PV backfeed + existing load breaker sizing

[jayzbond007]

Hi,


I am designing a PV array that will have (2) strings of modules, which will require 20A dpb each and consolidate into a load center. This will tie back to the main service panel (MSP). Since the MSP is full, I will move (1) 40A dpb circuits to the load center to make room and replace that breaker in the MSP with a breaker size that can accomodate both the PV and the existing 40A circuit. My question is, how do you size that? Since the PV is a backfeed and the existing 40A is a load, do I take my 40A for the PV + the existing 40A and use an 80A dpb?


Any help is greatly appreciated.


Thanks!

 

[ggunn]

Hi,


I am designing a PV array that will have (2) strings of modules, which will require 20A dpb each and consolidate into a load center. This will tie back to the main service panel (MSP). Since the MSP is full, I will move (1) 40A dpb circuits to the load center to make room and replace that breaker in the MSP with a breaker size that can accomodate both the PV and the existing 40A circuit. My question is, how do you size that? Since the PV is a backfeed and the existing 40A is a load, do I take my 40A for the PV + the existing 40A and use an 80A dpb?


Any help is greatly appreciated.


Thanks!

What inverter(s) are you using?

 

[GoldDigger]

Hi,


I am designing a PV array that will have (2) strings of modules, which will require 20A dpb each and consolidate into a load center. This will tie back to the main service panel (MSP). Since the MSP is full, I will move (1) 40A dpb circuits to the load center to make room and replace that breaker in the MSP with a breaker size that can accomodate both the PV and the existing 40A circuit. My question is, how do you size that? Since the PV is a backfeed and the existing 40A is a load, do I take my 40A for the PV + the existing 40A and use an 80A dpb?



Any help is greatly appreciated.

Thanks!

You can use a 40A breaker, but the AHJ may require the feeder wires to be sized for 80A .

 

[jayzbond007]

PV backfeed + existing load breaker sizing

Hi,

I am using Enphase microinverters for this sytem.

 

[jayzbond007]

PV backfeed + existing load breaker sizing

Okay, so you are saying that I can just use a 40A which is the same 40A for the existing load, but why wouldn't you include the PV backfeed (assuming the AHJ doesn't have any requirements)? I am looking at this as though the conductors are taking both loads in either direction, so wouldn't you use an 80A instead and size conductors to handle the added loads?

 

[ggunn]

Hi,

I am using Enphase microinverters for this system.

Ah. We usually mean a series connection when we say strings; Enphase are all in parallel. Hence my confusion. Carry on.

 

[ggunn]

Hi,


I am designing a PV array that will have (2) strings of modules, which will require 20A dpb each and consolidate into a load center. This will tie back to the main service panel (MSP). Since the MSP is full, I will move (1) 40A dpb circuits to the load center to make room and replace that breaker in the MSP with a breaker size that can accomodate both the PV and the existing 40A circuit. My question is, how do you size that? Since the PV is a backfeed and the existing 40A is a load, do I take my 40A for the PV + the existing 40A and use an 80A dpb?


Any help is greatly appreciated.


Thanks!

Are the busbars in your MSP rated at least 200A?

 

[ggunn]

You can use a 40A breaker, but the AHJ may require the feeder wires to be sized for 80A .

If he has loads in the subpanel he is splitting out to accommodate the 40A of PV, and he uses a 40A breaker in the main to feed the sub, wouldn't the minimum rating for his subpanel busbars and the conductors between the main and the sub be (40A + 40A)/1.2 = 66.7A?

 

[jayzbond007]

PV backfeed + existing load breaker sizing

The bus bar rating is 100A. This connection I am designing is a line side connection at the MSP.

 

[GoldDigger]

Okay, so you are saying that I can just use a 40A which is the same 40A for the existing load, but why wouldn't you include the PV backfeed (assuming the AHJ doesn't have any requirements)? I am looking at this as though the conductors are taking both loads in either direction, so wouldn't you use an 80A instead and size conductors to handle the added loads?

By the nature of the Grid Tied Inverter, the voltage will be exactly in phase with the line voltage and so any loads with PF=1 will exactly cancel some portion of the current sourced by the GTI.
In that case, either you will have
1. 40A going through the breaker and feeder to the loads, with the PV off, or
2. 40A going through the breaker and feeder from the GTI toward POCO, or
3. Anywhere in between those two cases, with the net current through feeder and breaker being less that 40A and with varying direction (referenced to the phase of the voltage) depending on the relative size of the GTI output and the load at that moment.

If you introduce a very very poor power factor load (lets say 100% inductive, for a PF of zero) then you could have the vector sum of the two currents going through the breaker and feeder, for a total of 40 x sqrt(2) = 56.5A. Very unlikely, but it would blow the breaker at the main.
If the load has a PF of ~.5 or better the cancellation of the resistive portion of the load current by the GTI output will still leave the net current lower than 40A.

For the feeder, on the other hand, some AHJs will assume that someone might later tap that feeder somewhere in the middle with a load that could then draw a total of 80A without tripping either set of breakers. Only half that current will be coming from either side of the tap, but.....
YMMV.

 

[jaggedben]

I am designing a PV array that will have (2) strings of modules, which will require 20A dpb each and consolidate into a load center. This will tie back to the main service panel (MSP). Since the MSP is full, I will move (1) 40A dpb circuits to the load center to make room and replace that breaker in the MSP with a breaker size that can accomodate both the PV and the existing 40A circuit.

