difference between electronic meter and Sunny Boy Control values

[Melodie de l'Epine]

We now have several examples on grid connected PV system of 1 to 3kWp
where there is a difference in the number of kWh measured by the
electronic meter installed by the utility, and the inverter.

for example:
electronic utility meter / SMA SWR700 inverter reporter by a Sunny Boy
Control

june 03: 128 / 120
july 03: 135 / 120
september 03: 108 / 100
october 03: 57 / 50

the values are quite close, so it is hard to say which value is most
likely to be correct - and asking the utility to check their meter is
quite expensive, if the meter is ok.

And in another example, this time with a Gridfit 2200 (Total Energie)
inverter, the inverter reports nearly twice as much production as the
electronic utility meter (102kWh/fortnight as opposed to
66kWh/fortnight). In this case, the utility says that they have checked
the meter and that it is properly calibrated, BUT a quick simulation
shows that the production reported by the inverter is the most likely.

We also have seen several systems with both disk meters and electronic
meters (one private and one belonging to the utility) with, again, a
difference in favour of one or the other, generally a difference of 1 to
10 kWh/month

Until recently most systems here relied on disk meters for monitoring
production values, so we don't really have much experience with
electronic meters.

Could there be a problem with the inverters' built-in meters? Or are
electronic meters notoriously unreliable? Maybe disk meters have a
tendency to undervalue?

Any experience, insight, pointers etc would be helpful.
thanks!

Melodie

 

[Melodie de l'Epine]

Solar Guppy a écrit :

Hi Melodie,

Welcome to the word of Manufacture reported data ...

The SMA products report 2-3% high from my testing on over 5 different units
... I've directly spoken to them and they say the customers are "happy" with
how the units report data ... read the owner like seeing big numbers ..The
Xantrex Suntie is within 1% ...

OK, but.... the Suny Boy is telling us less than the utility meter! Are
they just being cautious? This seems to be the wrong way 'round!

Other manufactures an 2X worse in my testing , sorry , I can't release who
as I am under an NDA on the testing .. but I've done extensive testing on
Fronius , PVPowered , SMA and Xantrex with lab grade power analyzers (.08%
accuracy).

Believe the Utility kWhr meter , it is the true measure of kWhr to the grid

Of course! but if that is the case, I'd like to know why the inverter in
the second recorded case is telling us such big numbers! Here we are
looking at a differnce of over 30%, compared to the inverter.

Maybe the inverters performance is suffering? The only thing I can think
of is that the inverter is reporting dc side kWh and functioning at a
very poor level of performance (around 70%), which could explain the
losses. The technical doc says over 90% efficiency at 20% of maximum power.

True, the GrdiFit inverter is a newcomer to the market, but still!!

 

[Pom-pom-pom]

We now have several examples on grid connected PV system of 1 to 3kWp
where there is a difference in the number of kWh measured by the
electronic meter installed by the utility, and the inverter.

for example:
electronic utility meter / SMA SWR700 inverter reporter by a Sunny Boy
Control

june 03: 128 / 120
july 03: 135 / 120
september 03: 108 / 100
october 03: 57 / 50

the values are quite close, so it is hard to say which value is most
likely to be correct - and asking the utility to check their meter is
quite expensive, if the meter is ok.

And in another example, this time with a Gridfit 2200 (Total Energie)
inverter, the inverter reports nearly twice as much production as the
electronic utility meter (102kWh/fortnight as opposed to
66kWh/fortnight). In this case, the utility says that they have checked
the meter and that it is properly calibrated, BUT a quick simulation
shows that the production reported by the inverter is the most likely.

We also have seen several systems with both disk meters and electronic
meters (one private and one belonging to the utility) with, again, a
difference in favour of one or the other, generally a difference of 1 to
10 kWh/month

Until recently most systems here relied on disk meters for monitoring
production values, so we don't really have much experience with
electronic meters.

Could there be a problem with the inverters' built-in meters? Or are
electronic meters notoriously unreliable? Maybe disk meters have a
tendency to undervalue?

Any experience, insight, pointers etc would be helpful.
thanks!

Melodie

<IRONY>
Do you suggest that the utility could cheat?
Come on, we all know how open and transparent is the special utility you are
talking about! They wouldn't mislead people, especially those who installed
alternative energy sources ;-)
</IRONY>


There is a strong lack of knowledge in electricity metering from the average
customer. Most people don't know the difference between a Volt and a Watt,
and don't have a clue on how many Wh or TWh they consume each month.

