GEC for solar installations

[Werone]

The grounding electrical conductor sizing from the code book seems silly for solar installations.

Lets look at a 25 panel Enphase microinverter installation.

One panel, 60 cell poly, under fault conditions can output a maximum current about 1.072 times above its rated STC current at time 0, or when the short occurs. I am guessing that three 60 HZ cycles later the voltage drops to zero and the current is minimal, probably equivalent to the number of photons hitting the surface of the Silicon, since the voltage and the current are directly proportional then I expect the current to be miniscule. (All UL listed inverters have Ground Fault detection circuits.) Remember the panels would need to all have this ground fault at the same time.

The granularity of an enphase inverter system has 25 discrete inverter/panel systems that makes the probability of simultaneous failure very low, but even so, the 25 simultaneous DC ground faults would possibly output 200 amps for 3 cycles or .05 seconds after which the current would drop to near zero. Why do I need a GEC conductor sized at 8 AWG? Seperately derived systems usually have a generator or battery system that can output a much larger (100+ times) fault current than a solar system, and that is what this rule had in mind when it was conceived.

The fault current for many inverters does not go above its rating, a 8000 watt SMA (240 VAC) inverter for example has a fault current specification of 33.33 Amps, yet I am required to put in a 8 AWG wire for seperately derived systems. Remember that the nature of the PV panels to drop voltage and current after about 5 one hundredths of a second. The enphase microinverters output fault current is like one amp for three 60 HZ cycles.... it just seems that we are following a rule blindly or am I missing something? Why do I need an 8 AWG GEC from my inverters to the main panel?

 

[c_picard]

Build a case and propose a code change.

What size equipment ground would you run for a 8k inverter? With correction factors(and other articles) applied you may be looking at an 8awg anyhow. Outside of conduit you need a 6 regardless. Oh, and try to find a fitting for attaching to a grounding electrode for anything smaller.

 

[iwire]

The GEC has absolutely nothing to do with the fault current the system is capable of. Fault current is the job of the EGC.

 

[Werone]

The GEC has absolutely nothing to do with the fault current the system is capable of. Fault current is the job of the EGC.

What is the role of the GEC in this type of installation? Why is it important, why can't I just use an equipment ground?

 

[Werone]

Build a case and propose a code change.

What size equipment ground would you run for a 8k inverter? With correction factors(and other articles) applied you may be looking at an 8awg anyhow. Outside of conduit you need a 6 regardless. Oh, and try to find a fitting for attaching to a grounding electrode for anything smaller.

The equipment ground at 10 AWG would be more than enough. I would not use a GEC. I would run an equipment ground to the main ground bus of the homes panel or the businesses panelboard. Why do I need a GEC for this type of installation? The currents are very low, and the fault currents are only sustainable for hundredths of a second. What benefit or additional safety or gain in performance do I get from running an uninterrupted wire that as of right now seems to only add cost?

 

[Marvin_Hamon]

The equipment ground at 10 AWG would be more than enough. I would not use a GEC. I would run an equipment ground to the main ground bus of the homes panel or the businesses panelboard. Why do I need a GEC for this type of installation? The currents are very low, and the fault currents are only sustainable for hundredths of a second. What benefit or additional safety or gain in performance do I get from running an uninterrupted wire that as of right now seems to only add cost?

You need a DC GEC, and that's what we are talking about in this thread, because it is an NEC requirement and because it is a safety issue. The DC GEC from the inverter to ground is for the DC side so the inverter AC output is not relevant. The AC output of the inverter is grounded through the existing AC service entrance.

The purpose of the GEC is to ensure that any equipment that you might touch in the PV system is at the same potential as the ground under your feet. This is good because it keeps people from being shocked. It should not carry any PV system fault current since the ground is a very poor conductor compared to an EGC. The EGC is tasked with carrying fault current for the purpose of actuating any overcurrent protecting devices. The installation requirements for the GEC are different from the EGC to ensure that the GEC stays connected even if equipment is disconnected or EGC's are disconnected. The GEC is also tasked with grounding high voltages that might occur from a close lightening strike or an unintended contact with high voltage lines. These require a different path than the EGC has to provide.

The Soares big book of grounding is a great reference for anyone who whats to understand more about both system and equipment grounding issues.

 

[ggunn]

You need a DC GEC, and that's what we are talking about in this thread, because it is an NEC requirement and because it is a safety issue. The DC GEC from the inverter to ground is for the DC side so the inverter AC output is not relevant. The AC output of the inverter is grounded through the existing AC service entrance.

The purpose of the GEC is to ensure that any equipment that you might touch in the PV system is at the same potential as the ground under your feet. This is good because it keeps people from being shocked. It should not carry any PV system fault current since the ground is a very poor conductor compared to an EGC. The EGC is tasked with carrying fault current for the purpose of actuating any overcurrent protecting devices. The installation requirements for the GEC are different from the EGC to ensure that the GEC stays connected even if equipment is disconnected or EGC's are disconnected. The GEC is also tasked with grounding high voltages that might occur from a close lightening strike or an unintended contact with high voltage lines. These require a different path than the EGC has to provide.

The Soares big book of grounding is a great reference for anyone who whats to understand more about both system and equipment grounding issues.

I think the issue here is that it is a microinverter installation and the only DC in the system is in the short cables from each module to its inverter.

 

[iwire]

What is the role of the GEC in this type of installation?

The NEC says the following about the function of the GEC

250.4(A) Grounded Systems.
(1) Electrical System Grounding. Electrical systems that
are grounded shall be connected to earth in a manner that
will limit the voltage imposed by lightning, line surges, or
unintentional contact with higher-voltage lines and that will
stabilize the voltage to earth during normal operation.

