Load side feeder tap NEC 705.12(D)(2)(2) and 240.21(B)(2)

[sarahleighty]

I am trying to determine the required ampacity of a feeder tap conductor for a load side tap (assuming it is allowed by AHJ) to interconnect a 4200W (22A max current) to an existing service conductor directly downstream from a 100A main breaker. (service conductor ampacity unknown). I am seeing this done in the field and local AHJs are allowing for system POC/commissioning. I am trying to assess if it is acceptable based on a thorough look at NEC 705.12(D)(2)(2). When installers are doing this they are illustrating in design plans a line side tap incorrectly, and minimum wire size is often a #8.

My interpretation after looking at 705.12(D)(2)(2) and 240.21(B)(2) leaves me to believe that the ampacity of the conductors running from the load side feeder tap to the 25A fuses in the PV disconnect (assuming a length of not over 25 feet) must be #2AWG: 100A + (22A x .33) = 108A minimum ampacity (#2AWG)

Can anyone confirm or give me a professional opinion? Is the #8 Ok as long as the existing service conductors can handle 100A +22A PV? Or are the #8 conductors (minimum allowed ampacity for supply side interconnection) in danger? Is #2 AWG correct?

 

[ggunn]

I am trying to determine the required ampacity of a feeder tap conductor for a load side tap (assuming it is allowed by AHJ) to interconnect a 4200W (22A max current) to an existing service conductor directly downstream from a 100A main breaker. (service conductor ampacity unknown). I am seeing this done in the field and local AHJs are allowing for system POC/commissioning. I am trying to assess if it is acceptable based on a thorough look at NEC 705.12(D)(2)(2). When installers are doing this they are illustrating in design plans a line side tap incorrectly, and minimum wire size is often a #8.

My interpretation after looking at 705.12(D)(2)(2) and 240.21(B)(2) leaves me to believe that the ampacity of the conductors running from the load side feeder tap to the 25A fuses in the PV disconnect (assuming a length of not over 25 feet) must be #2AWG: 100A + (22A x .33) = 108A minimum ampacity (#2AWG)

Can anyone confirm or give me a professional opinion? Is the #8 Ok as long as the existing service conductors can handle 100A +22A PV? Or are the #8 conductors (minimum allowed ampacity for supply side interconnection) in danger? Is #2 AWG correct?

You also have to look at what is downstream from the tap. It can get the rating of the service disco plus 125% of your inverter current and may need OCP.

 

[jaggedben]

In my opinion it's not possible to take the literal meaning of the language 705.12(D)(2)(2) too seriously. For example, the language could be read as saying that 240.21 calculates the rating of a feeder OCPD, but that's not true.

My opinion is that they intended that you take the feeder OCPD, and then add 125% of the inverter output, and then pretend that sum is the rating of an OCPD that is protecting the feeder for 240.21(B) calculations. In your example, that would be (100+(22*1.25))*.33 for a 25ft tap. Which would be 42 amps and be fine at 8awg. A ten foot tap could use 12awg.
And that's surely just as safe as any other 240.21(B) tap, because the inverter can barely contribute any meaningful additional fault current.

There is no minimum size if we're talking load side.

 

[sarahleighty]

In my opinion it's not possible to take the literal meaning of the language 705.12(D)(2)(2) too seriously. For example, the language could be read as saying that 240.21 calculates the rating of a feeder OCPD, but that's not true.

My opinion is that they intended that you take the feeder OCPD, and then add 125% of the inverter output, and then pretend that sum is the rating of an OCPD that is protecting the feeder for 240.21(B) calculations. In your example, that would be (100+(22*1.25))*.33 for a 25ft tap. Which would be 42 amps and be fine at 8awg. A ten foot tap could use 12awg.
And that's surely just as safe as any other 240.21(B) tap, because the inverter can barely contribute any meaningful additional fault current.

There is no minimum size if we're talking load side.

Thanks. Yes- agreed on the bit about interpreting 705.12 (and some other aspects of the code) and confusion in language. My concern is fault current from the utility on the tapped conductors, which would be potentially present with a line side tap or load side tap. Seems that if #8 is required minimum for line side tap, that it would be the same for a load side tap.