This is different with gas in cars. Most people know approximatly their
yearly consumption, albeit with a small calculation (such as "My car is 10
L/100, I drive 15 000 Km a year, that's 1500 L of gas). They would notice
the slightest change in gas-mileage or if the gas pump tried to cheat.

Nothing with electricity. The utility does all the job, and eventually
transfers the money directly from the customer's bank account. Nobody ever
check anything.

I recently bought a 'Kill-a-watt' meter and show it to friends, or just
yesterday, to a manager of my municipality. They all seem to discover
something completly new and are astonished when they find how much energy a
single device uses.

If we want a better energy behavior (savings, renewables...), we ought to
introduce a culture of electricity metering. And empower people to check the
validity of their meters...

 

[John Beardmore]

Solar Guppy a écrit :

OK, but.... the Suny Boy is telling us less than the utility meter! Are
they just being cautious? This seems to be the wrong way 'round!

What exactly is being measured ? Watts ? VA ?

Could power factor be an issue ?


Cheers, J/.

 

[[email protected]]

Ha again Guys & Gals;

(Quote from lower down)

">I recently bought a 'Kill-a-watt' meter and show it to friends, or
just

yesterday, to a manager of my municipality. They all seem to discover
something completly new and are astonished when they find how much energy a
single device uses."

While on the subject of energy hog appliances.

I set up 170 watts of my panels with a 100 amphour battery and a
(cheepo) 400 watt inverter to power father's pond pump, to let him see
a real life example of my panels at work.

As soon as I plug in the pump (witch is very small), my little
inverter's fan starts and while it never really complains, the fan is
always on (witch has never happened before and I assume is not really
good),

When I get to about 6 amps (from my demo array, whitch is usually at 7
or 8 amps on a clear day, 1.5 ish if it is overcast) the battery
starts holding its own and the voltage remains constant until about
4:30 (at whitch time I switch the pond pump to the grid). This pump
only pumps about 15 gpm, but we were both amazed at how much energy it
uses.

Keep smilin' and solarin'

Glenn

PS. Would a larger (cheepo, as I am pooro Ha Ha) inverter be more
efficiant on this pump? Over the winter I will be shopping for a more
efficiant pump.

 

[daestrom]

Ha again Guys & Gals;

(Quote from lower down)

">I recently bought a 'Kill-a-watt' meter and show it to friends, or
just

While on the subject of energy hog appliances.

I set up 170 watts of my panels with a 100 amphour battery and a
(cheepo) 400 watt inverter to power father's pond pump, to let him see
a real life example of my panels at work.

As soon as I plug in the pump (witch is very small), my little
inverter's fan starts and while it never really complains, the fan is
always on (witch has never happened before and I assume is not really
good),

When I get to about 6 amps (from my demo array, whitch is usually at 7
or 8 amps on a clear day, 1.5 ish if it is overcast) the battery
starts holding its own and the voltage remains constant until about
4:30 (at whitch time I switch the pond pump to the grid). This pump
only pumps about 15 gpm, but we were both amazed at how much energy it
uses.

Keep smilin' and solarin'

Glenn

PS. Would a larger (cheepo, as I am pooro Ha Ha) inverter be more
efficiant on this pump? Over the winter I will be shopping for a more
efficiant pump.

You don't give a lot of specifics, but if your panel is 170 watts, the
inverter is 90%, the motor is 90% and the pump is 90% (somewhat optomistic,
but not *impossible*), that leaves 124 watts for pumping water.

124 watts will pump 15 gpm at about 44 ft head (19 psi). You don't say what
the head is you're pumping at, but for a pond pump, this seems kind of
reasonable to me.

You may not be able to do much better than that.

daestrom
P.S. 124 watts running 24/7 for 365 days @ $0.10/kwhr would be ~$108/yr
(assuming it doesn't freeze up in winter ;-)

 

[Melodie de l'Epine]

John Beardmore a écrit :

What exactly is being measured ? Watts ? VA ?

Could power factor be an issue ?


Cheers, J/.

The Gridfit manual says it displays kWh - we are measuring & comparing
production, and not power.


Melodie

 

[Melodie de l'Epine]

John Beardmore a écrit :

What exactly is being measured ? Watts ? VA ?

Could power factor be an issue ?


Cheers, J/.

If I understand Power Factor correctly, no it shoudln't be an issue - we
should be not far from 1.

