What do you think of this proposed work around for load side taps?

[electrimate]

The designers at my company think this is a legal way to get around the 20% rule for load side taps.

The job is an apartment complex with multiple tie-ins at the meter centers. They are doing this in order to use larger inverters.
My opinion is that it is not legal, as the tap is on the load side of the service disconnect.
I have submitted this sketch to the AHJ, but have not heard back from him.

 

[beanland]

705.12(D)(2)

705.12(D)(2)

Since the 100A fused disco is now the overcurrent device feeding the existing 100A apartment panel, the addition of the inverter has not caused a violation. However, the code article applies to "conductor" as well. If you have a 100A and 30A OCPD feeding the same wire, and allowing for the 120% rule, that wire has to be rated for (100 + 30) / 1.2 = 108A. Seems that 115A rated wire meets that requirement. Seems that this complies with this code section.

 

[electrimate]

It is now the overcurrent device, but it is not the Service Disconnect.

I agree that the conductor is not overloaded. But I believe the bus of the panel still could be. 100 amp fuses do not instantly blow at 100 amps. It takes a while for them to heat up enough to finally blow. During the time it takes for them to blow, the buss of the panel could see more than 120% of the 100 amps. An example of this is when an AC compressor goes phase to phase. I've changed many burnt up breakers from that, and have even had to change out load centers because the buss stabs for the AC breaker were burnt up.

 

[don_resqcapt19]

.. During the time it takes for them to blow, the buss of the panel could see more than 120% of the 100 amps. An example of this is when an AC compressor goes phase to phase. ...

How is that any different from when the PV breaker is installed in the panel as permitted by the code rule"?

 

[BillK-AZ]

Since the 100A fused disco is now the overcurrent device feeding the existing 100A apartment panel, the addition of the inverter has not caused a violation.

Not Correct. The following is per NEC 2011, but the NEC 2008 requirements are essentially the same for this requirement.

Art 705.12(D)(2) "Bus or Conductor Rating. The sum of the ampere ratings of overcurrent devices in circuits supplying power to a busbar or conductor shall not exceed 120 percent of the rating of the busbar or conductor."

A 5000W 240V inverter will require a 30A overcurrent device. The limit of the sum of the breakers is 120% of 100A or 120A. The 30A overcurrent device exceeds this if the 100A breaker is not downsized. No amount of intermediate panels can increase this, you must use the overcurrent device that is on the output of the inverter in the calculation. Prior to the 2008 NEC the situation was worse from a PV point of view.

 

[beanland]

Where is the violation?

Where is the violation?

If I have a 100A fuse feeding a 100A panel and 100A wire, no violation.
If I have a 100A breaker and 30A breaker feeding wire rated 130A, no violation.
If I have a 100A breaker and 30A breaker feeding wire rated 110A, no violation because of the 120% rule.
If I have a 100A breaker and 30A breaker feeding wire rated 100A, violation.
The 100A fused disco is the device protecting the 100A panel and 100A wire, not the 100A & 30A devices upstream of that.
It is the device that operates that matters.

 

[electrimate]

I agree with BillK and what he said about breaker sizes made me think of an easy solution. Just change the 100amp breaker to a 90amp and then you don't need the added disconnect, which I don't think fixes the violation.

My only concern is if there will ever be a problem with downsizing to 90amps on a typical one or two bedroom apartment.

 

[electrimate]

Another point that I think the designer missed is that he is on the load side and the Service Disconnect and is feeding the 100amp disconnect with 100 amps and 30 amps, which is more that the allowed 120%. Even though he thinks the load center is protected by the new 100 amp disconnect, the feed to the new 100 amp disconnect is in violation of 705(D)(2).

 

[jaggedben]

I agree with BillK and what he said about breaker sizes made me think of an easy solution. Just change the 100amp breaker to a 90amp and then you don't need the added disconnect, which I don't think fixes the violation.

BillK is almost always right, but in this case he's wrong, or at least jumping to conclusions. Perhaps he assumed that the 100A breaker at the meter is on a 100A rated busbar, but that is not stated and is unlikely to be the case given that this is an apartment complex.

You haven't been very specific about the arrangement of the meters and breakers at the service equipment. But assuming each 100A breaker is indeed the service disconnect, I don't see a problem.

My only concern is if there will ever be a problem with downsizing to 90amps on a typical one or two bedroom apartment.

You'll have to do a load analysis on the given apartment, but I'd suppose problems with this would be unusual.

Another point that I think the designer missed is that he is on the load side and the Service Disconnect and is feeding the 100amp disconnect with 100 amps and 30 amps, which is more that the allowed 120%. Even though he thinks the load center is protected by the new 100 amp disconnect, the feed to the new 100 amp disconnect is in violation of 705(D)(2).

The code simply requires that all the conductors between the two fused discos and the 100A breaker be rated according to the 120% rule. The load center is protected by the 100A disconnect (notwithstanding your concerns about delayed fuse action, the code makers were aware of that when they made the rule). Likewise, each downstream circuit supplied out of the load center is protected by its own OCPD.

 

[tallgirl]

How is that any different from when the PV breaker is installed in the panel as permitted by the code rule"?

Because the bus bars aren't taking the pounding, that's why.

The 120% rule is there to protect bus bars. In this instance, the bus bars are protected by the 100A breaker from having more than 100A dumped backwards into them.

In the drawing above, the unprotected wires (as with the unprotected bus bars -- bus bars HAVE NO OCPDs IN THEM!) is the stub between the Polaris tap and the 100A OCPD to the apartment. The maximum available current is 130A and unless there's a tap rule at work (I don't think so), so that conductor would have to be rated 120% of 130A.

Could be wrong, but that's how I read that.

 

[jaggedben]

The maximum available current is 130A and unless there's a tap rule at work (I don't think so), so that conductor would have to be rated 120% of 130A.

Actually it's the other way around, the conductor would have to be rated at least 83.3% of the 130%, so that 130A is no more than 120% of the rating of the conductor. I'm sure you just mistyped.

 

[Smart $]

Actually it's the other way around, the conductor would have to be rated at least 83.3% of the 130%[A], so that 130A is no more than 120% of the rating of the conductor. I'm sure you just mistyped.

That is correct (with the exception of your typo ).

The sum of OCP ratings cannot be greater than 120% of the bus or conductor rating. Conversely, the bus or conductor rating cannot be less than 83.3% of the sum of OCP ratings.

115A?120%=138A maximum sum of OCP ratings

(100A+30A)?83.3%=108.3A minimum conductor rating​

 

[jaggedben]

Thanks. What's with me and the typos lately?