Method to safegaurd a PV System from back feeding the utility?

[Anode]

I've never encountered this before, but this is the response
I received from a third party engineer reviewing our plans to
interconnect a 150kW system.

"It is the intent of the design that there will be no
bi-directional breakers installed on this project and that
the PV system WILL NOT BACKFEED to the utility at any
time. The PV system installer shall be responsible to
provide evidence to the Utility that safeguards have
been incorporated into the PV system design to either
shut down the PV system or prevent access power
beyond that which is consumed directly by the facility
from being generated."
​

First, I can easily provide an hourly table with max ac out for
each hour and each month of the year, to then prove 112kW
would be peak output at any given time.

My question is how should we go about providing a safegaurd
as described in their response? I've never heard of anything
like this. Obviously I don't think they are talking about inverters
back feeding in the event of a grid outage, they want the
system to essentially not be net-metered.

Can you seasoned veterans lay some experience on me?

 

[GoldDigger]

I've never encountered this before, but this is the response
I received from a third party engineer reviewing our plans to
interconnect a 150kW system.

"It is the intent of the design that there will be no
bi-directional breakers installed on this project and that
the PV system WILL NOT BACKFEED to the utility at any
time. The PV system installer shall be responsible to
provide evidence to the Utility that safeguards have
been incorporated into the PV system design to either
shut down the PV system or prevent access power
beyond that which is consumed directly by the facility
from being generated."
​

First, I can easily provide an hourly table with max ac out for
each hour and each month of the year, to then prove 112kW
would be peak output at any given time.

My question is how should we go about providing a safegaurd
as described in their response? I've never heard of anything
like this. Obviously I don't think they are talking about inverters
back feeding in the event of a grid outage, they want the
system to essentially not be net-metered.

Can you seasoned veterans lay some experience on me?

There are devices called Network Protectors or Network Protector Relays which specifically sense reverse direction power flow (by sensing the phase between volts and amps) and give an alarm or open the network connection (which would, of course, shut down the GTIs and potentially the whole facility.)
Your biggest problem, IMHO, is to make sure that any such trip does not cut off power to your local loads in the process of shutting down the inverters.

Now you can also use CTs and appropriate hardware/logic to sense the difference between local load and PV output and selectively shut down some fraction of the inverters when they get too close to each other.

It is not just net metering they are worried about (nominally); they actually may be concerned about the stability of the grid if you suddenly start dumping full power back into it.
Think of the effect on the grid of suddenly and without warning shutting down your facility completely while it was running full blast without PV support. Then double that. That is what they should be concerned about.

 

[Engineer#4827502]

A reverse power relay to shut the system down should satisfy the utility. You may be able to install another control scheme without a relay but POCOs love and trust relays.

I've seen this required on utility owned campus distributions systems and network distribution systems. The campus systems, the utility did not want to run studies to see the impact of the PV system. Network distribution systems can be disrupted by new sources that could feed other loads. Networks are typically only found in dense urban areas.

If the issue the POCO has is not one of the above, you may be able to prove to them that the PV production will never exceed the load and that would be good enough. A backfeed event on a network system COULD take down a big part of a city.

 

[jaggedben]

"It is the intent of the design that there will be no
bi-directional breakers installed on this project and that
the PV system WILL NOT BACKFEED to the utility at any
time. ...​

​

Blue and red are not the same thing. Just sayin'. Do you think that they really mean the blue part, or just the red part? If they really mean no bi-direction breakers then that might preclude some solutions that would work if only what happens at the service is important. It depends as well on how you plan to interconnect.

Other than that, my only comment is that maybe this system should have storage.

Sorry that I have no concrete advice.

 

[Anode]

Now you can also use CTs and appropriate hardware/logic to sense the difference between local load and PV output and selectively shut down some fraction of the inverters when they get too close to each other.

Thank you, GD. What would turn the inverters back on? And would I be correct in presuming that the building automated system could potentially solve some of this perhaps, I guess with respect to the fact that it will be monitoring power usage already?

A reverse power relay to shut the system down should satisfy the utility. You may be able to install another control scheme without a relay but POCOs love and trust relays.

I've seen this required on utility owned campus distributions systems and network distribution systems. The campus systems, the utility did not want to run studies to see the impact of the PV system. Network distribution systems can be disrupted by new sources that could feed other loads. Networks are typically only found in dense urban areas.

If the issue the POCO has is not one of the above, you may be able to prove to them that the PV production will never exceed the load and that would be good enough. A backfeed event on a network system COULD take down a big part of a city.

The issue is the third party engineer. The utility will do a light interconnection study, and in my beginning conversations a month ago with an engineer who would review that application, she made it seem like this 150kW was a drop in the bucket for this facility, which is a very large public transfer station. Since the building is not connected to the downtown Seattle network, she said this was really a small system and wouldn't be a problem. There was no mention of any sort of power relay system, nor was there in the specifications of the project as far as I have read, multiple times.

The issue is we have not yet completed the interconnection study, perhaps if we can make them copacetic with the plans and project, we can provide something to the third party engineering firm to resolve the concern.