 

[James Baber]

From Jim Baber
I have a problem understanding Melodie's problem and I am very familiar with the
reporting from my Sunny Boy Control. I have been making system reports and
grid comparisons daily for 442 days as of today. I must admit that I have a
much larger system than Melodie does (10kW) and am using 4 of the Sunny Boy
Model SWR2500U inverters they have been running for 10882 hours, and have
produced 22643 kWh to date. I do not understand where she is reading the total
wattage fed onto the grid by her Sunny Boy inverter. The Sunny Boy Control I
have can only report the total kWh produced by the inverter(s), NOT how much is
being fed onto the grid. IT DOES NOT MEASURE the same thing as the utilities
meter(s).

If Melodie's inverter is wired correctly, it's output is fed into her system
(the load at the breaker box) between the utilities meter and the load at the
breaker box.

It is interesting to note that I do own a EXTECH 382065 a True RMS Clamp on
power meter that records its readings for quite a while, and I have actually
recorded the Wattage on the utilities service into my home to 2 decimal point
accuracy. The utilities wattmeter is only reading to the whole number, but it
appears to be within a 0.15 kWh of the clamp on meters reading over any given 48
hour period.

Just for giggles for the past 73 days the utility owes me ($92.12) for power I
have put onto the grid. I am on Time of Use metering and they have to give me
credit at the $0.33 / kWh rate for any peak use power I put on the grid. The
peak use period is from noon to 6 PM while the off peak is everything else. (the
rate for off peak is $0.09 / kWh). Guess what time of day my solar makes the
most power? works well for me, I have actually used 1,492 kWh of their power,
yet they owe me money, neat (for a change.

My calculations are accurate enough that I can determine my bill within a +/-
$0.15 range before the meter reader gets out of my nextdoor neighbors driveway.
It took me about 5 months to figure Pacific Gas & Electric's system out, but
I do understand it now. My spreadsheet for the calculations takes 1 page,
theirs takes 10. I actually do use their published rate structure, with the
only problem being the frequency that they change their structure and / or
rules. So far they have changed 4 times since December, usually retroactively,
and doesn't that mess up a spreadsheet. Believe me, their meter's accuracy
isn't the big problem. 2 of my own questions caused some of those @#$%^& rule
revisions to clarify actions. They didn't do what the rules said on the bills.

We now have several examples on grid connected PV system of 1 to 3kWp
where there is a difference in the number of kWh measured by the
electronic meter installed by the utility, and the inverter.

for example:
electronic utility meter / SMA SWR700 inverter reporter by a Sunny Boy
Control

june 03: 128 / 120
july 03: 135 / 120
september 03: 108 / 100
october 03: 57 / 50

the values are quite close, so it is hard to say which value is most
likely to be correct - and asking the utility to check their meter is
quite expensive, if the meter is ok.

And in another example, this time with a Gridfit 2200 (Total Energie)
inverter, the inverter reports nearly twice as much production as the
electronic utility meter (102kWh/fortnight as opposed to
66kWh/fortnight). In this case, the utility says that they have checked
the meter and that it is properly calibrated, BUT a quick simulation
shows that the production reported by the inverter is the most likely.

We also have seen several systems with both disk meters and electronic
meters (one private and one belonging to the utility) with, again, a
difference in favour of one or the other, generally a difference of 1 to
10 kWh/month

Until recently most systems here relied on disk meters for monitoring
production values, so we don't really have much experience with
electronic meters.

Could there be a problem with the inverters' built-in meters? Or are
electronic meters notoriously unreliable? Maybe disk meters have a
tendency to undervalue?

Any experience, insight, pointers etc would be helpful.
thanks!

Melodie

I would be glad to help off the group if you prefer.
--

Jim Baber

(see my 10kW grid tied solar system at "www.baber.org")

1350 W Mesa Ave.
Fresno CA, 93711
(559) 435-9068
(559) 905-2204 cell

Email [email protected]

 

[Melodie de l'Epine]

James Baber a écrit :

From Jim Baber
I have a problem understanding Melodie's problem and I am very familiar
with the reporting from my Sunny Boy Control. I have been making
system reports and grid comparisons daily for 442 days as of today. I
must admit that I have a much larger system than Melodie does (10kW) and
am using 4 of the Sunny Boy Model SWR2500U inverters they have been
running for 10882 hours, and have produced 22643 kWh to date. I do not
understand where she is reading the total wattage fed onto the grid by
her Sunny Boy inverter. The Sunny Boy Control I have can only report
the total kWh produced by the inverter(s), NOT how much is being fed
onto the grid. IT DOES NOT MEASURE the same thing as the utilities
meter(s).

