New Code Requirements

[Sunking]

Seems local FD are fed up with PV systems. A new code requirement can virtually stop all new installations or make them extremely expensive. Makes no difference is grid tied or stand alone.

690.12 PV Arrays on Buildings Response to Emergency
Shutdown. . For PV Systems installed on roofs of buildings,
photovoltaic source circuits shall be deenergized from
all sources within 10 seconds of when emergency shutdown
is initiated or when the PV power source disconnecting
means is opened. When the source circuits are deenergized,
the maximum voltage at the module and module conductors
shall be 80 volts or less.

It is going to be real interesting to see how installers comply with the new rules. Should slam the breaks on all new installs.

[Volusiano]

Which local FD for which city?

[bcroe]

80 volts

Seems local FD are fed up with PV systems. A new code requirement can virtually stop all new installations or make them extremely expensive. Makes no difference is grid tied or stand alone.

690.12 PV Arrays on Buildings Response to Emergency
Shutdown. . For PV Systems installed on roofs of buildings,
photovoltaic source circuits shall be deenergized from
all sources within 10 seconds of when emergency shutdown
is initiated or when the PV power source disconnecting
means is opened. When the source circuits are deenergized,
the maximum voltage at the module and module conductors
shall be 80 volts or less.

It is going to be real interesting to see how installers comply
with the new rules. Should slam the breaks on all new installs.

Is there any plan on how this could/should be done? A couple
panels in series can hit 80 V. Guess you could bring each pair
of series panels to a multi pole switch, that could break a string
into small sections. Shorting a string is no guarantee, cause the
full voltage could reappear at a break.

My first thought, connect a box across each pair of panels. With an
isolated signal (12V perhaps) applied, the box would do nothing to
interfere. But with the control voltage (wired to all such boxes)
removed, each box would fail safe into a short circuit across its
panels.

And how does the new reg apply to a new system inspection,
or test to see it really meets the requirement? Ground mount
are not affected?
Bruce Roe

[inetdog]

Ground mount
are not affected?

Correct. Not the same type of hazard to fire fighters.
They are not likely to climb on a ground mount to ventilate it, and if they are worried about high voltage they can just not put water on it and let it burn.

[J.P.M.]

Seems local FD are fed up with PV systems. A new code requirement can virtually stop all new installations or make them extremely expensive. Makes no difference is grid tied or stand alone.

690.12 PV Arrays on Buildings Response to Emergency
Shutdown. . For PV Systems installed on roofs of buildings,
photovoltaic source circuits shall be deenergized from
all sources within 10 seconds of when emergency shutdown
is initiated or when the PV power source disconnecting
means is opened. When the source circuits are deenergized,
the maximum voltage at the module and module conductors
shall be 80 volts or less.

It is going to be real interesting to see how installers comply with the new rules. Should slam the breaks on all new installs.

Perhaps it would be possible for someone (Naptown ?) in the business to get in contact with SEIA or some such industry group and see what they know.

[inetdog]

Is there any plan on how this could/should be done? A couple
panels in series can hit 80 V. Guess you could bring each pair
of series panels to a multi pole switch, that could break a string
into small sections. Shorting a string is no guarantee, cause the
full voltage could reappear at a break.

My first thought, connect a box across each pair of panels. With an
isolated signal (12V perhaps) applied, the box would do nothing to
interfere. But with the control voltage (wired to all such boxes)
removed, each box would fail safe into a short circuit across its
panels.

And how does the new reg apply to a new system inspection,
or test to see it really meets the requirement?
Bruce Roe

1. The new regulation applies to the permit inspection for a NEW installation in an area that has adopted the 2014 NEC version. Quite a few states will not do that for three or four years. California has just adopted the 2011 and a few states are still back in 2008. But it will cause a lot of related equipment to come on the market soon.
2. The 2014 rule exempts wiring which is within 10 feet of the array and at the same time less than 5 feet inside a building. So shutdown at a combiner box or putting your GTI on the roof should satisfy it just fine. Microinverters and some optimizers also comply already. It is interesting that one section explicitly exempts wiring within 10 feet of the array while another section specifically mentions modules and module connectors. I suspect that this is just "modular" writing so that the exemption can be easily removed in the 2017 NEC.
3. So you can still have 600 volts on the internal wiring of the array, as long as you disconnect at least one end of the string. Ungrounded arrays may require a disconnect at each end in case there is a ground fault at the same time.
4. It could, however, add a few hundred dollars (or more) to the cost of a string inverter system, based on installing combiners which include contactors.

Take a look at http://www.youtube.com/watch?v=sUXShMZJorQ for a video on the subject which includes the chair of the NEC panel that wrote the change.

[bcroe]

1. The new regulation applies to the permit inspection for a
NEW installation in an area that has adopted the 2014 NEC version.

Take a look at http://www.youtube.com/watch?v=sUXShMZJorQ for a video
on the subject which includes the chair of the NEC panel that wrote the change.

No sound on that video, on my Apple. Bruce Roe

[inetdog]

No sound on that video, on my Apple. Bruce Roe

Works fine on my Mac using Firefox 26.0 and OS 10.6.8. Since it uses Flash, it may not play correctly in Safari.

[Sunking]

Which local FD for which city?

Any state or city that adopts NEC 2014 code cycle.

[Sunking]

Is there any plan on how this could/should be done? A couple
panels in series can hit 80 V.

Sure it is easy to do, already equipment out on the market to do just that.

