Centerfeed busbar and 120% rule

[SunFish]

I am designing a job and the customer has a combined meter/main service panel. The 200 A main breaker feeds the CENTER of the busbar instead of the top as one normally sees.

Can I still use the 120% rule for backfed breakers with a centerfed busbar? We are on the 2014 NEC.

 

[Smart $]

2014

705.12(D)(2)(3)...

(d) Connections shall be permitted on multiple-ampacity
busbars or center-fed panelboards where designed under engineering
supervision that includes fault studies and busbar load
calculations.

 

[jaggedben]

You can under 2014 if you get an engineer to approve it with fault studies and loading calculations. See the new 705.12(D)(2)(3)(d).

 

[jaggedben]

Dang it, Smart$ beat me while I was getting out the book! :lol:

 

[Zee]

Dang it, Smart$ beat me while I was getting out the book! :lol:

If he reduces the main breaker size to 175A can he add 25A of PV?

 

[GoldDigger]

If he reduces the main breaker size to 175A can he add 25A of PV?

Yes, he can do that without requiring engineering support, since it would only total 100% of the bus rating. But the load calculation would have to support reducing to 175.

However, he should also check whether the center fed bus is actually rated for 125A. Not as likely to be the case in a center fed panel as in relatively recent end fed panel, but worth checking out just the same.

 

[jaggedben]

...However, he should also check whether the center fed bus is actually rated for 125A. Not as likely to be the case in a center fed panel as in relatively recent end fed panel, but worth checking out just the same.

He mentioned a 200A MCB. 100A MCB in a 125A is common and gives some leeway. 200A MCB in a 225A panel is rare, I have never seen one we didn't install.

 

[Carultch]

I am designing a job and the customer has a combined meter/main service panel. The 200 A main breaker feeds the CENTER of the busbar instead of the top as one normally sees.

Can I still use the 120% rule for backfed breakers with a centerfed busbar? We are on the 2014 NEC.

A pessimist would say there is no opposite end. An optimist would say there are two opposite ends.

The NEC2014 allows you to interconnect in several of these these exotic situations, as long as the safety is assessed under engineering supervision. The new rule simply says that the NEC cannot cover every possible situation in a nice "cookie cutter rule". Rather than having it be an automatic show-stopper, it allows you to introduce a licensed engineer to the project, and find a solution that works.

My opinion would be that as long as you only use one half of the center fed panelboard, and treat the other half as if all the breakers were locked in the OFF position, then you should have a working solution. As long as loads don't contribute significant fault current, the loads on the other half of the panelboard, when activated, will only reduce current on the interconnected side.

 

[cowboyjwc]

I don't see where it says that you are allowed to use the 120% rule for a center fed.

I've been reading it as you may derate the main breaker if you get load calcs and a fault study.

Now just to clarify, we are not on the 2014 so I only know what I read in the different articles.

 

[jaggedben]

...

My opinion would be that as long as you only use one half of the center fed panelboard, and treat the other half as if all the breakers were locked in the OFF position, then you should have a working solution. As long as loads don't contribute significant fault current, the loads on the other half of the panelboard, when activated, will only reduce current on the interconnected side.

I agree with the rest of your comments, but I think you have this part exactly backwards. It's precisely the loads on the opposite side of the main that create a potential problem. If these loads add up to more than the main breaker, then they could draw additional current from the inverter(s). That would mean more current passing through the busbar at the location of the main than the busbar is rated for. The opposite end rule was put in precisely to prevent this sort of situation. The more common situation would be where the inverter output might be put in the middle of the busbar, but the principle applies equally to the main.

A rule that would make sense is for the loads on the opposite end of the busbar to not add up to more than the rating of the busbar. But that's not in any version of the code. The 2014 code goes sort of halfway there by allowing interconnection on a centerfed busbar if all the loads plus inverter(s) do not exceed the rating of the busbar. But that's going to be rarely encountered, at least in residential situations.

 

[Carultch]

I agree with the rest of your comments, but I think you have this part exactly backwards. It's precisely the loads on the opposite side of the main that create a potential problem. If these loads add up to more than the main breaker, then they could draw additional current from the inverter(s). That would mean more current passing through the busbar at the location of the main than the busbar is rated for. The opposite end rule was put in precisely to prevent this sort of situation. The more common situation would be where the inverter output might be put in the middle of the busbar, but the principle applies equally to the main.

A rule that would make sense is for the loads on the opposite end of the busbar to not add up to more than the rating of the busbar. But that's not in any version of the code. The 2014 code goes sort of halfway there by allowing interconnection on a centerfed busbar if all the loads plus inverter(s) do not exceed the rating of the busbar. But that's going to be rarely encountered, at least in residential situations.

