Mike's take on 230.71

[c_picard]

In his PV book (Figure 230-2), it is stated that the six handle rule does not apply to the PV disconnect because it is not a service disconnect.

While I agree, it's not for the same reason. I think if it is disconnecting service entrance conductors, it's still a service disconnect, and is not one of the exceptions listed in 230.71.
The reasoning behind the six handle exception is, as I see it:

230.71 Maximum Number of Disconnects.
(A) General. The service disconnecting means for each service
permitted by 230.2, or for each set of service-entrance
conductors permitted by 230.40, Exception No. 1, 3, 4, or 5,
shall consist of not more than six switches or sets of circuit
breakers, or a combination of not more than six switches and
sets of circuit breakers, mounted in a single enclosure, in a
group of separate enclosures, or in or on a switchboard. There
shall be not more than six sets of disconnects per service
grouped in any one location.
For the purpose of this section, disconnecting means
installed as part of listed equipment and used solely for the
following shall not be considered a service disconnecting
means:
(1) Power monitoring equipment
(2) Surge-protective device(s)
(3) Control circuit of the ground-fault protection system
(4) Power-operable service disconnecting means

230.40 Number of Service-Entrance Conductor Sets.
Each service drop, set of overhead service conductors, set
of underground service conductors, or service lateral shall
supply only one set of service-entrance conductors.
Exception No. 5: One set of service-entrance conductors
connected to the supply side of the normal service disconnecting
means shall be permitted to supply each or several
systems covered by 230.82(5) or 230.82(6).

230.82 Equipment Connected to the Supply Side of Service
Disconnect. Only the following equipment shall be
permitted to be connected to the supply side of the service
disconnecting means:

(6) Solar photovoltaic systems, fuel cell systems, or interconnected
electric power production sources.


The above would allow for six PV service disconnects. The real question is would all disconnects be required to be co-located?
Or would the PV disconnect(s) need to be near one another, but not necessarily near the buildings main(s)?

I'd like to see them all in the same place. Whatsayou?

 

[GoldDigger]

The above would allow for six PV service disconnects. The real question is would all disconnects be required to be co-located?
Or would the PV disconnect(s) need to be near one another, but not necessarily near the buildings main(s)?

I'd like to see them all in the same place. Whatsayou?

If your goal is to cut off all power to loads, then it is not necessary to open PV output circuits connected to a line-side tap on the service.
If your goal is to cut off power to all wiring serving the building or passing through it, then the PV breakers would have to be part of that count.
Unless there are good operational or economic reasons to do differently, I would prefer to see a single line-side tap protected by the main CB in an AC combiner panel (sub-panel). Then separate PV equipment would connect to individual breakers in the combiner, without a six handle limit.
If you do not have a line side tap, but have the PV back feeding a main panel, then since that panel has a main disconnect you don't need to count anything else.

I would prefer to see them collocated, at least the single breaker associated with the line-side tap with the service disconnect(s).

 

[shortcircuit2]

IMO...a line side connection to existing service entrance (for...Solar photovoltaic systems, fuel cell systems, or interconnected electric power production sources.) just creates an additional service disconnect and is subject to the 6-switch rule with respect to the existing service. The addition of the new service switch must be grouped with the other service disconnects.

I base my reasoning on the basis that there is only one SERVICE POINT, thereby it is one service and must adhere to the laundry list applying to services. If an additional service point were added to the building, then we could consider the new equipment a 2nd additional service and could have up to 6 switches.

There has been much debate on this subject. In the interest of safety for first responders to a building that is on fire, it is best to have all disconnects of a service grouped, thereby allowing speedy shutdown without confusion.

shortcircuit

 

[GoldDigger]

There has been much debate on this subject. In the interest of safety for first responders to a building that is on fire, it is best to have all disconnects of a service grouped, thereby allowing speedy shutdown without confusion.
\

But removing the service voltage from a PV system does shut it down completely. With the service still intact, the only power remaining in the building will be on the wiring between the PV and the service. Still a potential hazard, just not as great. With a line-side tap, the disconnect definitely should be grouped at the main.
Usually first responders will also want to remove the DC side of the PV, especially if the DC goes to a string inverter inside the building. And they usually do that by taking a fire axe to the panels rather than trying to find a DC disconnect if they are going to be up on the roof with the panels.

 

[c_picard]

IMO...a line side connection to existing service entrance (for...Solar photovoltaic systems, fuel cell systems, or interconnected electric power production sources.) just creates an additional service disconnect and is subject to the 6-switch rule with respect to the existing service. The addition of the new service switch must be grouped with the other service disconnects.

I base my reasoning on the basis that there is only one SERVICE POINT, thereby it is one service and must adhere to the laundry list applying to services. If an additional service point were added to the building, then we could consider the new equipment a 2nd additional service and could have up to 6 switches.

There has been much debate on this subject. In the interest of safety for first responders to a building that is on fire, it is best to have all disconnects of a service grouped, thereby allowing speedy shutdown without confusion.

shortcircuit

While I agree in principle, the code as written does not support that view.

 

[Smart $]

...


The above would allow for six PV service disconnects. The real question is would all disconnects be required to be co-located?
Or would the PV disconnect(s) need to be near one another, but not necessarily near the buildings main(s)?

I'd like to see them all in the same place. Whatsayou?

I believe the part of 230.71 which states, "mounted in a single enclosure, in a group of separate enclosures, or in or on a switchboard" requires the up-to-six PV disconnects to be in the same location... but not in any particular location other than they CANNOT be in the same location as the up-to-six "normal" service disconnecting means if the combined total is over six.