If you take a manual reading from the Sunny Boy control, and a manual
reading from your production utility meter within about 1 minute of each
other on the same frequency (once a week, once a fortnight, month,
whatever), you can then calculate how many kWh you have produced over
that period of time (new reading - old reading = period production).

1/by reading from the Sunny Boy control you should be able to see how
many kWh your inverter has produced*
2/by reading from the utility meter you should be able to see how many
kWh the utility meter says you have injected into the grid.

If you are configured to sell ALL of your production to the grid (ie not
use any in you internal home system), then they should be the same,
shouldn't they?

I would expect that a couple of meters of cables (where the cables have
been correctly sized) and the few accessories (switche, fuses, whatever
etc) wouldn't amount to even 1kWh lost over a week or even a month.



*this is part of my problem - is the inverter telling me kWh entering
the inverter or kWh exiting the inverter? Becasue this could be one
reason why the readings differ.

If Melodie's inverter is wired correctly, it's output is fed into her
system (the load at the breaker box) between the utilities meter and the
load at the breaker box.

Where I am, when connecting PV to the grid you have three options
1- connect PV to interior of home circuit and sell surplus electricity
only, as you seem to suggest (I am not really clear about a lot of
technical terms in english, sorry; I presume "breaker box" is where all
the fuses are,and that it is the first "private" "destination" for
electricity bought from the "public" grid, once the electricity has gone
through the "public" meter)
2- connect PV to a branch in the main line created by the utility, sell
all the electricty (pv must be sold by same legal entity that consumes
electricity on this line)
3- get a new, separate main line installed and connect PV to this new,
separate main line (at a given physical site the consumer and producer
can be separate legal identites as there are two main lines)




Melodie

 

[James Baber]

Reply from Jim to Melodie's response

James Baber a écrit :

I emphasize again the Sunny Boy Control can only measure the power that the
inverter(s) it controls has created.

If you take a manual reading from the Sunny Boy control, ....

true for the total PV supplied power

......and a manual reading from your production utility meter within
about 1 minute of each other on the same frequency (once a week,
once a fortnight, month, whatever), you can then calculate how many
kWh you have produced over that period of time
(new reading - old reading = period production).

Not true for the production utility meter.

This will always be 'only' the power that you are actually 'still buying'from
the utility. It has nothing to do with what you are producing from your PV
system, IF you are wired as you are supposed to be per the manual. It will not
show your production at all.

1/by reading from the Sunny Boy control you should be able to see how
many kWh your inverter has produced* YES
2/by reading from the utility meter you should be able to see how many
kWh the utility meter says you have injected into the grid. NO

If you are configured to sell ALL of your production to the grid (ie not
use any in you internal home system), then they should be the same,
shouldn't they?

This would be true, only if you had 2 utility meters. One that is used to
measure what you are using on your internal home system, and the second used to
measure all of the output from the PV system as it is put into the grid. The
net difference between the two meters would be what you would actually have to
pay for. It could be done this way, but I am sure you would have problems
getting the utility to agree to it because of the billing complexity.

I would expect that a couple of meters of cables (where the cables have
been correctly sized) and the few accessories (switchs, fuses, whatever
etc) wouldn't amount to even 1kWh lost over a week or even a month. True.

*this is part of my problem - is the inverter telling me kWh entering
the inverter or kWh exiting the inverter? Becasue this could be one
reason why the readings differ.

Where I am, when connecting PV to the grid you have three options
1- connect PV to interior of home circuit and sell surplus electricity
only, as you seem to suggest (I am not really clear about a lot of
technical terms in english,

Yes, number 1 is the typical normal installation in California where the law
requires the utility to effectively credit the customer for all power sold back
to the grid.

sorry; I presume "breaker box" is where all
the fuses are, and that it is the first "private" "destination" for
electricity bought from the "public" grid, once the electricity has gone
through the "public" meter)

Yes that is what I meant

2- connect PV to a branch in the main line created by the utility, sell
all the electricty (pv must be sold by same legal entity that consumes
electricity on this line)

This is true if there is a 2nd utility meter to measure the sold back power.
Then your original argument that the "2nd" utility meter would read the same as
the inverters meter readings is true.

3- get a new, separate main line installed and connect PV to this new,
separate main line (at a given physical site the consumer and producer
can be separate legal identites as there are two main lines)

This also would be true, IF there was a 2nd utility meter to measure the sold
back power separate from your home.

In your cases 2 and 3 there must be a 2nd meter, but it might be built into the
same housing, although I have not seen such a unit.

