Here's a question

Status
Not open for further replies.

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
OK, so under the 2014 NEC we can finally have legit AC combiner panels under 705.12(D)(2)(3)(c), which says the bus rating must be greater than or equal to the total of all the breaker ratings in the panel, both supply and load, disregarding the breaker that connects the panel to the service. Groovy.

But... I have an AC combiner that combines three single phase inverters into a three phase service - 60A AB, 30A BC, and 60A CA. By the letter of the code my panel busbars must be rated at least 150A, but the most that hits any one bus is 120A, so doesn't it make sense that I should be able to use a 125A panel?
 

Carultch

Senior Member
Location
Massachusetts
OK, so under the 2014 NEC we can finally have legit AC combiner panels under 705.12(D)(2)(3)(c), which says the bus rating must be greater than or equal to the total of all the breaker ratings in the panel, both supply and load, disregarding the breaker that connects the panel to the service. Groovy.

But... I have an AC combiner that combines three single phase inverters into a three phase service - 60A AB, 30A BC, and 60A CA. By the letter of the code my panel busbars must be rated at least 150A, but the most that hits any one bus is 120A, so doesn't it make sense that I should be able to use a 125A panel?

At a minimum, by the letter of this rule, you shouldn't have to include breakers connected to unrelated phases in this calculation. So there is no reason to require you to include the BC breakers as part of the A-phase calculation. Because BC breakers cannot contribute current to phase A. I see no problem with using a 125A panel, no matter how unwilling your inspector is to understand what is actually happening in the panel.

Even then, the AB and CA breaker ratings do not add up directly, because the voltages and corresponding currents are out of phase. They add up by the sqrt(AB^2 + CA^2 + AB*CA) formula instead, which gives you 104 Amperes.
 
Last edited:

Carultch

Senior Member
Location
Massachusetts
OK, so under the 2014 NEC we can finally have legit AC combiner panels under 705.12(D)(2)(3)(c), which says the bus rating must be greater than or equal to the total of all the breaker ratings in the panel, both supply and load, disregarding the breaker that connects the panel to the service. Groovy.

But... I have an AC combiner that combines three single phase inverters into a three phase service - 60A AB, 30A BC, and 60A CA. By the letter of the code my panel busbars must be rated at least 150A, but the most that hits any one bus is 120A, so doesn't it make sense that I should be able to use a 125A panel?


For a single phase example, suppose you have a collection of 120V inverters and 120V loads. And they are distributed across line 1 and line 2.

Only the line 1 breakers will either source to or draw current from line 1.
Only the line 2 breakers will either source to or draw current from line 2.
Double pole breakers can source/draw to/from both lines, and will do so at the same time for 240V circuits.

This code rule requires you to add up all breakers on each line, whether 1-pole or 2-pole, and confirm that they do not exceed busbar ampacity. Breakers connected to the other line can be ignored, until you are doing calculations for the other line.
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
I also think you can use a 125A panel.

While the letter of this rule lets you out of the 120% rule, it doesn't let you use vector math to reduce the size of the busbar.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
In theory if you combine just the wrong mix of low power factor (some lag, some lead) loads or sources you could get the vectors to line up so that the magnitude added directly.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Thanks for your responses. I had a bit of a DUH moment when I started reading them; the relevant section of code - 705.12(D)(2)(3)(c) - speaks of busbar ratings, not panel ratings. The maximum sum of the ratings of breakers landing on any single busbar in my design is 120A, so clearly I am good with a 125A panel.
 

zman990

Member
Location
US
I feel they need to open this up and allow us to use combiner panel.
then add a subpanel with a load off an existing residential service panel and not worry about the main breaker for added sub panel and do out back feed breaker or point of connection at New sub.
Example: main all in one is rated 200/200 I pull an ac unit out of this panel and land a 150 bus rated mlo load center and put a 100 amp breaker at opposite end of main breaker in main service to protect the 150 amp bus and you now can back feed up to 80 amps of solar into a 200/200 main service.
I only say this because in my jurisdiction this is preferred over line side taps.
I just did a 125amp back feed into a 225 bus with a 125 breaker protecting the bus of new sub panel the main service was rated at 400 amps.
I ran it through a lot of test and also did a load calculation on the the panel and it calculates out to 225 amps.
If I add all breakers it is way over 400 closer to 500 without solar breaker but load calculations don't lie.
So 125+225 is 350 so there is no way it can overload the bus anywhere it should have label saying do not add loads without doing load calculations with solar included or something.
I have used this method for over a year and from what I am told is that it has been done for over 5 yrs now.
So it is obviously safe and I have informed them it is not to code and they say it is safe so ok

Sent from my HTC One_M8 using Tapatalk
 

zman990

Member
Location
US
Also a peak power demand from utility should work as well the derating of breakers is completely unnecessary and load calculations and different labels should be added instead of derate

Sent from my HTC One_M8 using Tapatalk
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
I feel they need to open this up and allow us to use combiner panel....

The 2014 code has the provision that allows this. See 705.12(D)(C)(3). You have to put a warning label on the combiner.

Example: main all in one is rated 200/200 I pull an ac unit out of this panel and land a 150 bus rated mlo load center and put a 100 amp breaker at opposite end of main breaker in main service to protect the 150 amp bus and you now can back feed up to 80 amps of solar into a 200/200 main service.