 

[George Stolz]

In his PV book (Figure 230-2), it is stated that the six handle rule does not apply to the PV disconnect because it is not a service disconnect.

While I agree, it's not for the same reason. I think if it is disconnecting service entrance conductors, it's still a service disconnect, and is not one of the exceptions listed in 230.71.

It's not a service disconnect because the PV is not a service. Article 230 does not apply unless you intend to tie it to the line side of a service disconnect (parallel with service conductors.)

Service. The conductors and equipment for delivering electric energy from the serving utility to the wiring system of the premises served.

PV cells doth not a utility make.

 

[shortcircuit2]

It's not a service disconnect because the PV is not a service. Article 230 does not apply unless you intend to tie it to the line side of a service disconnect (parallel with service conductors.)



PV cells doth not a utility make.

I agree PV is not a service, it is a power production source. With a grid tied PV system, the PV will not operate without the grid. In a grid tied PV system, the PV equipment does use power from the grid. Check the specs on the GTI and you will see a power consumption value, although small.

Grid tied PV equipment is part of the wiring system of the premises served. It is connected to the service of the premises and IMO counts as one of the 6 disconnects when connected to the service to the premises and must be grouped with other disconnects of the service it is connected to.

 

[George Stolz]

I must be behind on my PV stuff. I've never seen a PV system on the line side of a service disconnect, and have always seen what I've heard referred to as "grid intertied PV" on the load side of a service disconnect, generally on the load side of a feeder, and can sell power back to the grid. Shutting off power to the service disconnect results in a safeguard within the inverter shutting down its contribution to the grid. Sometimes, even in this condition, it can continue to supply loads to sub panels served by the inverter, because it all resides on the load side of the service disconnect and has different terminals for utility (input) and load (load).

What I am describing does not fall within the scope of Article 230.

Are you guys seeing a lot of PV installed on the line side of service disconnects, essentially dumping directly into the grid?

 

[iwire]

Are you guys seeing a lot of PV installed on the line side of service disconnects, essentially dumping directly into the grid?

Not a lot but for sure some. That can eliminate the need for replacing the service panel.

 

[shortcircuit2]

Are you guys seeing a lot of PV installed on the line side of service disconnects, essentially dumping directly into the grid?

Whether it is connected line or load, I was taught that it dumps directly back to the grid. My understand is this is done by the inverter pushing back to the grid by having a slightly higher voltage than the grid thereby forcing power back into the grid. The net meter measures this and offsets what the building is consuming...

 

[George Stolz]

Whether it is connected line or load, I was taught that it dumps directly back to the grid.

Agreed, but making the leap to calling a PV disconnect a "service disconnect" is akin to saying that since a receptacle receives its power from the grid, it is a service disconnect. If the PV system is on the load side of the service disconnect, then whatever disconnects Article 690 requires you to install are not bound by the rules of Article 230.

Not a lot but for sure some. That can eliminate the need for replacing the service panel.

Appreciated - but I'm not convinced that people earlier in the thread were addressing PV on the line side of a true service disconnect, instead lumping all PV together, hence my belaboring the point.

 

[Smart $]

Whether it is connected line or load, I was taught that it dumps directly back to the grid. My understand is this is done by the inverter pushing back to the grid by having a slightly higher voltage than the grid thereby forcing power back into the grid. The net meter measures this and offsets what the building is consuming...

In simple theory, you can't "push" current onto the grid at the same time local consumption is greater than the PV output. And when the PV output is greater than local consumption, there is no need for net metering, per se. The meter only has to "spin backwards" so to speak.

When you get beyond simple theory, there a instances where the simple is not entirely true because of local consumption power factor. It depends on the time frame being discussed.

 

[Smart $]

Agreed, but making the leap to calling a PV disconnect a "service disconnect" is akin to saying that since a receptacle receives its power from the grid, it is a service disconnect. ...

For line side connections, many consider it a service disconnect. IMO, a disconnect with service conductors connected to one side is a service disconnect, no matter what's connected to the other side.

 

[meternerd]

Are you guys seeing a lot of PV installed on the line side of service disconnects, essentially dumping directly into the grid?[/QUOTE]

I see both line side and backfed downstream of the main. In talking to manufacturers, the main reason for line side connection is so the 120% rule 690.64 B2 does not apply. If you connect to a 200A panel on the load side of the main, you are limited to 40A max. If you want to put in a larger system, line side is a cheap way to do it without upsizing the service panel.

 

[meternerd]

In simple theory, you can't "push" current onto the grid at the same time local consumption is greater than the PV output. And when the PV output is greater than local consumption, there is no need for net metering, per se. The meter only has to "spin backwards" so to speak.

When you get beyond simple theory, there a instances where the simple is not entirely true because of local consumption power factor. It depends on the time frame being discussed.

At our utility, we only do Net metering. But we do separately record Kwh Delivered, Received and Net. Net is just the difference between the other two. If power is delivered to the customer, Kwh is added to the Net register and the Delivered register. If power is received, the meter subtracts Kwh from the Net, but adds that amount to the Received register. Any excess power the customer produces is credited to the account twice per year based on the wholesale rate. Easy for us to do because we have all electronic meters. With electromechanical meters, the meter would just spin forward or backward and we would bill or credit on the monthly reading. We don't require or monitor any production meters the customer may choose to install, so we have no way to know how much total solar power the customer produces. Hope this isn't too far off the subject.