What does a GFI actually protect in Solar DC Circuits?

[Paul_NJ]

I understand that DC GFI's are required by the NEC in the panel-to-controller circuit when the panels are roof mounted. I've read that DC GFI's, unlike their AC counterparts, do not have the function of personnel protection in terms of tripping under mA leakage, but rather trip if there is 1/2 amp flow to ground due to a high resistance fault that might otherwise cause high current flow and lead to a fire or further damage. I've also read that there are many conflicting opinions whether these devices actually are needed; and that folks often include one because the NEC says so but not because they think it does anything, or understand what it does.

Can someone shed light upon what the function of DC GFI's really are in a solar application. Seems to me that breaking the circuit at the charge controller doesn't provide much protection to the panels and wiring up on the roof if there is a short in between, as they are going to pump current right into the fault regardless. We're installing a solar set-up in Africa, strictly 12 volt DC with a Morningstar MPPT controller, so NEC requirements aren't an issue. My partner doesn't believe a GFI really does anything so it isn't necessary. I'm not convinced either way: if a GFI is providing real protection I'm all for it. But I'd like to understand what I've got to defend.

Can someone shed some light on this?

[Sunking]

Hey Paul you are not quite fully understanding the electrical code. 690.5 states:Roof mounted DC PV arrays SHALL BE provided with DC Ground -Fault protection to reduce fire hazards. You are assuming it is required to use a Ground Fault Circuit Interrupter, but it is not required as GFCI is only 1 of many over current protection and ground fault detection devices.

What it is telling you is you must have ground-fault detection with a interruption device of some sort. I have no idea of where you came up with DC GFCI needs at least a 1/2 amp of imbalance to operate.

[Mike90250]

some high mucky-mucky has a factory that makes DC GFCI ?

There are indeed two schools of thought on this, and I've not studied it well myself.

[jerbucc]

GFI in solar

I understand that DC GFI's are required by the NEC in the panel-to-controller circuit when the panels are roof mounted. I've read that DC GFI's, unlike their AC counterparts, do not have the function of personnel protection in terms of tripping under mA leakage, but rather trip if there is 1/2 amp flow to ground due to a high resistance fault that might otherwise cause high current flow and lead to a fire or further damage. I've also read that there are many conflicting opinions whether these devices actually are needed; and that folks often include one because the NEC says so but not because they think it does anything, or understand what it does.

Can someone shed light upon what the function of DC GFI's really are in a solar application. Seems to me that breaking the circuit at the charge controller doesn't provide much protection to the panels and wiring up on the roof if there is a short in between, as they are going to pump current right into the fault regardless. We're installing a solar set-up in Africa, strictly 12 volt DC with a Morningstar MPPT controller, so NEC requirements aren't an issue. My partner doesn't believe a GFI really does anything so it isn't necessary. I'm not convinced either way: if a GFI is providing real protection I'm all for it. But I'd like to understand what I've got to defend.

Can someone shed some light on this?

This is a good question, that no one seemed to answer. You are correct that higher power solar inverters typically trip abone 5 amps(not 1/2). All they do is blow the fuse disconnecting the dc usually both plus and minus from the inverter. It doesnt realy protect much other then alerting that there is over 5 amps flowing along the ground. This would occur if there is a short between a plus DC line and the EGC or if the inverter was somhow backfeeding current. It doesn't prevent the panels of course from still generating. Theres a lot of talk about this lately. See Bill Brooks's article regarding PV fires.

[BritishPete]

I understand that DC GFI's are required by the NEC in the panel-to-controller circuit when the panels are roof mounted. I've read that DC GFI's, unlike their AC counterparts, do not have the function of personnel protection in terms of tripping under mA leakage, but rather trip if there is 1/2 amp flow to ground due to a high resistance fault that might otherwise cause high current flow and lead to a fire or further damage. I've also read that there are many conflicting opinions whether these devices actually are needed; and that folks often include one because the NEC says so but not because they think it does anything, or understand what it does.

Can someone shed light upon what the function of DC GFI's really are in a solar application. Seems to me that breaking the circuit at the charge controller doesn't provide much protection to the panels and wiring up on the roof if there is a short in between, as they are going to pump current right into the fault regardless. We're installing a solar set-up in Africa, strictly 12 volt DC with a Morningstar MPPT controller, so NEC requirements aren't an issue. My partner doesn't believe a GFI really does anything so it isn't necessary. I'm not convinced either way: if a GFI is providing real protection I'm all for it. But I'd like to understand what I've got to defend.

Can someone shed some light on this?

There are two types of GFCI protection (Likely called Earth Leakage Protection in Africa). One is for Personnel protection and typically set at 4-6mA in the US and 30 mA in many European countries. The other is for Equipment protection and is normally set at 75 mA or 300 mA although higher currents are also available. Fusing (protecting) the ground (Earth) itself would completely negate the point of having a ground (EGC) in the first place.

In my opinion you are correct (especially for a 12V system); Ignore the GFCI protection making sure you have a good EGC.