Interestingly enough my own meter has 2 separate dials, but this is setupso one
dial reads total kWh and the other reads kWh between noon and 6 PM on weekdays.
This is how my utility bills for peak hour usage and off peak hour usage.
They both can and do run backwards when I am putting power on the grid. Actually
it has 4 - a third shows the instantaneous kW currently being drawn from or sent
to the grid, the 4th is not turned on but it can show the power factor for the
customer (I know what this is, see my website for an explanation, It is why I
bought the expensive clamp on meter I mentioned)

Melodie

--

Jim Baber

(see my 10kW grid tied solar system at "www.baber.org")

1350 W Mesa Ave.
Fresno CA, 93711
(559) 435-9068
(559) 905-2204 cell

Email [email protected]

 

[Melodie de l'Epine]

James Baber a écrit :

Reply from Jim to Melodie's response

I emphasize again the Sunny Boy Control can only measure the power that
the inverter(s) it controls has created.

And there is only 1 inverter in the system, so it is measuring all the
PV production

true for the total PV supplied power




Not true for the production utility meter.

This will always be 'only' the power that you are actually 'still
buying' from the utility. It has nothing to do with what you are
producing from your PV system, IF you are wired as you are supposed to
be per the manual. It will not show your production at all.

There are lot's of "manuals" around - and they all are different. We
have a different manual here in France (you can consult it (in French
only) at:
http://www.grd.edf.fr/sge/viewDocum...chniques_et_cntractuelles_de_raccordement.pdf
The manual has mistakes, but considering the resistance the national
utility has displayed in the past to grid connected systems, it is a
miracle that they wrote and published it at all.

This would be true, only if you had 2 utility meters. One that is used
to measure what you are using on your internal home system, and the
second used to measure all of the output from the PV system as it is put
into the grid. The net difference between the two meters would be what
you would actually have to pay for. It could be done this way, but I am
sure you would have problems getting the utility to agree to it because
of the billing complexity.

Three utility meters, in fact. One on a separate line to measure
consumption, and two on the production line: one to measure the
inverter's consumption (which must be less than 1% of the production
otherwise you must pay consumption connection fees) and one to measure
production.

Yes, number 1 is the typical normal installation in California where the
law requires the utility to effectively credit the customer for all
power sold back to the grid.

Credit, but not buy at different rates. Here they buy at a higher price
than the consumer pays for their electricity, which is very cheap. (we
are after all, the world nuclear capital - and because there is so much
over production, it has to be sold of somehow!)

Yes that is what I meant


This is true if there is a 2nd utility meter to measure the sold back
power. Then your original argument that the "2nd" utility meter would
read the same as the inverters meter readings is true.

There is a second utility meter, but as it is on a separate, independant
line I didn't bother mentioning it. Sorry.

And they are electronic meters, so when they measure, they measure each
"direction" separatly, unlike disc meters (in fact, they have been
configured to only measure one direction and you need two here, because
the local utility doesn't want to have to redesign thier software to
deal with both negative and positive numbers on the same meter)

This also would be true, IF there was a 2nd utility meter to measure the
sold back power separate from your home.

In your cases 2 and 3 there must be a 2nd meter, but it might be built
into the same housing, although I have not seen such a unit.

Interestingly enough my own meter has 2 separate dials, but this is
setup so one dial reads total kWh and the other reads kWh between noon
and 6 PM on weekdays. This is how my utility bills for peak hour usage
and off peak hour usage. They both can and do run backwards when I am
putting power on the grid. Actually it has 4 - a third shows the
instantaneous kW currently being drawn from or sent to the grid, the 4th
is not turned on but it can show the power factor for the customer (I
know what this is, see my website for an explanation, It is why I bought
the expensive clamp on meter I mentioned)

I would imagine that most electronic meters have enough memory "slots"
to measure off-peak and peak consumption/productiona as well as
instantaneuouse power and period high & low values. Once you've put the
chip in, you may as well use it!



So, know that I have given all the information required, we are back to
square one - we have an inverter saying the PV system is producing
nearly twice as much as what the utility meter is saying, and I can't
figure out why.

 

[Pom-pom-pom]

(...)
There are lot's of "manuals" around - and they all are different. We
have a different manual here in France (you can consult it (in French
only) at:
http://www.grd.edf.fr/sge/viewDocument?contentId=Commun/Documents/conditions
_techniques_et_cntractuelles_de_raccordement.pdf
The manual has mistakes, but considering the resistance the national
utility has displayed in the past to grid connected systems, it is a
miracle that they wrote and published it at all.
(...)

