Bidirectional meter registers

Status
Not open for further replies.

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
This is not a work related question, rather a technical question on my new PV system.

I'm on Duke Energy in Greenville SC. The solar installer and his electrician installed a 4.25 kW (DC) system which has been in operation for about 3 weeks. After city inspection, Duke installed a new meter about 2 weeks ago, single meter which is in my service line, thus without direct knowledge of solar production.

The meter, an Itron Sentinel, sequentially displays 8 values, date, time, 2 in kWh, 2 in kW, and 2 in kVArh. My system uses Enphase micro-inverters, and the monitoring software give me their (Enphase) production information. I'm trying to correlate the information.

As of November 21 evening, Duke's meter gave:

210 kWh
97 kWh
1.35 kW
2.58 kW
31 kVArh
22 kVArh

Enphase tells me I have generated about 190 kWh since the meter was installed. My historic consumption in winter months is around 400-500 kWh. I cannot understand how the 210 kWh can have been determined, UNLESS my consumption has been about double the norm (210+190). I can see how I might have pushed 97 kWh net back, genertaing power for about 9 hours/day, with the average for those 9 hours most of the days over consumption for those hours.

Do any of you know what Duke has programmed these meters to display?

Thanks!
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
151124-0001 EST

The average number of hours per year over a 4 year period is 8766, and an average month is 730.5 hours.

If your average winter consumption has been 500 kWh/month, then average power has been about 685 W. My average is more in the range of 1250 W. But I have a refrigerator, a freezer (only one presently), a furnace blower motor, several computers, printer, and lights.

I will make a wild guess that your solar system might produce an average of 4*6 kWh per day, or 24 kWh. I used 6 hours instead of your 9 hours because the production curve is approximately a half sine wave. You estimate solar production of 190 kWh in 14 days or 13.6 kwh per day. So my estimate of production is almost double your estimate. Your home's daily consumption should average 24*0.69 = 16.6 kWh. This is less than my wild guess of production, and more than your estimated production.

I will guess the meter is not bi-directional, and you might have two tier pricing.

Assume that during 6 hours of the day you feed power to Duke, and supply all of your home's needs, then during 18 hours of the day Duke supplies you power. That would amount to Duke supplying 0.685*18 = 12.3 kWh per day or 172 kWh in 14 days. I am not using a good model, but it might give you some ideas.

See if you can find a used TED 1000 system with software and you could do your own monitoring.

You can try some experiments with your meter to try to figure out the readings. Use something like a controlled 2 to 4 kW load for experiments.

See http://beta-a2.com/energy.html for some of my TED plots. This data was when I was running two freezers, and more computers 24 hours/day than my present consumption. When I shutdowm most equipment at night I can get in the range of 24 kWh/day.

.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
I wouldn't assume the utility meter was at zero when they installed it. Hopefully they tested it before they installed it, and I'm not sure there is any way to reset these.

You might be able to call and get the initial reading.
 

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
151124-0001 EST

I will make a wild guess that your solar system might produce an average of 4*6 kWh per day, or 24 kWh. I used 6 hours instead of your 9 hours because the production curve is approximately a half sine wave.
PVWatts guestimates about 17 kWh/day, annual average. I used 9 hours because my generation was "active" for 9 hours per the generation report ... but time integrated average is less, of course, even less than your 6 hour number. My panels are roof-mounted at 50 degrees which reduced annual production by about 3%, but improves winter production.

The kWh generated came from the Enphase inverters. See the attachment.

I will guess the meter is not bi-directional, and you might have two tier pricing.
I will guess you think I can't read ... or that Duke lies. Their literature says "Your existing electric meter will be replaced with a bidirectional meter capable of measuring this two-way flow of electricity"

I find you exceptionally intelligent, but to ignore what I gave as facts is really condescending. You said that you didn't know what the registers represent which is sufficient.

Thanks!

George
 

Attachments

  • EM2_small.jpg
    EM2_small.jpg
    23.5 KB · Views: 0

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
I wouldn't assume the utility meter was at zero when they installed it.
I wouldn't assume either; I watched her set it and observed that it was zero. Shuld have included that info. Frankly, I expected they to match the old meter to start ... but no.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I wouldn't assume either; I watched her set it and observed that it was zero. Shuld have included that info. Frankly, I expected they to match the old meter to start ... but no.
Setting all registers in a meter to zero is easy. Setting them to a specific value is probably not even possible.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
I wouldn't assume either; I watched her set it and observed that it was zero. Shuld have included that info. Frankly, I expected they to match the old meter to start ... but no.

