a truck passing by with a linear amp on their CB radio, you may never know that is what caused the trip).

Consider my sump pump. Here I have a GFCI that tripped every couple of months. Reset it and it's fine until it trips again a couple months later. Am I to believe that my sump pump somehow has an intermittent ground fault that only happens every couple of months? That is, the sump pump has a ground fault that then clears up for the next 2 months as soon as I press the reset button?

I can't even imagine how it would be possible to have such an intermittent ground fault, considering that the GFCI is located right next to the sump pump so that the only possible sources of a ground fault are the sump pump motor and it's cord. Inspection of the cord revealed no problems (as would be expected, being less than a year old). The only way I can see a sump pump motor developing a ground fault is if water got inside, which would result in a ground fault until the water dried out, which isn't going to happen when the pump is submerged.

Since I could not find a cause for the GFCI tripping, I did what I have done other times when I could not identify the cause of the GFCI trip: Replace the GFCI.

As with the other times I have done so, the replacement GFCI corrected the problem.

Conclusion: There was an internal problem with the GFCI that caused it to trip when it should not have tripped.
 
Originally Posted By: bmwpowere36m3
I have a higher chance getting zapped on a light switch which is not GFCI protected.

Classic rationalizing. If standing where skin is wet, then that light switch also must be on a GFCI. And the switch is also safety (equipment) grounded. That is required human safety - twice over.

Touch a sump pump case with wet hands and that pump must not threaten human life. That sump pump's GFCI is reporting a human safety defect. Instead, one invents "nuisance tripping" to justify "kill the messenger" - remove that GFCI. Irresponsible rationalizing that unnecessarily threatens human life. Because one does not know how to solve problems - only knows how to cure symptoms.
 
Originally Posted By: westom


Where does a defect more likely exist? In the transmitter.


Nope. The transmitter is doing exactly what it's supposed to: Put RF energy into an antenna.

That the RF energy causes a GFCI to trip means that the problem is in the GFCI.

(Note that the transmitter is NOT powered through the GFCI, and in this case, was running off batteries).
 
Originally Posted By: researcher
now let me ask this.. would the "problems" that GFCI have also be applied to AFCI outlets?


Yes, in that a poorly-designed AFCI can trip when it should not, just as a poorly-designed GFCI will.

Yes, in that an AFCI contains electronic circuitry that can degrade/fail over time, just as a GFCI does.
 
Originally Posted By: brianl703

Yes, in that a poorly-designed AFCI can trip when it should not, just as a poorly-designed GFCI will.

How does an RF transmitter cause a trip? Correlation never creates a fact. He assumed the transmitter will always cause a problem. IOW he saw a symptom. Then used speculation to conclude it is acceptable.

GFCIs are routinely found in transmitter stations where radiated power is massive. GFCIs do not trip. Yes, it is possible that electronics inside one GFCI were defective - a manufacturing defect. Everyone has seen them. Even counterfeit electrolyte in capacitors caused electronics to fail many years later. Another manufacturing defect that we all know about. That does not say GFCIs are defective.

Electronics should work for many decades without failure. Manufacturing defects can exist. But GFCIs rarely fail. That GFCI apparently was reporting a defective sump pump. So he removed the GFCI - killed the messenger - to deny a human safety threat exists.

Always first identify and then fix a defect. If a transmitter might cause a GFCI to trip, then the reason why is learned. Never once saw a portable transmitter adjacent to a GFCI trip that GFCI (or AFCI). Never cure symptoms. GFCI is not defective - a pump apparently is.
 
As I stated previously: GFCIs are required for sump pump outlets not because a sump pump ground fault might exist, but because someone may use that outlet as a general-use receptacle.

A hard-wired sump pump is not required to be on a GFCI. Only a cord-and-plug connected sump pump.

His pump maybe defective. Mine definitely isn't.

Electronics do not work for many decades without failure UNLESS designed that way---and that increases cost. Any use of electrolytic capacitors, for example, dooms the circuit to an eventual failure when, not if, but when the rubber seal on that capacitor fails.

On GFCI failure rate, if I had to choose between "never, rarely, sometimes, often, always" I would say that they "sometimes" fail.

In my experience, as the age of the GFCI approaches 10 years, the failure rate is about 100%. This is for GFCIs of a particular brand produced in the mid 90s. Newer production GFCIs may be more reliable--GFCIs produced in the mid-00s of the same brand have not needed replacement yet.
 
Originally Posted By: westom
GFCIs are routinely found in transmitter stations where radiated power is massive. GFCIs do not trip.


I'd have to think transmitter stations have more power line filtering and probably ran the wiring inside of conduit. Thus lowering picked up RF. All wiring inside of a metal box with good shielding and ground practices vs typical house with none.
 
Originally Posted By: brianl703
An interesting read on what conditions can cause nuisance tripping.