You may also find pointers from here:
http://www.edf.fr/index_mv.php4?coe_i_id=14011. Maybe you already know it.
I found by chance this link: http://www.edf.fr/download.php4?coe_i_id=14009
Then googled back to find pages linking to it. Not an easy task.
"They owe us more than light, they said, pfff!"

Then I found Hespul and the light there was much brighter. Thank you very
much!

Pascal

 

[James Baber]

From Jim Baber
Melodie, I have a much better understanding of your problem now. I have to
complement you for doing such a good job of communicating a difficult problem in
the English language (which is the poorest structured of any language that I am
at all aware of), I just wish I was as fluent in French, or German, or even
Spanish, which we do see a lot of in this part of the United States.

Melodie de l'Epine concluded with:

So, know that I have given all the information required, we are back to
square one - we have an inverter saying the PV system is producing
nearly twice as much as what the utility meter is saying, and I can't
figure out why.

Now that I see how you are configured, I am familiar with the SMA (Sunny Boy
SWA2500U which is specifically built for USA standards) inverter you are using
inverter equipment that is quite likely very different.

* For one possible difference the USA version does claim a power factor of 1
(ideal) and my tests did confirm that for all 4 of my inverters had that power
factor

* Another possible factor, is as you suspect bad wattmeters, for this you need
ask your local electricians if any of them can test the wattmeters in all of the
circuits using a clamp on wattmeter like I described earlier. (Probably cheaper
than having the utility to do the test.)

* If they seem to understand what you are talking about you might ask them to
also check the power factor for all 3 of your metered circuits at the same time.
They may be able to explain power factors to you, however, do not be surprised
if you only get a blank (I don't have any idea what this person is talking about
look) from the electrician. This is not a French / English translation problem,
out of maybe 30 electrical engineers / technicians that I have talked with, only
7 even knew the term (2 were knowledgeable but didn't have test equipment), 5
from the utility knew how to measure it, but not what it was. It can make a
difference but not as much as you suggest you have. By the way that clamp on
test meter is quite expensive ($850.00 US) so don't be surprised if they don't
have one.

Our little conversation has been fun, I hope I do not come across as a cocky
know it all Yank, I do not mean to. I hope I have been of some assistance

Jim

There are lot's of "manuals" around - and they all are different.

I'll bet there are. In your choice of language

We have a different manual here in France (you can consult it (in French
only) at:
http://www.grd.edf.fr/sge/viewDocum...chniques_et_cntractuelles_de_raccordement.pdf

I Wish I could, I looked but the French is beyond me.

The manual has mistakes, but considering the resistance the national
utility has displayed in the past to grid connected systems, it is a
miracle that they wrote and published it at all. > US

Particularly, considering some of the German attitudes. (The US manual was
rewritten in California I understand)
I would think that to be true, based on my previous reply below.

Melodie:
Three utility meters, in fact. One on a separate line to measure
consumption, and two on the production line: one to measure the
inverter's consumption (which must be less than 1% of the production
otherwise you must pay consumption connection fees) and one to measure
production.



Credit, but not buy at different rates. Here they buy at a higher price
than the consumer pays for their electricity, which is very cheap. (we
are after all, the world nuclear capital - and because there is so much
over production, it has to be sold of somehow!)

Sounds good to me, I wish that were true were. The best a Californian can do is
get what they would have paid for the same amount of power. (and that is
better than most states)

There is a second utility meter, but as it is on a separate, independant
line I didn't bother mentioning it. Sorry.

And they are electronic meters, so when they measure, they measure each
"direction" separatly, unlike disc meters (in fact, they have been
configured to only measure one direction and you need two here, because
the local utility doesn't want to have to redesign thier software to
deal with both negative and positive numbers on the same meter)



I would imagine that most electronic meters have enough memory "slots"
to measure off-peak and peak consumption/productiona as well as
instantaneuouse power and period high & low values. Once you've put the
chip in, you may as well use it!

--

Jim Baber

(see my 10kW grid tied solar system at "www.baber.org")

1350 W Mesa Ave.
Fresno CA, 93711
(559) 435-9068
(559) 905-2204 cell

Email [email protected]

 

[daestrom]

Three utility meters, in fact. One on a separate line to measure
consumption, and two on the production line: one to measure the
inverter's consumption (which must be less than 1% of the production
otherwise you must pay consumption connection fees) and one to measure
production.

<snip>

My goodness! The utility certainly doesn't make it easy to hook up PV in
France, do they? I suppose you had to pay for the installation of the
second line and all those extra meters?

So, know that I have given all the information required, we are back to
square one - we have an inverter saying the PV system is producing
nearly twice as much as what the utility meter is saying, and I can't
figure out why.

