Problems interpreting Utility API information for Max Demand

[Solar Noob]

Hello again, after posting my last question involving the best way to calculate load draw from existing loads, I have hit another brick wall when it comes to correctly calculating the load draw of the panel schedule. Adhering to Article 220.87, I have pulled the Utility API information, however the problem now seems to be that the information they provide does not allow me to calculate the actual kW demand. The only information they seem to give is in kWh, and this seems to only be a problem with residential accounts, as the commercial accounts always include max kW demand over the course of a year.

My question is when working with residential accounts, is there an error in the way I ask for the Utility API, or is it just by nature of the information usually provided to residents that doesn't usually cover the max kW demand like commercial accounts? If this is the case, is the only real option to set a recording ammeter or power meter for 30 days?

 

[electrofelon]

Hello again, after posting my last question involving the best way to calculate load draw from existing loads, I have hit another brick wall when it comes to correctly calculating the load draw of the panel schedule. Adhering to Article 220.87, I have pulled the Utility API information, however the problem now seems to be that the information they provide does not allow me to calculate the actual kW demand. The only information they seem to give is in kWh, and this seems to only be a problem with residential accounts, as the commercial accounts always include max kW demand over the course of a year.

My question is when working with residential accounts, is there an error in the way I ask for the Utility API, or is it just by nature of the information usually provided to residents that doesn't usually cover the max kW demand like commercial accounts? If this is the case, is the only real option to set a recording ammeter or power meter for 30 days?

Many residential meters do not record demand.

 

[Solar Noob]

If that is the case, is the only real option to calculate the demand fully is to set the recording ammeter or power meter? I know there are load calculators that use articles 220.83 to calculate existing loads, but I am assuming these are not valid ways to calculate load demands for existing service panels.

 

[GeorgeB]

Even with commercial/industrial accounts, the "max" is usually a 15 minute average. Although almost impossible to really occur, there theoretically could be usage at 3 times that rate for 5 minutes and no usage for 10 minutes.

From a SOLAR SIZING criteria, those shouldn't be of any significance. For residential, the local tariff will often limit your annual production to some percentage of consumption, that is they won't allow full (or any) credit for generation exceeding consumption on some basis, here in Duke Energy territory with my net-metering tariff, annual. For commercial/industrial, some % of previous year's consumption negotiated with your POCO would work.

If you are looking at using solar charged batteries to power the property, the real maximum consumption would be valuable info; only a recording meter of some sort would accomplish that, and considering heat and air conditioning as commonly the highest loads, 1 month wouldn't be a lot of good. If it is 3 in the afternoon on a 100F day and the home has 2 clothes dryers and a Tesla, odds are 99+% that you won't be able to really know the maximum.

So ... what are your goals?

 

[Solar Noob]

It would vary from job to job, but being able to see if derating the main breaker would be an option for jobs that don't meet the 120% rule for standard panels (200A service panels with 200A main breakers) is the main goal.

 

[electrofelon]

It would vary from job to job, but being able to see if derating the main breaker would be an option for jobs that don't meet the 120% rule for standard panels (200A service panels with 200A main breakers) is the main goal.

Yeah if you can't get demand data from the utility, it's either put a logger on there yourself or do a NEC calculation ( which will often be way way high not always useful).

 

[pv_n00b]

For a dwelling, you can go by 220.83 or use 220.87 but that requires either one year or 30 days of recorded demand. That's for adding new loads to an existing panel. You have not said what you need this for and you might be able to get away with something else.
And I see you are trying to use a variation of my user name. I'll be keeping an eye on you.

 

[jaggedben]

If you have 15min kwh data then arguably you can use it. Just multiply by 4 to get kW. Your AHJ may need convincing but it's legit in my opinion.

Read this thread:

220.87

The utilities in this area can provide fifteen minute usage data for the previous year. Can you use the highest 15 minute kWH data for determining the maximum load on the service when you are adding additional loads such as an EV?

forums.mikeholt.com

And/or this one:

Utility kWH vs kW

Hello I just need a quick sanity check on this. When looking at an existing building sometimes we are lucky enough to get utility bill which include both the kWH and kW for the service. As far as I know, the peak kW*1.25 is what in calculations for the existing loads per 220.87. Is there...

forums.mikeholt.com

 

[wwhitney]

If you have 15min kwh data

The OP says "problem with residential accounts," so do you know if any of the CA POCOs provide 15 minute data on residential accounts? IIRC I can only get 1 hr data on my account.

Thanks,
Wayne

 

[jaggedben]

The OP says "problem with residential accounts," so do you know if any of the CA POCOs provide 15 minute data on residential accounts? IIRC I can only get 1 hr data on my account.

Thanks,
Wayne

I only know about PG&E, and yes it's usually hourly, but I've seen a couple exceptions. So don't assume.

