GFCI Breakers and Inductive Loads

[websary]

Hi,

Any help with this question is hugely appreciated! I work for a commercial refrigeration company, and we have a customer who is saying that their codes require the use of a GFCI breaker on the circuit that will power our units. These unit will consist of a few 208V single phase compressors and a few low wattage 120V fan motors. The units are refrigerated display cases in a grocery store. They have some water inside of the case due to the melting of frost on the coil when the unit goes through a defrost, but it is separated from the electrical and I've never heard of this requirement. Two questions:

1. Does anyone know if NEC requires GFCI in this application? Does this seem like a reasonable code requirement? The location is Seattle, WA.

2. I am guessing that the having these compressors and motors on a GFCI breaker could cause nuisance trips, but I don't know that for sure. What are your thoughts?

Thanks!

Bill

 

[synchro]

Are the compressor motors powered through VFDs (variable frequency drives)? They are known to increase the likelihood of GFCI tripping.

 

[don_resqcapt19]

I can't think of what in the NEC would require GFCI protection for those units in a grocery store, unless they are in the deli kitchen. Also, most of the GFCI rules only apply to cord and plug connected equipment. The larger refrigerated cases are typically hard wired.

 

[websary]

Thanks for the replies guys. The compressors are not on VFDs just straight on / off through a contactor. The loads are not perfectly balanced because the 120 components (fans, lights, heaters) are distributed across the two legs for that bring 208 to the compressor. Any concerns about tripping in that situation?

You are right about the wiring. It's hard wired with no plug. Seems like an odd requirement for GFCI, right? I think I'll have to ask to see the specific code the customer is referring to.

Thanks!

 

[winnie]

If your customer has an internal requirement or 'code' that requires an installation which exceeds NEC requirements, don't try to argue with them; that is an upsell and you should bill accordingly.

However you should request from the customer the exact requirement which they need to meet. For example the requirement might be something along the lines of 'provide GFCI or hard wire', in which case the normal hard wiring is sufficient. Or the requirement might be for 'GFPE' which is less sensitive and thus less prone to nuisance tripping. Or the requirement might permit assured grounding rather than the use of GFCI. Do your best to discover what the customer is really asking for.

If GFCI is required, try to divide your loads so that normal leakage current doesn't cause tripping.

-Jon

 

[mbrooke]

From an physics standpoint GFCIs should not be on fridges. Capacitive reactance to ground will cause nuisance tripping in varying degrees.

 

[winnie]

From an physics standpoint GFCIs should not be on fridges. Capacitive reactance to ground will cause nuisance tripping in varying degrees.

I would modify that statement.

From a physics standpoint, capacitive reactance to ground results in inherent leakage from circuit conductors to ground. This leakage must be considered when applying GFCIs to fridges.

IMHO a GFCI is a very effective approach to preventing electric shock. Furthermore it is entirely possible to design systems to arbitrarily reduce capacitive reactance to ground, however the costs of doing so must be balanced against other equally effective approaches to preventing electric shock which may be more effective in a given application.

For example if you have a solid EGC then leakage which would certainly trip a GFCI presents no shock hazard. 50 mA leaking from a door seal heater to a solid ground will not create any sort of shock hazard (though arguably with modern insulation that much leakage probably indicates a damaged heater). So some sort of automatic mechanism that ensures the continuity of the EGC, combined with much less sensitive ground fault detection would provide the same safety as a GFCI.

For reducing capacitive reactance to ground, one simple technique is to have thicker insulation between energized conductors and grounded metal. A more complex technique is to place an electrostatic shield connected to the grounded conductor between circuit elements and grounded metal. A technique used in some appliances is to have components with switched bonding, where component housings are only grounded when access covers are removed (IMHO this is a sketchy approach, something that meets the letter but not the spirit of the rules).

Addressing websary: The requirements for GFCIs are expanding, and even if they are not actually required now, they may be in the future. As this is your business you probably want to be ahead of the curve, learning how to make your equipment work with GFCIs, but also trying to influence the rules requiring GFCIs.

-Jon

 

[ptonsparky]

Too often, we as ECs are stuck between the theory that AFCI and/or GFCI is not required or needed, manufacturers that indicate their equipment should not be used with GFCIs and the reality that laws may require them.

I agree, suggest the customer provide you with his references.
There is a big difference between GFCI protection for personnel, Ground Fault protection of equipment, or none at all.

 

[mbrooke]

I would modify that statement.

From a physics standpoint, capacitive reactance to ground results in inherent leakage from circuit conductors to ground. This leakage must be considered when applying GFCIs to fridges.

IMHO a GFCI is a very effective approach to preventing electric shock. Furthermore it is entirely possible to design systems to arbitrarily reduce capacitive reactance to ground, however the costs of doing so must be balanced against other equally effective approaches to preventing electric shock which may be more effective in a given application.

For example if you have a solid EGC then leakage which would certainly trip a GFCI presents no shock hazard. 50 mA leaking from a door seal heater to a solid ground will not create any sort of shock hazard (though arguably with modern insulation that much leakage probably indicates a damaged heater). So some sort of automatic mechanism that ensures the continuity of the EGC, combined with much less sensitive ground fault detection would provide the same safety as a GFCI.

For reducing capacitive reactance to ground, one simple technique is to have thicker insulation between energized conductors and grounded metal. A more complex technique is to place an electrostatic shield connected to the grounded conductor between circuit elements and grounded metal. A technique used in some appliances is to have components with switched bonding, where component housings are only grounded when access covers are removed (IMHO this is a sketchy approach, something that meets the letter but not the spirit of the rules).

Addressing websary: The requirements for GFCIs are expanding, and even if they are not actually required now, they may be in the future. As this is your business you probably want to be ahead of the curve, learning how to make your equipment work with GFCIs, but also trying to influence the rules requiring GFCIs.

-Jon

Agree.

You can reduce capacitance by not grounding internal parts (I am against this as IMHO a greater hazard is presented), thicker insulation, smaller MOVs, ect however there will always be a practical limit when the number of conductors and length are taken into account inside the appliance.

A ground assurance device would be ideal and I am all for this but as we know the NEC sets their bets on only one scheme.

 

[ramsy]

In my experience, the most common shock hazard with refrigeration equipment is the lighting. Whether internal lamps or external light switches, moisture builds up and ultimately becomes a shock hazard.

Repair solutions have required arranging raceways to drain, or removal of internal lighting not listed for the environment.