So as an example to what you are saying:

Suppose we have 80A of Solar to interconnect at a centerfed 400A panelboard with a 400A main. You are telling me about a situation where the upper half of the panelboard attempts to consume 480A of current, and the lower half where the solar is interconnected isn't consuming anything. In this case, I see exactly what you are saying. That solar plus main supply can exceed the busbar ampacity in the upper half of the busbar. Point taken. Nicely done.

Maybe my opinion could still be true, as long as the sum of the breakers on the opposite half of the busbar doesn't exceed the busbar ampacity.

Such that on this 400A panelboard, I could have 4 qty 100A all-phase load breakers on the upper half, 6 qty 100A all-phase load breakers on the lower half, and still interconnect 80A of solar on the lower half.

 

[Smart $]

...Suppose we have 80A of Solar to interconnect at a centerfed 400A panelboard with a 400A main. ....

I think the advocates of these restrictions are overly concerned.

Let's look at a scenario for a standard service which overshadows this scenario, yet we don't see anyone jumping through hoops to make changes...

Let's say 1200A service. Splits to six (6) 400A panelboards and appropriately sized service entrance conductors. A green electrician says, "Wow! How are the 1200A service conductors protected with 2400A of service disconnecting means?" Code 'says' they are protected by the calculated load not being greater than 1200A. It's been done forever without a significant number of faults.

Using the same logic we should be able to supply a 200A distribution panel with 200A of utility and 200A of PV... as long as the load is not greater than 200A. :blink:

 

[Carultch]

Using the same logic we should be able to supply a 200A distribution panel with 200A of utility and 200A of PV... as long as the load is not greater than 200A. :blink:

I would think the same thing, if it weren't for 705.12(D).

I can't think of how one could construct Kirchoff's laws, such that an ordinary panelboard is fed by a breaker limited 200A from a source on both opposite ends, and end up with any part of it exceeding 200A. The current will start at 200A on one end, diminish to zero somewhere in the center, and then return to 200A on the opposite end. And that works if there is 400A of total load, which will be very unlikely.

 

[Smart $]

I would think the same thing, if it weren't for 705.12(D).

I can't think of how one could construct Kirchoff's laws, such that an ordinary panelboard is fed by a breaker limited 200A from a source on both opposite ends, and end up with any part of it exceeding 200A. The current will start at 200A on one end, diminish to zero somewhere in the center, and then return to 200A on the opposite end. And that works if there is 400A of total load, which will be very unlikely.

Put both sources at the same end. A max 200A of load would limit bus current to a max of 200A under nominal conditions. I understand the 200A can be exceeded on an other than nominal condition, but so can the service example I mentioned.

 

[jaggedben]

I would think the same thing, if it weren't for 705.12(D).

I can't think of how one could construct Kirchoff's laws, such that an ordinary panelboard is fed by a breaker limited 200A from a source on both opposite ends, and end up with any part of it exceeding 200A. The current will start at 200A on one end, diminish to zero somewhere in the center, and then return to 200A on the opposite end. And that works if there is 400A of total load, which will be very unlikely.

John Wiles has stated that one concern is that thermal transfer in the setup you describe has not been studied and is not part of UL's tests. I feel this may be an unnecessary concern but since I'm not an engineer I can't really argue about it. But one route to allowing such a setup could be a new product standard for panelboards to be tested to.

Put both sources at the same end. A max 200A of load would limit bus current to a max of 200A under nominal conditions. I understand the 200A can be exceeded on an other than nominal condition, but so can the service example I mentioned.

I completely understand your point, but I feel the distinction between service conductors and loadside busbars isn't entirely arbitrary. Joe-Homeowner who knows nothing about code and calculated load is a lot more likely to put in a spa circuit that overloads his service breaker than to redo his service. And he's far less likely to have a service with six panelboards, and a facility with such a service is far more likely to have work done to code. What it comes down to is that on any (compliant) existing service where the main OCPD hasn't been tripping it ought to be safe to add up to 100% of the rating in interactive sources, as long as the owner will never add loads. How you write that last part into the code is the problem.

 

[Smart $]

John Wiles has stated that one concern is that thermal transfer in the setup you describe has not been studied and is not part of UL's tests. I feel this may be an unnecessary concern but since I'm not an engineer I can't really argue about it. But one route to allowing such a setup could be a new product standard for panelboards to be tested to.



I completely understand your point, but I feel the distinction between service conductors and loadside busbars isn't entirely arbitrary. Joe-Homeowner who knows nothing about code and calculated load is a lot more likely to put in a spa circuit that overloads his service breaker than to redo his service. And he's far less likely to have a service with six panelboards, and a facility with such a service is far more likely to have work done to code. What it comes down to is that on any (compliant) existing service where the main OCPD hasn't been tripping it ought to be safe to add up to 100% of the rating in interactive sources, as long as the owner will never add loads. How you write that last part into the code is the problem.