Now we can argue for year about if a GEC can do all those things but it matters not, the NEC says we have to do it.

why can't I just use an equipment ground?

I can't answer that and I doubt anyone can give you an actuate answer beyond 'Because the NEC says you can't.


I don't want to seem rude but I find it funny you are worried about running an 8 AWG, that is small stuff, get back to me when it is 300' of 3/0 copper for a GEC.

 

[Marvin_Hamon]

I think the issue here is that it is a microinverter installation and the only DC in the system is in the short cables from each module to its inverter.

The length of the DC conductors does not matter, if they are there then the system requires all the DC grounding covered in the NEC. There is no exclusion for DC conductors less that a particular length.

This is one of the problems with some of the currently advertised "AC PV Modules" that are just DC modules with a preinstalled microinverter. These are not AC PV modules by the NEC definition since they are not a listed unit with no accessible DC wiring, so they still have to have all the DC grounding of a standard string or central inverter PV system. It's going to be difficult to explain to people the difference when real AC PV Modules are more commonly available.

 

[Marvin_Hamon]

I don't want to seem rude but I find it funny you are worried about running an 8 AWG, that is small stuff, get back to me when it is 300' of 3/0 copper for a GEC.

I guess when the biggest conductor you are ever running is a #8 then it seems massive.

It is interesting how sometimes the default thinking is "If I don't understand how a requirement works then I don't need to do it" instead of "If I don't understand how it works I'm going to do it every time so I don't find out the hard way why someone who might know more than me says to do it." I see this all the time.

 

[Werone]

I learn from taking a position, or postulating. I thank you all for your insight, I hope that my question helps others understand the meaning or reasoning behind the NEC code. The basic question here arose from the fact that the currents are low and can be handled by a smaller gauge wire. I want efficient systems, and I wanted to know why the 8 AWG wire was required and why I needed a GEC for this particular application. I know if lightning hit my array not even a GEC of a much larger size would work, I know that if I had a short on every single panel the short would be local to the distance between the panel and the inverter, the DC wiring, and the GEC would not direct the short to ground since it must travel back to the negative side of the panel. Its like adding a Grounding Wire to the negative terminal of a battery of a UPS and taking that wire to the main panels bus bar. A battery has alot more current than solar systems.

 

[iwire]

It is interesting how sometimes the default thinking is "If I don't understand how a requirement works then I don't need to do it" instead of "If I don't understand how it works I'm going to do it every time so I don't find out the hard way why someone who might know more than me says to do it." I see this all the time.

I agree, and in my opinion everyone of us is guilty of it at times. To think it is only others that think that way is wishful thinking.

When I recognize I am trying to rewrite the code to fit my ideas I try to remember that the code is a consensus standard more than 100 years in the making. It is not perfect but the code making panel members certainly had more information than I do about the need for a code section.

 

[Marvin_Hamon]

I agree, and in my opinion everyone of us is guilty of it at times. To think it is only others that think that way is wishful thinking.

When I recognize I am trying to rewrite the code to fit my ideas I try to remember that the code is a consensus standard more than 100 years in the making. It is not perfect but the code making panel members certainly had more information than I do about the need for a code section.

Too true iwire, guilty as charged. As we get more experienced and actually understand why a requirement is in the code then we start to ask ourselves if that requirement is relevant to a particular installation, is the intent satisfied by something else we are already doing, or can we provide an equivalent level of protection another way that is maybe less expensive and easier to do. Then we can make our case to the AJH for an exception to the code. While blindly following the code will make for a relatively safe installation, sometimes it is overly conservative. But only years of experience and training equips us to make that call. I request alternate materials and methods exemptions for code requirements on occasion and usually have no trouble making my case.

 

[Marvin_Hamon]

I learn from taking a position, or postulating. I thank you all for your insight, I hope that my question helps others understand the meaning or reasoning behind the NEC code.

It is a great question Werone. Grounding in general is difficult for most people to get straight. The purpose of the grounding electrode system and the grounding electrode conductors are particularly difficult since they are not directly involved in clearing a fault or conducting fault current from the system they are protecting.

 

[jaggedben]

since the voltage and the current [from a PV panel] are directly proportional then I expect the current [after an inverter has shut down due to a ground fault] to be miniscule.

Some small corrections: the voltage and current from a PV panel are inversely proportional, roughly speaking. If ground fault detection on the inverter successfully opens the source circuit, then current will be zero, but there are scenarios where current could still flow after that happens. For example, if you touch the system.

You need a DC GEC, and that's what we are talking about in this thread, because it is an NEC requirement and because it is a safety issue. The DC GEC from the inverter to ground is for the DC side so the inverter AC output is not relevant. The AC output of the inverter is grounded through the existing AC service entrance.

The purpose of the GEC is to ensure that any equipment that you might touch in the PV system is at the same potential as the ground under your feet.
...

Actually the requirement for a #8 GEC on micro-inverter systems comes from the AC side, namely table 250.66. If micro-inverter systems were only required to size the GEC from the roof according to requirements for DC, the GEC could be smaller (leaving aside other considerations in the code, such as 250.64). (Of course, in most cases it would just bump down to #10, the size of most solar panel leads.)

The OP might have something of a point concerning micro-inverter systems, since it would take multiple coincidental failures to create a situation where the potential between the PV equipment and the ground beneath your feet was greater than about 40V DC. His point might even extend to solar in general. Why can DC systems can have smaller GECs than AC systems, and why should the AC side of utility interactive solar systems have to size a ground wire according to 250.66 and not 250.122?