Also in one of those other threads I talked about multiplying the hourly kwh consumption by four to calculate a (way way way too conservative) worst case assumption that the max hourly kwh was all consumed in a 15min period. For many not-yet-electrified or even partially electrified single family homes the max hourly kwh in the data is under 6kw, i.e the calc comes out to 100A or less. So if the home already has a 200A service, it really proves you can add more load, or downsize a main breaker. Whether you can convince an AHJ this meets the letter of 220.87 I don't know, haven't had to deal with that discussion.

 

[jaggedben]

Yeah if you can't get demand data from the utility, it's either put a logger on there yourself or do a NEC calculation ( which will often be way way high not always useful).

The NEC calc will almost always be much higher than real demand but for a fully electrified 2000sq ft single family home it will still often not exceed 150A under the optional methods. For a non electrified home it will likely be under 100A. At least that's so in California, where our OP is. The optional methods are really not that hard to do and perfectly useful for demonstrating compliance in most situations.

 

[RumRunner]

I’m confused as if being left punch-drunk like a barfly looking for the nearest restroom.

You must be referring to programming protocol when you say API.

Seriously speaking:

The app you are trying to access needs a “key” .
API is a protocol in computer programming that enables one application to communicate with other apps in other platforms.

API , (Application Programming Interface) is a protocol that programmers used to allow them to write programs to interoperate with other computers i.e. Apple to IBM to Microsoft, et al.

In your case, you are acting as a client to access the server that holds the relevant data that you need for the demand calculation.

In some organization, some data are protected or in their discretion the data have very little use for consumer, the client. They are essentially the server.

During the writing of the program at this stage-- or after the program is considered usable and ready to be deployed--you as valued client can be given the access “key” (authorization) and you will have the rights and privileges to these data.

In the absence of these given rights . . . the app is off limits. This is the reason you can't get what you want.

 

[pv_n00b]

I’m confused as if being left punch-drunk like a barfly looking for the nearest restroom.

Utility API is a bit of a misnomer. They provide two products. an API that someone can use in an application to access utility customer data and they have a website where you can set up an account and get utility information on a customer, with the customer's permission. The OP is most using the website interface or an app that incorporated the API, not the API.

 

[zbang]

Drop 'API' from the conversation- it's a method to get the data, not the data itself; that will remove some of the confusion.

 

[pv_n00b]

I only know about PG&E, and yes it's usually hourly, but I've seen a couple exceptions. So don't assume.

Also in one of those other threads I talked about multiplying the hourly kwh consumption by four to calculate a (way way way too conservative) worst case assumption that the max hourly kwh was all consumed in a 15min period. For many not-yet-electrified or even partially electrified single family homes the max hourly kwh in the data is under 6kw, i.e the calc comes out to 100A or less. So if the home already has a 200A service, it really proves you can add more load, or downsize a main breaker. Whether you can convince an AHJ this meets the letter of 220.87 I don't know, haven't had to deal with that discussion.

This does not grok to me. It still assumes that the power was on for 15 minutes and off for 45 minutes. But it could have been on for 10 minutes and off for 50 which would give a higher peak. I mean, it will give us a number for kW but I don't think we can say it means we have found the peak kW draw. It might be just as accurate to add up the three or four largest electric devices in the house and say all of them on at once is the peak.

 

[wwhitney]

This does not grok to me. It still assumes that the power was on for 15 minutes and off for 45 minutes. But it could have been on for 10 minutes and off for 50 which would give a higher peak.

Higher 1 minute or 1 second peak, but the same 15 minute peak. So jaggedben is just giving you a very conservative way to go from 1 hour data to 15 minute data. And based on the 220.87 exception, it appears that the "demand data" referred to is 15 minute data.

Cheers, Wayne

 

[jaggedben]

This does not grok to me. It still assumes that the power was on for 15 minutes and off for 45 minutes. But it could have been on for 10 minutes and off for 50 which would give a higher peak. I mean, it will give us a number for kW but I don't think we can say it means we have found the peak kW draw. It might be just as accurate to add up the three or four largest electric devices in the house and say all of them on at once is the peak.

As was discussed in one of the threads I linked to in post #8, utility peak demand data is likely just based on kWh over a 15min interval anyway. So then as far as 220.87 it would make no difference.

Adding up loads by nameplate would be done according to some other method in 220, and for residential will almost always be much higher than real peak demand. But 220.87 is what we're discussing.

Plus what Wayne said.

 

[electrofelon]

but I don't think we can say it means we have found the peak kW draw.

We are not usually looking for peak KW. The 1 year demand option in 220.87 is not peak either.

 

[pv_n00b]

We are not usually looking for peak KW. The 1 year demand option in 220.87 is not peak either.

220.87(1 ) The maximum demand data is available for a 1-year period.
Maximum is peak.

 

[electrofelon]

220.87(1 ) The maximum demand data is available for a 1-year period.
Maximum is peak.

Well it is peak in the sense of the highest of all the demand readings, but each reading is a 15 minute average not a peak.