2016 CEC Interpretation - Heads Up

Status
Not open for further replies.
C

c_picard

Senior Member
Location
USA
Just a polite heads up to PV integrators and respected AHJs in California - There is a "white paper" being circulated there that has at its foundation an erroneous interpretation of the definition on a PV system.

The assumptions that follow are leading some AHJs to believe that existing systems would need to be brought up to current code if a new system is installed on the same premises.

There is no indication the author consulted any of his peers before publishing and distributing this, therefore I do not see this as a consensus-based document that should be considered when applying the 2016 CEC. My hope is that those exposed will use the resources already available in CA, and seek guidance from other locales that have been navigating the 2014 NEC for years now.
 
c_picard said:
Just a polite heads up to PV integrators and respected AHJs in California - There is a "white paper" being circulated there that has at its foundation an erroneous interpretation of the definition on a PV system.

The assumptions that follow are leading some AHJs to believe that existing systems would need to be brought up to current code if a new system is installed on the same premises.

There is no indication the author consulted any of his peers before publishing and distributing this, therefore I do not see this as a consensus-based document that should be considered when applying the 2016 CEC. My hope is that those exposed will use the resources already available in CA, and seek guidance from other locales that have been navigating the 2014 NEC for years now.
Click to expand...
I encountered just such a situation here yesterday. The AHJ did not make us bring the old system up to 2014 code since it was installed under 2011 code. It's a good thing, that, since the old system did not have rapid shutdown.
 
ggunn said:
I encountered just such a situation here yesterday. The AHJ did not make us bring the old system up to 2014 code since it was installed under 2011 code. It's a good thing, that, since the old system did not have rapid shutdown.
Click to expand...

There's a compromise solution I'm supportive of for this scenario, that also happens to align well with the 2017 NEC language - install a label that illustrates which array or arrays have RSD capability, and which do not.
 
Can you post a link to the white paper? I’d like to see it. There has been a lot of discussion of how to identify structures that were built under the various RSS requirements and how to deal with mixed installations. As far as I know there is no consensus on what to do with mixed installations and now there can be up to 3 different kinds in one place. Most likely any guidance will come from the fire code end of things since it’s firefighter who will have to deal with the systems when they get to a fire. They need to know what to expect.

What an AHJ can require to be brought up to code is not a code question though since it’s not covered in the code. What triggers an AHJ to say any related part of a structure needs to be brought up to current code as part of new work is kind of up to the AHJ. I know a guy who had to pretty much rewire his older home to add outlets to rooms to meet current code because he upgraded the service panel. That was unexpected. So I can see AHJs requiring upgrades to older PV systems. Something to ask about before starting a job.
 
Last edited:
pv_n00b said:
Can you post a link to the white paper?
Click to expand...

I'd rather not amplify this person's signal here.

pv_n00b said:
Most likely any guidance will come from the fire code end of things since it’s firefighter who will have to deal with the systems when they get to a fire. They need to know what to expect.
Click to expand...

The good folks involved on the fire code side are historically reluctant to address electrical installation requirements, rightfully so in my opinion because correlating this stuff is hard enough as it is. I'll continue to advocate for a solution that resembles the 2017 NEC approach using a label that marks out which array(s) are equipped with what. Totally understand that this introduces yet more input for an officer to process, but arguably better than what we were doing.

pv_n00b said:
What an AHJ can require to be brought up to code is not a code question though since it’s not covered in the code. What triggers an AHJ to say any related part of a structure needs to be brought up to current code as part of new work is kind of up to the AHJ.
Click to expand...

Perhaps in some locales. The Code does in fact address alterations that trigger updating - AFCIs come to mind. Other AHJs have explicit admin rules that they can leverage to force updates, most I'm familiar with have an "imminent danger" clause.
 
c_picard said:
I'd rather not amplify this person's signal here.
Click to expand...

That’s not a very useful answer. There’s some “paper” out there from “someone” that leads AHJs to think that older PV systems should be brought up to code if a new one is added. With that kind of gloss over what exactly are you asking the forum? Are you just running around telling people the sky might be falling? Provide the specific arguments that the letter is making so we can effectively refute them.
 