You can't do this. Under any recent code cycle you are still limited to 40A breaker in the main panel. 2014 does it by 125% of the inverter output but the result is the same.

I just did a 125amp back feed into a 225 bus with a 125 breaker protecting the bus of new sub panel the main service was rated at 400 amps. I ran it through a lot of test and also did a load calculation on the the panel and it calculates out to 225 amps. If I add all breakers it is way over 400 closer to 500 without solar breaker but load calculations don't lie.

If the load calculations on the 400A service are nowhere near 400A then the proper way to do it under the code in this example is to downsize the main breaker at the service to 300A. (Too bad it's 5A short of being okay with a 350A breaker.)

So 125+225 is 350 so there is no way it can overload the bus anywhere it should have label saying do not add loads without doing load calculations with solar included or something.
I have used this method for over a year and from what I am told is that it has been done for over 5 yrs now.
So it is obviously safe and I have informed them it is not to code and they say it is safe so ok

Yes, it's been pointed out that if the main breaker has never tripped, then there should be no fear of ever overloading the busbar. But what if the client adds loads later? It may be safe until that happens, and anyway the code still doesn't allow what you're doing. I can certainly think of some 100A services I've installed on where if the client gets an electric car they'll need a service upgrade, but maybe no one will tell them that, and their busbar could get overloaded with the solar.

Also a peak power demand from utility should work as well the derating of breakers is completely unnecessary and load calculations and different labels should be added instead of derate

That could get complicated. Peak power demand could change, the installation needs to stay okay through any changes. See previous comment.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Correct, thanks.
You're welcome. I am intimately acquainted with this section of the code due to some projects I am presently working on. I had to tell my master electrician that he has to go to an 800A combiner panel because the total of the ratings of the breakers he needs to put in it is 620A. He wasn't happy. If that section of the code allowed us to use 125% of the inverter output instead of the sum of the breaker ratings, a 600A panel probably would have been compliant. Oh, well...
 

zman990

Member
Location
US
Pretty much what I'm trying to say is I know the backfeed breaker and the sub panel mcd need to abide by 120% rule we all know so well but given another example it will still pass inspection and be safe.
200/200 map land 100 mcb in msp then install sub and take out 40amp 2p and land in New sub the has a 200bus 100amp mcb and don't even do a load calculation and back feed 100 amp breaker in New sub essentially you are back feeding 100 amps into the 200/200 it passes all the time no issues.
It seemed crazy to me at first but without ability to line side tap this is what they want.
It makes sense that if solar is produced at max and loads need power they take solar first then grid the only possible way is if the msp some how needed 240 amps of power while solar was on it is not possible main 200 would trip this is a safe install. No derates


Sent from my HTC One_M8 using Tapatalk
 

zman990

Member
Location
US
The code needs to take the mcb protecting sub out for needing to calculate 120%

Sent from my HTC One_M8 using Tapatalk
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
The code needs to take the mcb protecting sub out for needing to calculate 120%

The breaker protecting the sub is used for the 120% rule for the sub busbar, and not for the 120% rule for the main busbar. That's the way it is now and I don't see why it should change.
 

zman990

Member
Location
US
It does count in 2014 except when all breakers don't add up to more than bus obviously they do in this case I am discussing.
I feel it should change to where the main doesn't count

Sent from my HTC One_M8 using Tapatalk
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
I feel it should change to where the main doesn't count

Your going to need more than a feeling to convince anybody.

There's been some talk of changing the rule from 120% to 150% which would make our lives a lot easier, but I don't get the logic of saying that the breaker from the utility side doesn't figure into whether a busbar can be overloaded. Part of the problem is these busbars are not extensively tested for different heat loading situations. For example they are tested for having the full rated current flowing to a load right next to the main feed, they are not tested for having the same amount of current flowing to a different load at the opposite end at the same time.
 

zman990

Member
Location
US
The reason is so that the amperage can be used up before leaving entering into the utility if it can handle full load right next to main why not the opposite end of the buss.


Sent from my HTC One_M8 using Tapatalk
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
The reason is so that the amperage can be used up before leaving entering into the utility if it can handle full load right next to main why not the opposite end of the buss.

Panelboards are not tested for solar backfeeding scenarios. That's what it comes down to.

See this article for a description of how panelboards are tested to UL standards.

So you have a 100A panelboard that is tested for the heat effects of having 100A flowing at one spot. That creates a certain amount of heat that affects the panelboard busbars and the thermal elements in the circuit breakers.

Now let's say you add a 100A solar source and another 100A load at the opposite end. Now you have 100A flowing in two different spots of a 100A busbar. The busbar has not never been tested for having 100A flow in more than one spot. Sure, there's only 100A at any given spot, but you still have twice as much heat being generated from current flow. That heat has to dissipate somewhere and it could likely heat up some spot on the busbar to a higher temperature than the initial load. You have to know something about what happens in that situation before you make changes to the code.

Maybe it would worthwhile for a solar industry associate to pay for some lab testing and present the results to the Code Making Panel. I'm somewhat doubtful that anything short of that will result in changes.
 
Status
Not open for further replies.
Top