OK that was just my first thought. Another simple question: Why are there 2 different numbers for KWh? It sounds like the 97 KWh number might be reasonable for 2 weeks use.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
Setting all registers in a meter to zero is easy. Setting them to a specific value is probably not even possible.

I'm not sure why they would even provide a reset. Just seems like something that people could take advantage of. They can always record the KWh reading during the install, and use the difference to calculate usage, just like they do every other month.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
151124-1433 EST

GeorgeB:

From the data that you provided and making some assumptions I was trying to figure some values that might correlate with the meter data. An important question is whether the meter is really bidirectional 100% of the time.

The enphase graph you provided adds some interesting information. I can't read the scale values, but it appears the vertical axis increments by 5 kWh. Thus, peak production is slightly over 20 kWh per day. Somewhat less than my 24 guess.

Eyeball averaging of the graph's last 15 days looks like about 18 kWh per day from your solar system. Your estimated use per day from other years is around 450/31 = 14.5 kWh/day. I believe you said the Duke meter has only been installed for about 2 weeks. Then I assume that corresponds with the last days of the enphase graph.

If the Duke meter is bidirectional, then I would expect a reading closer to zero. What am I missing, or is there a scale factor problem?

I have no way to test my DTE meter relative to bidirectionality, but this morning I did a comparison of my TED 1000 with the DTE meter because your post stimulated me to do a test I have put off. I don't excpect my TED 1000 to be as accurate as my DTE meter and I know from other tests that TED readings degrade as the reactive component increases. But even with the TED 1000 inaccuracies it can be an effective tool to study a utility meter, and your power and energy use.

This morning's tests in my home were:

Code:
   TED              DTE

  160 W           150 W        TED high by  7%,    almost everything turned off
 5075 W          5067 W        TED high by  0.2%,  large resistive loaded added

  980 W           885 W        TED high by 11%,  resistive load removed, freezer & furnace blower added
 1170 W          1055 W        TED high by 11%,    refrigerator added

I believe the near zero load has a fairly large inductive component. Some small transformers, and some non-linear loads. A separate 1000 system when measuring a 100 W resistive load is within about 1%.

The TED 1000 MTU (Measuring Transmitting Unit) outputs bidirectional power data, but their software in some modes does an absolute value conversion, and thus in these modes sign information is lost.

.
 

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
Thanks! Correlating the data is exactly what I'm trying to do. I trust Duke, but want to understand. That's a characteristic this engineer has <g>. My embedded comments in {}

151124-1433 EST

From the data that you provided and making some assumptions I was trying to figure some values that might correlate with the meter data. An important question is whether the meter is really bidirectional 100% of the time.{great question; itron suggests there is enough flexibility with the sentinel to do many things; see https://www.itron.com/na/productsAndServices/Pages/SENTINEL.aspx if interested; mine has cellular connection which the installer thought was used at least daily}

The enphase graph you provided adds some interesting information. I can't read the scale values, but it appears the vertical axis increments by 5 kWh. Thus, peak production is slightly over 20 kWh per day. Somewhat less than my 24 guess. {your guess would be more valid if we'd had good sunny days}

Eyeball averaging of the graph's last 15 days looks like about 18 kWh per day from your solar system. Your estimated use per day from other years is around 450/31 = 14.5 kWh/day. I believe you said the Duke meter has only been installed for about 2 weeks. Then I assume that corresponds with the last days of the enphase graph.{yes; I didn't note when they installed the meter, but I was present ... and nosy}

If the Duke meter is bidirectional, then I would expect a reading closer to zero. What am I missing, or is there a scale factor problem?{that's what i'm trying to figure out too; I doubt I QUITE accomplished 100% of my consumption, but even the 97 kWh seems ... maybe wrongly ... high}

I have no way to test my DTE meter relative to bidirectionality, but this morning I did a comparison of my TED 1000 with the DTE meter because your post stimulated me to do a test I have put off. I don't expect my TED 1000 to be as accurate as my DTE meter and I know from other tests that TED readings degrade as the reactive component increases. But even with the TED 1000 inaccuracies it can be an effective tool to study a utility meter, and your power and energy use.

This morning's tests in my home were:

Code:
   TED              DTE

  160 W           150 W        TED high by  7%,    almost everything turned off
 5075 W          5067 W        TED high by  0.2%,  large resistive loaded added

  980 W           885 W        TED high by 11%,  resistive load removed, freezer & furnace blower added
 1170 W          1055 W        TED high by 11%,    refrigerator added

I believe the near zero load has a fairly large inductive component. Some small transformers, and some non-linear loads. A separate 1000 system when measuring a 100 W resistive load is within about 1%.