That's not the problem. As you noted, Joe Homeowner don't know the Code and is likely to overload the panel without doing a Code compliant determination.

If I were to write the Code, I'd make it something simple, like a permanent plaque stating to the effect...

WARNING:
THIS EQUIPMENT FED BY MULTIPLE SOURCES.
SUM OF LOAD BREAKER RATINGS SHALL NOT
EXCEED _____A​

...where the blank is the maximum permitted under Code (likely the bus rating).

Would have to back it up with actual Code, and include requirement for source breakers to be permanently labeled

 

[jaggedben]

If I were to write the Code, I'd make it something simple, like a permanent plaque stating to the effect...

WARNING:
THIS EQUIPMENT FED BY MULTIPLE SOURCES.
SUM OF LOAD BREAKER RATINGS SHALL NOT
EXCEED _____A​

...where the blank is the maximum permitted under Code (likely the bus rating).

Would have to back it up with actual Code, and include requirement for source breakers to be permanently labeled

That would be a disaster. :happysad:

First, I'd say in at least 90% of the service panels I've looked at the sum of the branch breakers exceeds both the rating of the panel and the main OCPD. So you'd be killing the solar industry. We'd hardly be able to sell a single residential system without a service upgrade.

Second, there is no place in the code, AFAIK, which supplies the number to fill in your blank. The present approach of the code is simply to protect a panel board from the utility side with an OCPD that does not exceed the panelboard's rating. Once you have done that you can put in as many branch breakers as will fit. And in the many cases where electricians put in twenty 20A breakers for twenty different loads that are probably each 1-5A, you easily end up with breakers that add up to more than the panelboard rating even though the calculated load is a fraction of it.

So, such an approach is not going to work.

 

[Smart $]

That would be a disaster. :happysad:

First, I'd say in at least 90% of the service panels I've looked at the sum of the branch breakers exceeds both the rating of the panel and the main OCPD. So you'd be killing the solar industry. We'd hardly be able to sell a single residential system without a service upgrade.

Second, there is no place in the code, AFAIK, which supplies the number to fill in your blank. The present approach of the code is simply to protect a panel board from the utility side with an OCPD that does not exceed the panelboard's rating. Once you have done that you can put in as many branch breakers as will fit. And in the many cases where electricians put in twenty 20A breakers for twenty different loads that are probably each 1-5A, you easily end up with breakers that add up to more than the panelboard rating even though the calculated load is a fraction of it.

So, such an approach is not going to work.

That's the problem with initiating something like what I suggest. Too many people are rooted in conventional wisdom to date. Backtrack to a mutli-disconnect service panel with line-side PV system breaker... say 400 MLO with one 200A PV and 2?200A load breakers. Is there really any difference to that and having a 400A MCB panel with one 200A PV and 2? 200A load breakers, or 8?50A load breakers or 20?20A load breakers.

The idea is that the calculated load does not exceed the panel rating as the protection. Anyone that utilizes my suggestion would be less likely to "under-load" their circuits. What I suggested was just to get a push going in the right direction. Where the sum of load breaker ratings exceed bus rating, you can still have your 120% rule... no plaque required (i.e. the one I suggested). The precursor is already in the 2014 edition of 705.12(D)(2)(3)(c) only it is 'worse' because it includes the PV breaker in the sum.

 

[ggunn]

The idea is that the calculated load does not exceed the panel rating as the protection.

But as long as all the breakers are load breakers, there's not a problem with overloading a panel. If the load exceeds the rating, the main breaker will shut it down.

 

[jaggedben]

...

The idea is that the calculated load does not exceed the panel rating as the protection. ...

Then at least make your proposed change and warning label based on the calculated load and not the sum of the breakers.

It would be no piece of cake for a solar installer to come into an existing building, survey all the existing circuits and combine underloaded ones, relabel, respond to the customer when a mistake is made and a breaker starts nuisance tripping, etc. etc. You are more or less doubling the electrician labor, and that is basically our profit and ability to stay in business that gets wiped away right there. Forget whether it should have been done one way or another to begin with, you have to consider the (unnecessary) cost of retrofitting every house you might touch.

In any case, I'd much rather have the rules we have now, which make 90% of situations simple to analyze, than to completely turn things upside down and create more work on every job (load survey and calculations) just to make those last 10% possible on centerfed busbars or whatever else. And by the way, the 2014 code offers your solution as one option of several. No need for it to be the only option.

Too many people are rooted in conventional wisdom to date.

It's not necessarily a problem that people are rooted in conventional wisdom. It's a serious problem that everybody's houses are.