I'm not familiar with the SunnyBoy exactly, but a couple of ideas come to
mind...

Being off by nearly double, is it possible that one of the two meters is set
for the wrong voltage? I'm going to assume there are no
current-transformers and that both your meter and utility's meter are the
common, round type that plug into a base. You can probably check your
inverter's settings yourself, but the utility would have to check theirs and
that could cost you.

Depending on the exact internal design of your inverter, it could be a
rather 'simple' inverter that actually creates a lot of harmonics on the
line. I know that old-style revolving disk types of utility meters *will*
register properly with this sort of harmonic on the line. But perhaps the
electronic utility meter does not do so well?

Measure the DC power flowing into the inverter on a good sunny day (DC power
is simply volts times amps). Sometime when the conditions will remain
steady for an hour or more. Check the DC power a couple of times during a
two or three hour period and note the utility and sunny boy's kwhr meter
readings. Multiplying the time in hours (or fraction of hour) times the
average DC power flow times a guess for inverter efficiency (85%) should
show you which kwhr meter is close to being right. If the sunnyboy kwh
meter shows *more* than the energy flow into the inverter by these DC power
measurements, that's impossible so the kwh meter is bad. If the utility
meter shows less than half the energy that the DC power measurements
suggest, you would have had a fire or the utility meter is bad.

daestrom

 

[mel]

Thanks for your suggestions, Jim.

(I'd like to take credit for having learnt English very well... but
English is my mother tongue. Unfortunatly, my PV and electricty
vocabulary is only in French... which is why I sometimes have trouble
translating back and forth)

I had a quick read-around about power factor, and if I have correctly
understood, you are right, the power factor should be 1.

The PV systems aren't "under my nose", but I work for a non-profit
association that promotes PV, and basically serves as a free users help
service, when the installer and/or supplier gets out of their depth.
(thanks Pascal for the positive feedback - it's always nice to know that
some people appreciate us, even if EDF still isn't our biggest fan!)

I will see if we can get the owner to re-check their meters - but I
think I am leaning towards a fault in the inverter counter - the Grid
Fit inverters from Total Energie are really new on the French market,
and (as we have also seen with the Mastervolt inverters) there is a bit
of a "setteling in" period.

again, thanks for your time.

Melodie



James Baber a écrit :

 

[Melodie de l'Epine]

Well, after further investigation (and a friendly utility technician
that gasp! didn't ask the owner to pay anything), we have some more
information on the Total Energie inverter that was giving strange
readings compared to the utility meter.

The technician measured (how, I don't know ) that there is problem at
low power levels with the inverter : -40% dephased.

Going back to the inverter specifications, Total Energie note that when
the inverter is at high power levels (>50%) the power factor is 1. SMA
says a power factor of 1, without specifying any power levels.

So Watson here has deduced that Total Energie inverters dephase at low
power levels and they (TE) know it.

If I understand correctly, this is what is happening (any comments would
be most welcome - I'm new to the technical side)

We have the inverter measuring something (don't now what yet...), and
converting that into a P(w) as though the electricity was in phase (I
and U inphase), then once again converting into kWh.

Next to this we have the meter that measures intensity and voltage at
the same time (to give P(w)) - however, becasue the current is dephased,
and becasue the meter is not capable of extrapolating the results as if
they were in phase, it gives us a lower power level than the inverter (P
= U x I: if I is lower because it's out of phase, we will automatically
get a lower P value, and hence less kWh)

I suspect that if TE had not included a built-in meter, this would not
have posed any problems - because the owner would have installed thier
own meter between the inverter and the utility meter, and the two meters
would have read the same thing.

I guess our next step is to go back to TE and to ask them how they
programmed thier built-in meter, and what sort of performance there is
at low power levels.

Having learnt all this, I can now extrapolate to the SMA inverters, and
guess that the SMA also dephases at low power levels, but to a much
lesser extent - which wouldn't surprise me, considering how long they
have been on the market, I expect them to be more efficient than the TE
inverters, that have only been on the market for 6 months or so.


Once again, thanks for all your pointers

Mel









Melodie de l'Epine a écrit :

 

[[email protected]]

The technician measured (how, I don't know ) that there is problem at
low power levels with the inverter : -40% dephased.

Sounds like either a) the power factor (the ratio of real power in watts
to apparent power in VA) was 0.6 instead of 1 (ideal), or b) it was 0.4.
That in itself may not be a problem, unless (as you say) it somehow leads
to the TE overreporting delivered energy, compared to the utility meter.