The TED 1000 MTU (Measuring Transmitting Unit) outputs bidirectional power data, but their software in some modes does an absolute value conversion, and thus in these modes sign information is lost.

.

It's interesting that these itron (ex-Schlumberger) meters are made "locally"; their old plant was a major source of PCB contamination of several waterways here. They (unidirectional) are available cheap from "DIY" generation sites, but not with the Duke firmware I'm sure. see http://www.civicsolar.com/product/itron-c1s-cl200-centron-digital-kilowatt-hour-meter-240vac-2s-form

Thanks. I've asked the customer-owned-generation group at Duke the question; I doubt an answer, but maybe ...?
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
151124-2400 EST

GeorgeB:

I received my DTE smart meter about 4 years ago. About 6 months later I was able to get hourly data from my meter from DTE's website about a day and a half after the data was generated. Hourly data for energy use was provided. Apparently the data for the previous day was sent somewhat after midnight via a mesh network from home to home to some sort of repeater. Possibly a cellphone connection.

The on-line data and graphs were fairly good, and data could be down-loaded. September of 2013 the website data provided was changed. Better in some ways and not in other ways. This fall, October 2015, the on-line data has been eliminated.

Beginning a while back was a smart phone application to provide energy usage. This is quite lacking and data can not be downloaded. This application is called Insight. In combination with Insight there is now a box called Bridge that RF links directly to the smart meter and provides short time average data. Response time about 2 to 10 seconds. This I used to get the DTE data in today's experiment. Resolution is possibly 2 W but noisy.

You probably lack most of this capability from Duke at this time.

My meter is an Itron C2S0D 30 TA 1.0 Kh CA 0.5 FM2S 60 Hz. I can do crude short time average power measurement via a sequencing square dot on the meter.

You may be able to apply a large load on your meter for a controlled time and see how the meter reading changes. Also turn off virtually everything in your home when the solar is producing maximum power and see how the meter display changes. A one hour test period could be useful when generating about 4 kW.

The lost cost availability of Itron meters is interesting.

.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
In PG&E land the bidirectional smart meters separately add up net in and net out values separately. And they can accumulate both figures for three different rate periods in case you are on a TOU plan.
 

Rampage_Rick

Senior Member
My meter is an Itron C2S0D 30 TA 1.0 Kh CA 0.5 FM2S 60 Hz. I can do crude short time average power measurement via a sequencing square dot on the meter.
We have those same meters here in BC. The registers are configured to the utility's specs; ours display kWh in and kWh out on the regular 200A 1PH models. (though I've never come across one that shows more than 0 kWh out) The polyphase OpenWay meters we have also show kVArs and peak load in W.

As to the "spinning" squares on the LCD, I believe that each blip represents one watthour of consumption, so a full cycle of squares equals 6 watthours. In addition there's an IR LED on the top of the meter that blinks once per watthour (you can see it with a cell phone camera)
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
151125-0803 EST

Rampage_Rick:

The calibration on my meter is the same as yours. One dot step = 1 Wh.

A step period of 3.6 seconds = 1 kW. Or 21.6 seconds for 6 steps is 1 kW, an easier measurement for useful accuracy.

800 W is 27 seconds for 6 steps.
180 W is 20 seconds for 1 step.
4000 W is 21.6 seconds for 24 steps.

With a stopwatch one can probably get accuracy of average power of about 1% with a test period around 30 seconds.

An equation for average power measurement is

Pave = 3.6*Kh/T

Where
Pave in kW is the average power over the time of measurement T.
Kh is the meter scaling factor, 1 for our meters.
T is the measurement time divided by the number of dot steps in the measurement time.

.
 

GeorgeB

ElectroHydraulics engineer (retired)
Location
Greenville SC
Occupation
Retired
Duke was more forthcoming than I expected; should have asked them first. Their email said " Pretty soon you should be receiving a welcome packet that explains how to read the meter."

GREAT!

It went on to say that the 1st is how much I've consumed from them. The 2nd is how much I've supplied to them. So my "net" to pay for is the difference.

My generation (via Enphase) less what I've supplied them plus what they've supplied me is my consumption, so I've used a little more than my historical average.

The kW registers were not addressed. Nor were the kVArh, but I can ASSUME the same relationship as kWh. I'm a little surprised my reactive power is as high as it is, but I calculated 0.988 from the most recent (less than ideal resolution) numbers, and that's not bad.

Thanks everyone!

George
 
Status
Not open for further replies.
Top