We size PV breakers to be at a minimum a KAIC rating to match the primary service breaker, assuming it to be sized to accommodate the potential fault current from the transformer. We are required to have a minimum 60A enclosure for the supply side connection breaker / fuse, and a minimum #8 AWG. Load side taps also tap into the service conductors and seems they would be help to the same code standards, IF they are allowed. Perhaps that would make things simpler here.

I understand a 10 ft tap uses the 1/10 allowance so circuit calcs would be minimal- "240.21(B)(1) ...not less than one-tenth of the rating of the OCPD protecting the feeder conductors"- that doesn't make sense using the 25A fuses, so assuming they meant 100A main breaker- we have 10A, still doesn't make sense for tap conductor sizing.

240.21(B)(1) also says the ampacity must be not less than the "equipment containing the overcurrent device ... or not less than the rating of the OCPD .... So- rating of the OCPD protecting it is 25A but 30A bladed disco houses circuit fuses, so I guess we get to choose 25A or 30A. But we couldn't use a #12 for this circuit because of the restriction on that wire and OCPD allowance protecting it per 240.4(D)- #12 can only be installed with OCPD of 20A or smaller, so we would have to go to a #10 minimum.

So.... IF I can see that my feeder circuit is less than 10 feet, #10 is Ok in this case???

 

[jaggedben]

...My concern is fault current from the utility on the tapped conductors, which would be potentially present with a line side tap or load side tap. Seems that if #8 is required minimum for line side tap, that it would be the same for a load side tap.

I don't think that really follows. On a load side tap there is an OCPD head that is governed by the NEC. Not so on a line side tap.

... Load side taps also tap into the service conductors ....

No, by definition they do not. Service conductors end at the service disconnecting means and OCPD.

I understand a 10 ft tap uses the 1/10 allowance so circuit calcs would be minimal- "240.21(B)(1) ...not less than one-tenth of the rating of the OCPD protecting the feeder conductors"- that doesn't make sense using the 25A fuses, so assuming they meant 100A main breaker- we have 10A, still doesn't make sense for tap conductor sizing.

The red part is correct.

Most of the time your minimum size required for the inverter output (705.60) will be above the minimum size required by 240.21(B). As far as the tap rules making sense, all I can say is they have been that way, more or less the way they are, for quite a bit of time. The way 705 applies them to an inverter is where it gets squirly. What doesn't make sense, to me at least, is requiring the tap conductors to the inverter to be much bigger than on a load tap; the inverter output can typically provide only a tiny fraction of the fault current that the utility can, so safety wise it's not significantly different. Often 10% of (feeder OCPD+125% of inverter output) is less than the inverter output. If so, there's no sensible reason that the tap conductors should be bigger than that required of the inverter output. Hopefully your AHJ reads the language the same way. They added something about 705.60 in the 2017 code.

So.... IF I can see that my feeder circuit is less than 10 feet, #10 is Ok in this case???

Yes. In my opinion.

 

[sarahleighty]

I don't think that really follows. On a load side tap there is an OCPD head that is governed by the NEC. Not so on a line side tap.



No, by definition they do not. Service conductors end at the service disconnecting means and OCPD.



The red part is correct.

Most of the time your minimum size required for the inverter output (705.60) will be above the minimum size required by 240.21(B). As far as the tap rules making sense, all I can say is they have been that way, more or less the way they are, for quite a bit of time. The way 705 applies them to an inverter is where it gets squirly. What doesn't make sense, to me at least, is requiring the tap conductors to the inverter to be much bigger than on a load tap; the inverter output can typically provide only a tiny fraction of the fault current that the utility can, so safety wise it's not significantly different. Often 10% of (feeder OCPD+125% of inverter output) is less than the inverter output. If so, there's no sensible reason that the tap conductors should be bigger than that required of the inverter output. Hopefully your AHJ reads the language the same way. They added something about 705.60 in the 2017 code.



Yes. In my opinion.

Got it. I was thinking too hard about conductor tapping, indeed- the tap DOES occur downstream from the main breaker, not into the service conductor, and so it is protected by the main breaker from any fault on the utility side.
Thanks to everyone who helped me clear the waters on this!
With respect- many thanks.