Last edited:
That's a fair criticism, though I wasn't actually asking the forum for anything.

Here's the summary:

If we look at the NEW definition in the 2016 CEC of what a PV system is, it's clear that a building can only have one PV System. So, if there is an existing system that is altered, or has an additional subsystem installed - the entire SINGLE system (new + old subsystem) needs to comply with the current code. Subsystems must also have their AC outputs combined prior to their interconnection with the building service - no subsystem tying in at a detached garage subpanel when there is a subsystem on the house, for example.

Sharing the .pdf wouldn't be in anyone's best interest, including the author's.

Happy to continue the discussion offline.
 
c_picard said:
If we look at the NEW definition in the 2016 CEC of what a PV system is, it's clear that a building can only have one PV System. So, if there is an existing system that is altered, or has an additional subsystem installed - the entire SINGLE system (new + old subsystem) needs to comply with the current code. Subsystems must also have their AC outputs combined prior to their interconnection with the building service - no subsystem tying in at a detached garage subpanel when there is a subsystem on the house, for example.
Click to expand...

Hard to limit myself to polite terms. As you can see, the ramifications go beyond rapid shutdown.

Here's the definition that is apparently being referred to, which is now in Article 100. Same as 2014 NEC.

The total components and
sub-system that, in combination, convert solar energy
into electric energy suitable for connection to a
utilization load.
Click to expand...

I really can't see how that language leads to the ideas further up.

I have always felt that, under this definition, one inverter generally qualifies as one system. Turn it off or remove that system, and generally any others present are still suitable for connection to a utilization load. There are some situations that clearly ought to be treated differently: for example, microinverters on the same circuit sharing an overcurrent device and disconnecting means. Or systems where single-phase inverters are stacked to create split-phase or wye systems (some off-grid systems). I have felt for a while that various code sections should be revised accordingly to not suggest that multiple inverters are required to be combined before connection to distribution.
 
That definition also appears in NEC 2017. Additionally, 690.4(D) states:

(D) Multiple PV Systems. Multiple PV systems shall be permitted to be installed in or on a single building or structure. Where the PV systems are remotely located from each other, a directory in accordance with 705.10 shall be provided at each PV system disconnecting means.
Click to expand...
 
Most AHJs will probably consider the 2017 change when deciding how to interpret the 2014 edition. It doesn't make a lot sense to enforce to enforce an interpretation that is diametrically opposed to the most recent Code revision.

The 2017 Code also describes the identification and directory requirements where a building has multiple PV systems that meet different rapid shutdown requirements. You just need to clearly identify the subarrays that have NEC 2017 vs NEC 2014 vs nor rapid shutdown. You don't have to retroactively upgrade the existing systems.
 
jaggedben said:
I have always felt that, under this definition, one inverter generally qualifies as one system. Turn it off or remove that system, and generally any others present are still suitable for connection to a utilization load. There are some situations that clearly ought to be treated differently: for example, microinverters on the same circuit sharing an overcurrent device and disconnecting means. Or systems where single-phase inverters are stacked to create split-phase or wye systems (some off-grid systems). I have felt for a while that various code sections should be revised accordingly to not suggest that multiple inverters are required to be combined before connection to distribution.
Click to expand...

I think if the inverters are electrically combined before interconnecting to general loads, and have a master shutoff for the group of inverters, that they collectively would qualify as "1 system".
 
Carultch said:
I think if the inverters are electrically combined before interconnecting to general loads, and have a master shutoff for the group of inverters, that they collectively would qualify as "1 system".
Click to expand...

While I can see the sense in thinking that, I don't think that the code definition entails it. And I'm not sure I see a safety issue that would justify amending the code to make the definition more strict.

Moreover, it doesn't cover all cases sensibly. Say I sell new microinverter systems to two different homeowners, both of whom happen to have previously existing string inverter systems. Homeowner A has space in his service panel for both 2-pole breakers, so I install the new one there and call it a day. Homeowner B has no breaker spaces left, so I install a small 'combiner' sub and move the existing inverter output into it to make everything work. Does it really make sense to call B 'one system' and A 'two systems'?
 
Status
Not open for further replies.
Top