Unfortunately it does not say anything about why. It is a subjective warning about what may and may not work. And that, unfortunately, is vague, subjective reasoning that does not say why with numbers.

Those GFCI testers can only report some defects; can never say anything is good.

A sump pump can be leaking 4 milliamps constantly - and not trip a GFCI. One connects another appliance which causes a GFCI trip. Subjective reasoning (also called junk science) then blames the other appliance.

Meanwhile I built a tester than intentionally created an adjustable leakage current. So that I can measure how much leakage causes a GFCI to trip. Good GFCI trips at between 4 to 6 milliamps. If it trips lower, then we know some other appliance has a human safety threat. So it may cause 'nuisance tripping'.

All honest discussions define everything quantitatively. That means numbers are always included. No numbers is a first indication that it might be a scam.

Second, a transmitter should not create a 5 milliamp current on AC wires. GFCIs work just fine in the massive E-M fields even in transmitter facilities. However it can induce micro-amp currents. That combined with leakage elsewhere could result in 'nuisance tripping'. Solution- find and fix that unacceptable other leakage. Never foolishly make a conclusion from myths that blame 'nuisance tripping'.

One always learns what is defective long before making accusations or a conclusion. Classic junk science is to see a GFCI trip. Then wildly speculate it was caused by some other (perfectly safe) appliance or a radio transmitter.

That sump pump apparently has a human safety defect. So someone with contempt for human life used a classic urban myth (nuisance tripping) to justify removing a GFCI. Used a conclusion only from wild speculation to endanger human life.

Long before making such conclusions, the informed first learn the whys with numbers. One who cannot must find someone to teach or somehow to learn basic electrical concepts before making a conclusion. 'Nuisance tripping' is an 'easy to promote' urban myth.
 
Originally Posted By: westom


A sump pump can be leaking 4 milliamps constantly - and not trip a GFCI. One connects another appliance which causes a GFCI trip. Subjective reasoning (also called junk science) then blames the other appliance.


Any GFCI installed for a sump pump is to be on a dedicated circuit for use of the sump pump ONLY. This is code. Plugging another appliance into GFCI outlet dedicated for a sump pump is misusing the outlet, and is not permitted in any facility under my control.
 
Originally Posted By: brianl703
Any GFCI installed for a sump pump is to be on a dedicated circuit for use of the sump pump ONLY.

So you completely missed the points. Again:

A pump leaking four milliamps can create 'nuisance tripping'. It may trip that GFCI every day. Or it may trip it once after many weeks or once every six or twelve months. That 'nuisance trip' is reporting a defect.

Even the microamp current from a nearby transmitter could have that 'nuisance trip' ... that is actually reporting a potential human safety threat.

Second, conclusions only from observation are classic junk science. If the GFCI it tripping, then one finds a defect long before making any conclusion. 'Nuisance tripping' (as demonstrated by some contributors here) is typical of conclusions only from observation.

And third, I have never seen the National Electrical code require a sump pump on a dedicated circuit. Apparently that is either a local code or a workmanship standard. A good habit. But not required.
 
Last edited:
An AFCI detects high frequency currents to the load. There are some loads which do that "normally" and can cause unnecessary tripping of the AFCI. This is different from a GFCI. Ground faults are never normal.
 
Last edited:
Originally Posted By: mk378
There are some loads which do that "normally" and can cause unnecessary tripping of the AFCI.

A responsible post because it also says why. Some defices can create that high frequency noise. And then we found the defect that made that possible. For example, electronics that drove a paper shredder would trip that breaker.

We opened it. Clearly marked on its PC board were holes for the required line filter. We see this often. An engineer designed what is necessary. Then a business school graduate uses his cost control mentality to cut costs. Apparently that business school graduate discovered a shredder worked just fine without that line filter. So the filter was replaced with jumper wires. To increase profits.

I obtained and installed an exact same filter from an old TV. It even matched all holes. Then a breaker stopped tripping.

Don't blame the AFCI breaker. Blame the defective appliance that creates noise - that even FCC regulations say must not exist. What is the most common reason for such failures? Bosses who have no idea how the work gets done.
 
Thought I'd update this. I did some research on GFCI trip times. I found that GFCIs aren't required to trip instantaneously. For example, GFCI with a 6mA ground fault is allowed to take as long as 5 seconds to trip. The allowable trip time is shorter as the fault current goes up; for example, at a 20mA fault the GFCI may have to trip in 1 second. (The exact numbers are in a chart of fault current vs. allowed trip time in the UL 943 standard).

A GFCI that always trips instantaneously is compliant with the UL requirements, is cheaper to make (the circuitry is less complex), but is more likely to nuisance trip.

More expensive GFCIs that implement variable trip times are less likely to nuisance trip.

And that is why some brands of GFCIs are more likely to nuisance trip than others.

The more you know.
 
Back
Top