If I understand correctly, this is what is happening (any comments would
be most welcome - I'm new to the technical side)

We have the inverter measuring something (don't now what yet...), and
converting that into a P(w) as though the electricity was in phase (I
and U inphase), then once again converting into kWh.

The product of the current I in amperes and voltage V in volts (what's U?) is
called apparent power in volt-amps (VA), which is larger than real power in
watts, unless the current and voltage are in phase. Real power is IVcos(phi),
where phi is the phase difference. If the TE did not multiply by the cosine,
and the power factor were not 1, the TE would overreport delivered energy.

Next to this we have the meter that measures intensity and voltage at
the same time (to give P(w)) - however, becasue the current is dephased,
and becasue the meter is not capable of extrapolating the results as if
they were in phase, it gives us a lower power level than the inverter (P
= U x I: if I is lower because it's out of phase, we will automatically
get a lower P value, and hence less kWh)

Utility meters are supposed to measure and bill for real power even if
the power factor is not 1. A lot of loads (eg hair dryers on low-heat with
diodes in series) have non-unity power factors, and lot of customers would
complain if they didn't do that correctly. Then again, most customers have
electrical loads vs sources, and it is possible that an electronic meter
might measure real power accurately for a load but inaccurately for a source
with a non-unity power factor. Disk meters work well in either direction--
can you find one of them and hook it up in series with everything else?
From what you've said so far, it seems more likely that the problem is in
the inverter.

Having learnt all this, I can now extrapolate to the SMA inverters, and
guess that the SMA also dephases at low power levels, but to a much
lesser extent...

The SMA may have a different problem, since it under- vs over-reports...

You might just demand payment for the difference between the TE and EDF
readings ("Voleurs!" , or subtract that amount from the bill and let
EDF sort things out at their expense. This might only work once.

Nick

 

[James Baber]

Well, after further investigation (and a friendly utility technician
that gasp........

Congratulations Melodie.
You did a great job for your client with their complex problem, the dephasing
at low power is something I would not have looked at, in fact I believe my
system may not have that problem. Although, in my climate it produces at a low
level for such a short period of the day it still may be happening. With such a
low percentage of the time of a low level operation as I have, it would be
almost impossible to notice the error, unless I specifically look for it.

I don't monitor my inverter's Pac (instantaneous ac power production)
normally, I've always been interested in the net production over a period of
time, 1 day, 1 week ... Since my system does not function very long at low
levels I would never be likely to notice this phasing problem.

You have made me curious, and the next time I hook my recording wattmeter up,
I will change my Sunny Boy Control to monitor the Pac values and compare the two
wattages and the power factor, on a 1 minute by minute set of plots early in the
mornings for a few days.

It is a hassle to set my meter up for this kind of extended testing outside
where the meter is attached to the utility's service lines, I don't want my
meter damaged or stolen if left outside.
--

Jim Baber

(see my 10kW grid tied solar system at "www.baber.org")

1350 W Mesa Ave.
Fresno CA, 93711
(559) 435-9068
(559) 905-2204 cell

Email [email protected]

 

[john]

We now have several examples on grid connected PV system of 1 to 3kWp
where there is a difference in the number of kWh measured by the
electronic meter installed by the utility, and the inverter.

for example:
electronic utility meter / SMA SWR700 inverter reporter by a Sunny Boy
Control

june 03: 128 / 120
july 03: 135 / 120
september 03: 108 / 100
october 03: 57 / 50

the values are quite close, so it is hard to say which value is most
likely to be correct - and asking the utility to check their meter is
quite expensive, if the meter is ok.

And in another example, this time with a Gridfit 2200 (Total Energie)
inverter, the inverter reports nearly twice as much production as the
electronic utility meter (102kWh/fortnight as opposed to
66kWh/fortnight). In this case, the utility says that they have checked
the meter and that it is properly calibrated, BUT a quick simulation
shows that the production reported by the inverter is the most likely.

We also have seen several systems with both disk meters and electronic
meters (one private and one belonging to the utility) with, again, a
difference in favour of one or the other, generally a difference of 1 to
10 kWh/month

Until recently most systems here relied on disk meters for monitoring
production values, so we don't really have much experience with
electronic meters.

Could there be a problem with the inverters' built-in meters? Or are
electronic meters notoriously unreliable? Maybe disk meters have a
tendency to undervalue?

Any experience, insight, pointers etc would be helpful.
thanks!

Melodie

the differance is your losses from the conversion of dc to ac. quit
using the inverter and you won't have that anymore.