Basically same thing we have to do in Data Centers called a EPO or Emergency Power Off button aka Employee Pissed Off button. Where by it is a very large RED Button you push and everything is diconnected. Ever notice one at self service gas stations. Technically very simple to do. Each panel has a electro-mechanical disconnect switch added on to it. Gets real expensive real fast and a lot more wiring to control it all.

As of now if you have a fire at your house with panels on it the very first thing the FD will do is go on top if there is enough room between panel rows and smash all your panels with a fire ax. If there is not enough room for them to get up there, they sit back and watch your home burn to the ground and keep it from spreading to your neighbors house. Solar systems are getting some very bad reviews from insurance and FD's. Last year there was a very large fire at a meat plant with a large solar array. Fire Fighters could not get on the roof and had to watch a 300,000 square foot building burn down for two days.

[bcroe]

Disconnect

Sure it is easy to do, already equipment out on the market to do just that.

Basically same thing we have to do in Data Centers called a EPO or Emergency Power Off button aka Employee Pissed Off button. Where by it is a very large RED Button you push and everything is diconnected. Ever notice one at self service gas stations. Technically very simple to do. Each panel has a electro-mechanical disconnect switch added on to it. Gets real expensive real fast and a lot more wiring to control it all.

Sure we had the big button in all my labs. But it just opened the main breakers with a control coil. For
PV we have to go after each of possibly dozens (or hundreds?) of individual panels. Are you saying, there
is already a mechanical activator designed to disconnect panels? Does it default to open circuit, or do you
need intact wiring & power source to do the disconnect? Outdoor operation?

My premise is, a high voltage string must be made non functional, not just disconnected at the ends.

So far I'm still liking the idea of adding a fairly sensitive 24V relay to every pair of panels, a normally
closed contact just shorts them out when no control power applied. No high voltage circuit breaking. No
great cost, esp if built into this class panel. Just run a doorbell wire pair around. A solid state version
would use next to no power even with fail safe op.

I had been thinking about a method like this, to automatically throttle back some strings
(short a panel or 2) if power got a bit high on a sunny day. Shorting a couple panels in a
string of 12, would pretty much shut down that string without any high voltage switching
or big inductive transients. Bruce Roe

[inetdog]

My premise is, a high voltage string must be made non functional, not just disconnected at the ends.

That is a perfectly reasonable premise from a pure safety standpoint, but is it not what the 2014 NEC requires and will cost you a lot of money for a questionable increase in safety.

For now, a single contactor (or maybe two) at the combiner level will meet the requirements at the minimum cost and complexity.
If you feel you have to go to the module level, then the economics get tilted in the direction of microinverters, which do not put panels in a string at all and include all of the necessary circuitry for compliance.

[bcroe]

That is a perfectly reasonable premise from a pure safety standpoint, but is it
not what the 2014 NEC requires and will cost you a lot of money for a questionable increase in safety.

For now, a single contactor (or maybe two) at the combiner level will meet the requirements at the
minimum cost and complexity.
If you feel you have to go to the module level, then the economics get tilted in the direction of microinverters,
which do not put panels in a string at all and include all of the necessary circuitry for compliance.

I certainly don't understand the whole picture here. But a 400V string which happens to pick up a
ground (by damage or intent) is still a 400V hazard. Separating strings doesn't help much; even
one 8A string is a serious personal hazard. I wouldn't hazard it even at night. What will the
firemen think when they discover the facts about the economical but not necessarily effective new rules?

Are the proposed contactors normally unpowered, activated by an emergency? This mode would fail
if anything in the control went wrong. If contactors drop out fail safe with loss of power, they will be
an annoying use of continuous power.

These contactors will have to deal with interrupting high voltage DC, an expensive requirement. A very
small normally shorting relay at the panel would have no such high voltage need, no doubt redesigned
to very low power solid state in the terminal box before long.

The quick solution certainly would be microinverters, which have other advantages. With my nearest
panels 240' from my inverters, and the farthest some 400', I have a big appreciation for higher
voltage strings. Bruce Roe

[inetdog]

I certainly don't understand the whole picture here. But a 400V string which happens to pick up a
ground (by damage or intent) is still a 400V hazard. Separating strings doesn't help much; even
one 8A string is a serious personal hazard. I wouldn't hazard it even at night. What will the
firemen think when they discover the facts about the economical but not necessarily effective new rules?

Are the proposed contactors normally unpowered, activated by an emergency? This mode would fail
if anything in the control went wrong. If contactors drop out fail safe with loss of power, they will be
an annoying use of continuous power.

The expectation (which may or may not change) is that actual module level shutdown may be introduced in the 2017 NEC. Who knows?

I agree that just interrupting, or ungrounding, the string will not be perfect, but if is limits the exposure to an area within 10' of the panels that is still an improvement, and should be good protection for firefighters inside the building. (The requirement is no more than 10' from the array AND no more than 5' after entry into the building envelope.

Normally closed contactors certainly will not do the job, and powered contactors will consume power. One potential solution is the mechanism currently used in shunt-trip breakers. They latch mechanically but can be tripped by a pulse through an external contact. Shunt trip can be combined with the normal circuit breaker function at little extra cost, so you are not adding another layer of DC arc interrupting devices.
They would have to be manually reset, so there is no problem with needing power to close the switch so that you can produce power.

[Sunking]

I certainly don't understand the whole picture here. But a 400V string which happens to pick up a
ground (by damage or intent) is still a 400V hazard. Separating strings doesn't help much; even
one 8A string is a serious personal hazard.

I have not talked to others yet about exactly how to implement it, but it is not isolating a string, it is isolating every individual panel from what I can ascertain.