2022 NEC 220.87 Exception

[Misheru]

Can I use this calculation for determining existing loads if I have solar PVs but it is connected on the load side of the service. Per the exception "If the feeder or service has any renewable energy system (i.e., solar photovoltaic systems or wind electric systems) or employs any form of peak load shaving, this calculation method shall not be permitted."

Please shed some light

 

[jaggedben]

The answer is no, you cannot use it, that's what it says. (Although I think if you have data that allows you to calculate the complete consumption from all source you should be allowed to use it.)

 

[Elect117]

220.87 basically lets you use metering / billing information.

Every utility's credit program is a little different, and can be different between metering schedules. The information they give you might not give you the correct load.

For example, lets say they give you one years worth of metering information on 2 meters but you have 4 meter numbers. 1 for the demand load, 1 for the generation kwh, one for the demand kwh, and another for load vars.

How are we going to calculate the highest peak demand for a years worth of data? The demand kw will be constantly offset by the generation. So you would have to figure out the generation kw, but they don't have that metered. They only meter the kwh.

So if you were to try to estimate the generation kwh by dividing by time, you won't get a coinciding kw value for the demand load kw. The peak demand could be on the 3rd Monday of the month and you won't have the solar output demand.

Arguably, the solar system typically records metrics, but not always to the same precision as utility metering. And then you still have a problem trying to match up days and times down to the same window the meter captured the peak demand value.

Acrobatics can get you close, but never perfect and I think this exception exists for that reason.

Some utilities might meter a solar demand kw and capture at the same cycle the load demand.

There is also the issue of the solar outputting more KW than the load. That probably doesn't apply to this case.

 

[jaggedben]

220.87 does not specify the demand data has to be provided by a utility. But an AHJ could reasonably expect any data provided by others to be collected by equipment that meets the same standards for accuracy. That said, there have been other discussions here about how granular the data has to be in order to be used for 220.87, and I can't say there was any consensus or even much specific knowledge about how utilities collect demand data. Apparently some use 15min intervals, others 5min, and others a rolling 15min interval (i.e. the highest average of three consecutive 5min intervals). So I think if you have data that's about as accurate and precise it's up to your AHJ.

All that said, the exception is rather absolute, and doesn't allow for the possibility that your data might come from a source that includes metering the generation and accounting for that to calculate the total demand. So no matter how good your data is, the presence of solar *in the data* would prohibit using the 220.87 method. But it matters exactly what is measured, not just the presence of solar. For example if a service had two disconnects, one for solar and one for loads, a meter device on the loads feeder could collect eligible data. Because the exception refers to the service or feeder, and in that case the feeder would have no generation.

 

[Elect117]

For example if a service had two disconnects, one for solar and one for loads, a meter device on the loads feeder could collect eligible data. Because the exception refers to the service or feeder, and in that case the feeder would have no generation.

This would depend on the meter's physical location in the circuit. If I am understanding it correctly, then yes it would give you good information, but then the utility bill might not be lowered by the generation since the meter will see both as a single incoming feed. Unless they have changed the meters to have more jaw sockets so they can use more than one incoming line and only 1 outgoing line.

 

[jaggedben]

This would depend on the meter's physical location in the circuit.

Yes.

If I am understanding it correctly, then yes it would give you good information, but then the utility bill might not be lowered by the generation since the meter will see both as a single incoming feed. ...

My point is that the customer could install their own CT metering on the load feeder and solar, separately from what the utility measures. This is actually not that uncommon for solar installations. Again, 220.87 does not absolutely say the data has to come from a utility.

 

[Elect117]

My point is that the customer could install their own CT metering on the load feeder and solar, separately from what the utility measures. This is actually not that uncommon for solar installations. Again, 220.87 does not absolutely say the data has to come from a utility.

Ahhhh, ya. I agree.

We actually use something like that on car charging branch circuits for the City, for the EV credits, rather than install smart charger which are very expensive.