GFCI Receptacles and GFCI Circuit Breakers - Nuisance Tripping Concerns 1

[jmbelectrical]

Another engineer just performed a punch list inspection with the electrical contractor for a project that we recently completed. The contractor informed him that the GFCI receptacles serving the electric water coolers, which are required to be GFCI-protected as of the 2008 NEC, were frequently tripping. The issue was allegedly resolved by installing standard receptacles in conjunction with GFCI circuit breakers. There are several things that I don't understand.

1. How common is GFCI nuisance-tripping on small refrigeration equipment and other fractional horsepower motor loads? In addition to electric water coolers, the NEC now requires GFCI protection for vending machines and many other appliances. Furthermore, all 120-volt receptacles in commercial kitchen facilities, which often contain refrigerators, ice makers, etc. - require GFCI protection. It was my understanding that older GFCI and AFCI devices were subject to nuisance tripping on these types of loads, but I thought these issues were resolved by manufacturers long ago.

2. Why would a GFCI breaker necessarily be less prone to nuisance tripping? When considering leakage current via shunt capacitance, one would think that the reverse is true. I believe SquareD recommends a maximum one-way circuit length of 200 feet for their GFCI breakers. In a scenario in which I had an electric water cooler or similar appliance located more than 200 feet from its source panelboard, what would be the best strategy to provide reliable GFCI protection?

3. Are some GFCI receptacles less prone to nuisance-tripping than others? Are there certain manufacturers that anyone can recommend (or not recommend)?

4. How likely is it that the electric water coolers are simply defective? I believe they're all brand new, in excellent condition, and installed properly.

Thanks in advance for any input provided!

 

[MikeHalloran]

I'd consider the possibility that you have a water leak or a condensation source somewhere, and that a GFI protected circuit is actually getting wet.




Mike Halloran
Pembroke Pines, FL, USA

 

[waross]

For #2. mount the breaker in a sub panel close to the load.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[PHovnanian]

jmbelectrical, I've heard some stories about nuisance trips with motor loads. So its not just you.

A couple of theories: GFCI current sensors in receptacles, being smaller than breakers, may have magnetics that saturate more easily from motor start currents. Or, motor start capacitors may have some capacitance to ground that causes trips. Although this case would seem to cause nuisance trips for either receptacles or breakers equally.

 

[Skogsgurra]

I have experienced quite a few installations over here in Europe where RCDs (Residual Current Devices, our translation of GFCI) trip more or less continuously. This installation

http://www.youtube.com/watch?v=Cq3hlU-oU9E

has lots of immersed pumps and long motor cables protected by RCDs. When they were running it with ordinary frequency inverters, there was one guy needed to reset the RCD to be able to run the show. They were tripping all the time. Then, frequency inverters with pure sine-wave output were installed instead and the RCD-resetting guy can now do other things.

The problem is not leakage or water in the devices. It is about increased capacitive currents due to high-frequency contents in the motor voltage. That current comes eventually from the grid (rectified in the inverter's grid rectifier) and therefore cannot be distinguished from "true" ground faults.

The same thing happens on a smaller scale if there are frequency inverters connected to the PCC. HF pollution is spread from the PCC to all outlets and there, HF currents are flowing through cable and motor capacitances to ground. This indirect "ground current" is orders of magnitudes lesser than the current measured on inverter supply lines. But still problematic enough to cause nuisance tripping.

Improved filters between inverters and grid often helps. Sine-wave inverters always helps.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[dpc]

In my experience, use of GFCI circuit breakers generally results in MORE nuisance tripping, not less. We avoid these. GFCI receptacles are a mass produced commodity item and can be prone to problems. The first thing I would try is to simply replace the ones that seem prone to tripping. I would stick with recognized brands and purchase "specification grade" receptacles (not that this means much anymore).

Other than that, I suspect Gunnar is correct (as usual) in his assessment.

 

[jmbelectrical]

waross,

It's pretty rare that we design a 120-volt receptacle circuit with a one-way length exceeding 200 feet, but it does happen on occasion when we are faced with unusually stringent requirements (from an architect, interior designer, or some other building design professional) regarding where we place subpanels.

 

[jmbelectrical]

dpc,

I'm glad you mentioned "specification grade" receptacles, because I've been wondering about that for quite some time! What does it mean, exactly? I do know what hospital grade receptacles are, but, in terms of build quality and features, I'm not sure about specification grade.

 

[dpc]

I'm sure others will add comments on this, but "specification grade" generally refers to a higher quality component typically used in commercial or industrial installations. This is opposed to residential grade or the dreaded "contractor special" grade. There are federal GSA specs for receptacles, switches, etc that are generally referenced.

For receptacles, Hubbell has been the accepted quality standard that everyone else tries to compete against, even going so far as to copy the Hubbell model numbers.

Hospital grade is different. These have specific requirements for insertion and removal force. I would never specify these except in patient care areas. People hate them.