The bus bar rating is 100A. This connection I am designing is a line side connection at the MSP.

These two statements don't seem consistent, unless the 40A breaker you are moving is a service disconnecting means. (Is it for the air-conditioner or something?) If that is the case, have you checked whether you could install a 40-40 quad instead of moving the circuit? That would be easier. If it is not the case, you should more fully explain what you're doing.

 

[GoldDigger]

Okay, so you are saying that I can just use a 40A which is the same 40A for the existing load, but why wouldn't you include the PV backfeed (assuming the AHJ doesn't have any requirements)? I am looking at this as though the conductors are taking both loads in either direction, so wouldn't you use an 80A instead and size conductors to handle the added loads?

One more thing to keep in mind is that if you will be using the 120$% rule you must mount the backfed breaker to the end of the bus opposite the main.
Is that where the existing 40 is? Or is there space at the bottom?

 

[jayzbond007]

PV backfeed + existing load breaker sizing

The main service panel that I am tying into has only two circuits. One 50A to an existing sub panel and one 40A to A/C unit. The panel itself is 100A rated. No tandem breakers allowed on this panel. These are all service disconnects, so the 120 rule does not apply.

 

[shortcircuit2]

If you introduce a very very poor power factor load (lets say 100% inductive, for a PF of zero) then you could have the vector sum of the two currents going through the breaker and feeder, for a total of 40 x sqrt(2) = 56.5A. Very unlikely, but it would blow the breaker at the main.
If the load has a PF of ~.5 or better the cancellation of the resistive portion of the load current by the GTI output will still leave the net current lower than 40A.

The NEC hasn't considered your calculation above, right?

 

[jaggedben]

The NEC hasn't considered your calculation above, right?

Not directly, AFAIK, which is to say I don't think it appears in the substantiations for any proposals that became part of the code. However, the fact that the 120% rule is so conservative (i.e., it's not the 200% rule) means that in many cases it won't be an issue. John Wiles has stated that the reason for the conservatism has to do with how panelboards are tested by UL, but if I understand correctly it ends up also reducing the probability of power factor concerns.

 

[Smart $]

The main service panel that I am tying into has only two circuits. One 50A to an existing sub panel and one 40A to A/C unit. The panel itself is 100A rated. No tandem breakers allowed on this panel. These are all service disconnects, so the 120 rule does not apply.

The 120% rule will apply to the feeder and your new subpanel.

 

[tallgirl]

Not directly, AFAIK, which is to say I don't think it appears in the substantiations for any proposals that became part of the code. However, the fact that the 120% rule is so conservative (i.e., it's not the 200% rule) means that in many cases it won't be an issue. John Wiles has stated that the reason for the conservatism has to do with how panelboards are tested by UL, but if I understand correctly it ends up also reducing the probability of power factor concerns.

There are a number of issues which concern me, that I don't think the NEC is thinking about as much as it should be. For example, Exeltech, OutBack, SMA (and others, I'm sure) produce 120 volt grid-tied and grid-interactive inverters. One might think that a pair of 30A inverters (or two strings of Exeltech microinverters) could be put in a panel, with a 30A2P interconnection breaker, 30A conductors, etc. (derate as usual) and be fine, except that because they are 120 volt inverters, a neutral imbalance of twice that (60A) could happen if only half the inverters (one leg) were producing power and the other leg was at the limit for load.

 

[Smart $]

There are a number of issues which concern me, that I don't think the NEC is thinking about as much as it should be. For example, Exeltech, OutBack, SMA (and others, I'm sure) produce 120 volt grid-tied and grid-interactive inverters. One might think that a pair of 30A inverters (or two strings of Exeltech microinverters) could be put in a panel, with a 30A2P interconnection breaker, 30A conductors, etc. (derate as usual) and be fine, except that because they are 120 volt inverters, a neutral imbalance of twice that (60A) could happen if only half the inverters (one leg) were producing power and the other leg was at the limit for load.

Can you explain how you get 60A???

I'm just not seeing it.

 

[GoldDigger]

Can you explain how you get 60A???

I'm just not seeing it.

30A of unbalanced generation on L1 going through the neutral plus 30A of unbalanced load on L2 going in the same direction through the neutral. Relative phase angle of 180 (L1 versus L2) + 180 (in versus out) degrees, so the currents in the neutral add.
One strong reason to use a 120/240 GTI with phase voltage monitoring instead of separate 120 GTIs.

For three phase using multiple microinverters each with one line-to-line phase, the situation has not been thought through either.

And it gets really hairy when you try to use balanced three phase GTI on a high leg delta service with open delta using a large lighting pot and a small secondary phase pot from POCO.

Glad to see that tallgirl is aware of the issue and bringing it up!

 

[jaggedben]

30A of unbalanced generation on L1 going through the neutral plus 30A of unbalanced load on L2 going in the same direction through the neutral. Relative phase angle of 180 (L1 versus L2) + 180 (in versus out) degrees, so the currents in the neutral add. One strong reason to use a 120/240 GTI with phase voltage monitoring instead of separate 120 GTIs.

I'm with Smart$, not seeing it either. Seems to me the neutral current should still go in opposite directions and balance each other out. I'll wait and hope he draws the diagram though, he's better at that.

For three phase using multiple microinverters each with one line-to-line phase, the situation has not been thought through either.

??? Please explain what has not been thought through.