[RE-wrenches] NEC 2014 690.12 Rapid Shutdown

Bill Brooks billbrooks7 at yahoo.com
Thu Jan 16 09:45:46 PST 2014


Jeffrey,

 

Sounds like you need to get involved in the code making process since you
have so many good ideas on how to improve the language. I like
confrontational discussions as long as they lead to a better understanding
and constructive outcomes.

 

About 30 people worked on this language, so it is definitely not perfect.
However, I don't think it is quite as bad as you make it out to be. I wanted
to jump in since some of your conclusions were not correct.

 

This is a circuit requirement, not a disconnecting means requirement, since
it has to do with shock hazard of PV circuits in and around a building. This
is for firefighter safety. 30V is the international standard for touch safe
in a wet location. 240VA is to set a limit on the available power on a
circuit. Contactor combiners, which would be part of a compliant solution,
have 24V control circuits. The other reasoning for 240VA is that internally,
72-Cell PV modules can be divided into segments of this power level for the
foreseeable future (more on that another day).

 

If the conductors stay outside, you have 10' from the array to place your
shutdown device. On large central systems, this would likely be a contactor
combiner-most manufacturers sell these. If the conductors are going
immediately into the building, as with residential and integrated systems, a
shutdown device would have to be within 5' of entering the building. If goes
outside for a while, then inside the building, the total length could be no
more than 10' and no more than 5' inside the building-this is not additive.
Remember, all this is for firefighter safety.

 

As Brian Mehalic and others have pointed out, the language does not specify
where the shutdown initiating device is to be located. The lack of detail is
more for flexibility than it is to give an AHJ license to make an installer
do anything they want. 

 

With grid-tie only systems (no battery backup), it would be most convenient
and cost effective to have a system that initiates the shutdown on loss of
utility. In this way, a firefighter can do what they normally do, shut down
utility power to the building, and the rapid shutdown would automatically
initiate. This does not necessitate an additional disconnecting means for a
load-side PV connection. The main breaker could be the initiating device.
For a supply-side connection, the NEC already requires that the PV
disconnect switch be located adjacent to the service disconnecting means
(article 230).

 

The biggest issue with string inverters (central inverters) is that there is
a need to shutdown the capacitor input side of the inverter since that stays
energized for 5 minutes or more. The 10 seconds was to provide a means to
rapidly discharge the capacitors rather than requiring a relay or tripping
device. Doing something other than a relay will require a test laboratory to
evaluate the function-guess what?-we don't have a standard yet to evaluate
those products. Sounds like you might want to work on that committee.

 

It is more complicated for battery backup systems. Midnite Solar's birdhouse
products are the best I have seen so far to address this concern. Since dc
and ac circuits are not differentiated, battery backup systems need to have
a shutdown process that works independently of a utility outage for obvious
reasons, and it must shutdown both the dc circuits and the backup ac
circuits. A separate switch, like the birdhouse, would be necessary that
only controls these functions in an emergency situation.

 

Is the language not detailed-possibly. This was done to provide flexibility
rather than create problems. Fire departments have been requiring rooftop
disconnects for years in California. These disconnects are nearly worthless
from a shock prevention point of view since capacitors in the inverter stay
charged or there are multiple disconnecting means feeding each other. We
have been trying to hold the fire community off of rooftop disconnect
requirements so we could work on a solution that actually does what they
want it to do. There is a long discussion on this in the appendix of my
"Understanding the CalFire Guidelines" document on the SolarABCs website.

 

The 2014 NEC language was a compromise worked out with the solar industry
(yes string inverter companies as well) in response to the first version of
the proposal which was to require module-level shutdown. This is not
module-level shutdown, it is PV output circuit shutdown (combiner box
shutdown is another way to look at it). However, the 2017 NEC cycle is this
year and there was a lot of talk about requiring module-level shutdown this
time around. 

 

I hope this helps. I will be writing articles for IAEI journal and other
periodicals on this subject since it was a very far-reaching and potentially
confusing new requirement in the NEC. Thanks for your interest and let's
keep the constructive dialogue going on the subject. It is time to get
involved in the NEC update process again.

 

Bill Brooks.

 

From: re-wrenches-bounces at lists.re-wrenches.org
[mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Jeffrey
Quackenbush
Sent: Thursday, January 16, 2014 1:09 AM
To: RE-wrenches
Subject: Re: [RE-wrenches] NEC 2014 690.12 Rapid Shutdown

 

Wrenches,

 

There is no guidance in the Code text for where the shutdown should take
place. (1) says: "Requirements for controlled conductors shall apply only to
PV system conductors of more than 1.5m (5') in length inside a building, or
more than 3m (10') from a PV array."





So, the provisions apply if the circuit 10' from the array and 5' inside a
building, but no mention is made of where the shutdown actually needs take
place in the circuit. In the video Bill Brooks suggests that the shutdown
mechanism should also be placed within this 10'/5' boundary but that is just
an inference -- nowhere in the text is this actually specified. If that was
the intent of the Code committee, then they've done a poor job actually
expressing it in English.

 

I'm concerned that some AHJs will interpret this to exclude all central
inverter systems (without the addition of cost-inducing secondary DC-DC
converters like Tigo) because the combiner or junction box can be many feet
from the actual beginning of a home run under the array. Alternately,
permissive AHJs could allow this function to be fulfilled anywhere, meaning
that the implementation won't meet the intent of the writers.

 

I'm also concerned, as Isaac mentioned, that there are no requirements for
how the shutdown be initiated, or that it contains of the accessibility and
grouping requirements that are always included for disconnects. I really
think this should be treated and categorized as a disconnect requirement,
not a circuit requirement, because that is the ultimate function that's
intended. 

 

I'm surprised none of the inverter manufacturers have chosen to comment
here, as this could dramatically impact the sales of central inverters.

 

Jeffrey Quackenbush

 



---
This email is free from viruses and malware because avast! Antivirus protection is active.
http://www.avast.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20140116/895cf919/attachment-0003.html>


More information about the RE-wrenches mailing list