I have 4 Solectria 28kW inverters whose aggregate ac output are interconnecting with a 2000A main switchboard on a 200Amp breaker. Do you have any suggestions to a power relay or network protection system that would work specifically with those inverters? Where should I start looking in case we do have to take this route.

Thanks again for the responses.

 

[Anode]

[/INDENT]Blue and red are not the same thing. Just sayin'. Do you think that they really mean the blue part, or just the red part? If they really mean no bi-direction breakers then that might preclude some solutions that would work if only what happens at the service is important. It depends as well on how you plan to interconnect.

Other than that, my only comment is that maybe this system should have storage.

Sorry that I have no concrete advice.

I was responding while you were and missed it.

I was confused by the response, so I can't say what was meant. Storage is definitely not what they're after. Connection will be made to the end of the bus in the main 2000 Amp switchboard with a 200AF or Breaker. The initial riser for the project showed it being a fuse, if that helps...

 

[Engineer#4827502]

Are any existing breakers/meter that might be backfed rated for bi-directional feed? I wonder if this is the issue the engineer has noted. Is the 'no bi-directional breakers installed' a customer requirement?

It sounds to me like the POCO is OK with backfeed. Is there anything else about the existing electrical system or new PV system that seems non-ordinary?


EDIT: looks like it's a load side connection and probably has a GFP main. Is that main rated for backfeed? Maybe they don't want to replace it.

 

[Anode]

Are any existing breakers/meter that might be backfed rated for bi-directional feed? I wonder if this is the issue the engineer has noted. Is the 'no bi-directional breakers installed' a customer requirement?

It sounds to me like the POCO is OK with backfeed. Is there anything else about the existing electrical system or new PV system that seems non-ordinary?


EDIT: looks like it's a load side connection and probably has a GFP main. Is that main rated for backfeed? Maybe they don't want to replace it.

Sorry I misspoke before when recalling off memory. Just looked at the one-line for the building and the MSB is 3200Amp Rated. It has a 3200AF/3200AT ahead of the utility meter. The meter along with most of the building, do not exist yet, at least except on paper. The note for the utility meter says to coordinate installation and requirements with POCO...

 

[Engineer#4827502]

If the main is upstream of the utility meter, line side won't happen unless it's separately metered. I imagine your problem here is the backfeed ratings/listings of the proposed equipment.

 

[ggunn]

I've never encountered this before, but this is the response
I received from a third party engineer reviewing our plans to
interconnect a 150kW system.

"It is the intent of the design that there will be no
bi-directional breakers installed on this project and that
the PV system WILL NOT BACKFEED to the utility at any
time. The PV system installer shall be responsible to
provide evidence to the Utility that safeguards have
been incorporated into the PV system design to either
shut down the PV system or prevent access power
beyond that which is consumed directly by the facility
from being generated."
​

First, I can easily provide an hourly table with max ac out for
each hour and each month of the year, to then prove 112kW
would be peak output at any given time.

My question is how should we go about providing a safegaurd
as described in their response? I've never heard of anything
like this. Obviously I don't think they are talking about inverters
back feeding in the event of a grid outage, they want the
system to essentially not be net-metered.

Can you seasoned veterans lay some experience on me?

Third party... is this a PACE project? Is the "no backfeeding" requirement coming from the customer, the utility, or PACE? It strikes me as strange that a third party reviewer is telling you (the designer, I presume) what the intent of the design is.

 

[jaggedben]

I was confused by the response, so I can't say what was meant. Storage is definitely not what they're after. Connection will be made to the end of the bus in the main 2000 Amp switchboard with a 200AF or Breaker. The initial riser for the project showed it being a fuse, if that helps...

The only reason I mentioned storage is that I gather that some of the commercial scale storage systems that are aimed at demand management and reducing demand charges would probably be capable of being used in this application. That's probably a much more expensive solution than you had in mind, but it may speak to your question of what kind of technology manages the inverter turning back on.

With that said, I'm not an expert on this at all.

 

[Anode]

Third party... is this a PACE project? Is the "no backfeeding" requirement coming from the customer, the utility, or PACE? It strikes me as strange that a third party reviewer is telling you (the designer, I presume) what the intent of the design is.

Not a pace project. But let me clarify further.. The third party firm wrote the specifications for the PV project, and wrote the specifications for the entire project, which is a large new facility.

All that said, no where in the specifications does it support their response of it being 'intent of the design'. Also, in early talking with the utility about interconnection over 100kW, their seasoned engineer did not indicate any peculiarities with the size of the system, or the building's location to the area where backfeeding the grid is a critical (downtown network).

While I am still awaiting clarification from the third party firm, I wanted to reach out to the experts here on a potential solution in the event that they will hold us to this standard.

 

[Engineer#4827502]

With a controller and an inverter capable of curtailment, you should be able to develop a scheme that cuts production equal to or lower than consumption. This would require access to the data from the utility meter or a separate meter installed in series with the utility meter.