Subpanel OCPD and Feeder sizing calculations.

[Ctay005]

I'm currently working on a system where we are going to install a small sub-panel beside the MSP in order to make room for the PV breaker. We will be taking 20A and 30A breakers from the Main and hook them up in this sub-panel as well as the 30A of solar and then have a breaker in the Main to protect it all. The plan reviewer said that the Sub-panel Disconnect (OCPD) and wires need to be sized based on the 20A and 30A breakers as well as the 30A of solar. But my thought was that the 30A of solar doesn't need to be included in these calculations because the maximum draw in this panel will be 50A and whatever the Solar isn't powering will be drawn through the grid and if there is no draw then the solar will backfeed as normal to the MSP.

Am I correct in my thinking or do the feeders and OCPD need to be sized based off of 80A (20 + 30 +30 (solar)) like the AHJ said?

 

[jaggedben]

Short answer: You're correct.

The sub feeder needs to be sized for the calculated load, or the PV, whichever is greater, but not both. The subpanel busbar probably needs to be sized for the sub (OCPD+PV)/120%; the 2014 code also gives you a couple other options. Depending on the code cycle you are on, the inspector could also require that same calcuation for the sub feeder (2011 code or earlier). However no code cycle requires a full sum, and in no case would you be required to increase the rating of the sub OCPD beyond 50A.

 

[Smart $]

Short answer: You're correct.

The subfeed needs to be sized for the calculated load, or the PV, whichever is greater, but not both.

Definitely a short answer. However, it is an incorrect short answer.

At the very least, the feeder and bus cannot be less than 120% of main ocpd plus 125% inverter output... and also not less than the calculated load, factored an extra 25% for the continuous portion.

Depending on POC arrangement, it may need to be sized larger [705.12(D)(2)].

 

[Ctay005]

Definitely a short answer. However, it is an incorrect short answer.

At the very least, the feeder and bus cannot be less than 120% of main ocpd plus 125% inverter output... and also not less than the calculated load, factored an extra 25% for the continuous portion.

Depending on POC arrangement, it may need to be sized larger [705.12(D)(2)].

So based on this the feeders need to be sized to handle:
21A (Inverter output) x 1.25 = 26.25 + 50A (OCPD) = 76.25
76.25 x 1.20 = 91.5A

Correct?

 

[Smart $]

So based on this the feeders need to be sized to handle:
21A (Inverter output) x 1.25 = 26.25 + 50A (OCPD) = 76.25
76.25 x 1.20 = 91.5A

Correct?

76.25 ÷ 1.20 = 63.5A minimum...

So need at least a 100A panel because 60A panel is too small.

Under 2014 Code, that's only if PV is at opposite end of panel bus from main. Put the PV breaker anywhere else, the math is different, resulting in a greater minimum feeder size.

 

[Ctay005]

76.25 ÷ 1.20 = 63.5A minimum...

So need at least a 100A panel because 60A panel is too small.

Under 2014 Code, that's only if PV is at opposite end of panel bus from main. Put the PV breaker anywhere else, the math is different, resulting in a greater minimum feeder size.

63.5A... I knew my math seemed off. Okay that makes sense now. I should be good to go with a 100A panel fed by #6 cu feeders.

 

[Smart $]

63.5A... I knew my math seemed off. Okay that makes sense now. I should be good to go with a 100A panel fed by #6 cu feeders.

Beware of using a panel with 100A main breaker. Screws up the math, even if supplied with 50A/63.5A feeder. Better and less costly to use an MLO panel.

 

[iwire]

63.5A... I knew my math seemed off. Okay that makes sense now. I should be good to go with a 100A panel fed by #6 cu feeders.

How are you using 6 AWG for this?

 

[jaggedben]

Definitely a short answer. However, it is an incorrect short answer.

At the very least, the feeder and bus cannot be less than 120% of main ocpd plus 125% inverter output... and also not less than the calculated load, factored an extra 25% for the continuous portion.

Depending on POC arrangement, it may need to be sized larger [705.12(D)(2)].

Nothing wrong with my answer. In the 2014 code the 120% rule applies only to panelboard busbars. In previous codes it applies to both busbar and conductor. I said that.

 

[jaggedben]

How are you using 6 AWG for this?

Looks to me like 6awg is rated 65A at 75C, so unless he's using 60C rated wire he should be fine, under any code cycle. If on 2014 the wire could be 8awg if it's 75C rated.

Edit: since he's relocating circuits he might have a deration issue if he runs them all in the same raceway.

 

[Smart $]

Nothing wrong with my answer. In the 2014 code the 120% rule applies only to panelboard busbars. In previous codes it applies to both busbar and conductor. I said that.

Yeah... the ampacity required for other than using the 120% rule is the sum of 125% inverter output plus rating of ocpd protecting feeder. In OP's case, that'd be the 76.25A he calculated.

 

[Smart $]

63.5A... I knew my math seemed off. Okay that makes sense now. I should be good to go with a 100A panel fed by #6 cu feeders.

Beware of using a panel with 100A main breaker. Screws up the math, even if supplied with 50A/63.5A feeder. Better and less costly to use an MLO panel.

63.5A figure is good for busbar under 2014 Code... but as jaggedben brought up, the 120% rule no longer applies to the feeder. The thing is, Code only provides a rule for when inverter connection is NOT at opposite end, and no rule for when it is at opposite end. Safe bet is to use the 76.25A value unless we can get a consensus that a lesser value is permitted.

 

[jaggedben]

I hope we're all agreed that the sub does not require an 80A OCPD. That seemed to the main issue, about which the plan checker's interpretation, as stated to us, is wrong. Sub is fine with a 50A OCPD, (or perhaps even less if calculated load allows.) From there various math calculations follow, as have been covered, depending on component and code cycle.

 

[Ctay005]

I hope we're all agreed that the sub does not require an 80A OCPD. That seemed to the main issue, about which the plan checker's interpretation, as stated to us, is wrong. Sub is fine with a 50A OCPD, (or perhaps even less if calculated load allows.) From there various math calculations follow, as have been covered, depending on component and code cycle.

Thats what I was trying to get approval on and get my math correct to show the plan checker. Since we are running wires together derating will be need to be considered so #4 AWG should be more than enough to satisfy both the code and this picky plan checker.

Thanks