Samsung Refrigerator Tripping GFCI

Status
Not open for further replies.
So I have a refrigerator, 3 feet away from the edge of the sink in a new kitchen that I just installed (electric only obviously) and the refrigerator circuit is a 20 amp AFCI breaker going to a GFCI receptacle. The customer bought a Samsung four-door 28 ft.³ refrigerator and about five minutes after plugging it in the GFCI tripped. I’ll spare the troubleshooting details if you’ll take my word that I’m confident enough in the field, and I have isolated the problem at the refrigerator itself. Samsung‘s guidance is use a non-GFCI receptacle. Obviously that creates a problem with the NEC and liability on my end, but I also find it negligent on their end to have a product leaking somewhere and their response is to find a way to hook it up in a way that ignores it. Am I thinking right here? I don’t think they will provide me with a way to stop the leak, so has anyone dealt with this? The breaker is not tripping at all, so I’m thinking most likely somewhere along the line. The neutral is coming in contact with the chassis of the refrigerator or something to that effect.
 
You could recommend a plug and cord isolation transformer to the customer in this situation. That way, the receptacle will still be NEC compliant.
 
The receptacle is in the adjacent cabinet, but the refrigerator is immediately next to the sink so if there is a loss of neutral current through the ground or something like that, I think it would be very unwise to ignore it.
 
It appears that Samsung is using a VFD of sorts in their refrigerators to increase the energy star levels. VFD will almost always trip a GFCI. Look at issues created with the requirement for GFCI on AC units, that created a moratorium on implementation of the code requirement of certain AC units.
Sounds as if it is the VFD is kicking in after about 5 minutes of running.
 
It appears that Samsung is using a VFD of sorts in their refrigerators to increase the energy star levels. VFD will almost always trip a GFCI. Look at issues created with the requirement for GFCI on AC units, that created a moratorium on implementation of the code requirement of certain AC units.
Sounds as if it is the VFD is kicking in after about 5 minutes of running.
Interesting. If I could find some literature on this and can confirm the refrigerator is not actually leaking current or experiencing some kind of short then maybe I could figure out a way to remove the GFCI but it would really be much better if these things just worked with the devices our code makes us install. I appreciate the info I’m going to look into this VFD
 
I believe you were a cup short when you wrote this, could you translate? :)
I'll take a stab at it:
IIRC, someone here suggested getting the unit started and to temp sans GFCI, then using using the GFCI.
Get the refrigerator running and cooled to normal temperature on a non-GFCI-protected circuit first.
 
I believe probing both line ad return at the 4/5 minute mark with a regular recept.
There are handy 20a devices that seperate line and neutral- you just plug in and probe...but having the device hang out the wall works too..
 
My understanding is that there are several issues in play here.

1) Manufacturers not getting on board and designing their hardware from the get-go to not trip a GFCI when properly functioning.

2) GFCI requirements not really taking into account the safety of a system which has an intact safety ground connection.

3) Real leakage in new appliances associated with dampness in insulation, associated with transportation and storage. This dampness goes away once the appliances have had a change to run and heat up.

4) High frequency leakage associated with VFD and Inverter technology in appliances. This is real leakage current, but it has a different safety impact and should have different detection thresholds than 60Hz leakage.

The idea of temporarily plugging the unit in to a non-GFCI outlet, and letting it come to temperature, let the defrost and anti-condensation heaters run for a while, etc. specifically relates to item 3, and is a very good idea. This is apparently a very real issue with GFCI protection for electric ovens.

-Jon
 
Interesting. If I could find some literature on this and can confirm the refrigerator is not actually leaking current or experiencing some kind of short then maybe I could figure out a way to remove the GFCI but it would really be much better if these things just worked with the devices our code makes us install. I appreciate the info I’m going to look into this VFD
 
You could recommend a plug and cord isolation transformer to the customer in this situation. That way, the receptacle will still be NEC compliant.
Isolation transformer would not work.
The NEC requires power systems to be grounded if it will keep the voltage L-G voltage below 150V.
The NEC requires 120V receptacles to be GFCI protected when installe "within 6 ft" of a kitchen sink.
 
It appears that Samsung is using a VFD of sorts in their refrigerators to increase the energy star levels. VFD will almost always trip a GFCI. Look at issues created with the requirement for GFCI on AC units, that created a moratorium on implementation of the code requirement of certain AC units.
Sounds as if it is the VFD is kicking in after about 5 minutes of running.

It's also possible that the VFD is running the compressor at a higher speed initially to cool it down, and then it lowers the speed to maintain the temperature. The lower speed might then shift some of the common mode noise to lower frequencies which can pass through the low pass filtering of the GFCI detection circuitry and cause it to trip. If this was the case, initially cooling down the unit using a non-GFCI receptacle would not prevent subsequent tripping when powered through a GFCI. But even if it still tripped, such a test could provide more insight about what's going on.
 
Isolation transformer would not work.
The NEC requires power systems to be grounded if it will keep the voltage L-G voltage below 150V.
The NEC requires 120V receptacles to be GFCI protected when installe "within 6 ft" of a kitchen sink.
Using an approved external isolation transformer like this one, for example (or perhaps a more economic model): https://www.amazon.com/Tripp-Lite-IS1800HG-Isolation-Transformer/dp/B00008N6S7

The transformer could still be grounded and it is not installed as part of the building. Perhaps it could be placed 6' away from the sink to skirt that, if it applies to the appliance. Please let me know your take on this if I am mistaken - from the perspective of liability to the electrician, he has not made any installation against code this way, as his job stopped at the receptacle. It is up to the end user to operate the transformer, like any other appliance.
 
You’ve stated that the refrigerator is 3’ from the sink. But I don’t see where you mentioned where the receptacle is located in relation to the sink. Is GFCI protection even required?
 
You’ve stated that the refrigerator is 3’ from the sink. But I don’t see where you mentioned where the receptacle is located in relation to the sink. Is GFCI protection even required?
I wondered too. A good EG is all that may be required.

That from a guy, me, who had a GFCI ahead of a chest freezer because it's next to a utility room sink where I used to wash the legs and feet of grandchildren while they sat on it.
 
The receptacle is within 6 feet of the sink but would it even matter if the refrigerator is within 3 feet of the sink and not GFCI protected? I would think that a big metal item so close to the sink should be GFCI protected even if the receptacle is in the front yard.
 
Status
Not open for further replies.
Top