Greetings from cloudy Frederick, MD

[Naptown]

Have they provided any of the information to you regarding string and inverter layout?
It is something I would insist on if I were the purchaser and these type of questions came up.
If the system is up and running is there any variance between inverter outputs on a daily basis. In other words is one inverter putting out more or less than the other 2 by more than say 1 or 2%

[Naptown]

I had a bit of time today to run some numbers on your inverters and panels.
The inverter you specified in your post was a SB3000US If this is the correct inverter than the max input voltage is 500V
If by chance it is a SB300HFUS-30 the input voltage is 600V
I would hope it is the latter as the former is worse.
The sweet spot as far as configuration on the SB3000HFUS is 13 modules on 1 string. This falls within the voltage and input maximums.
You can add one more module to the string and still be within the sweet spot array to inverter size (+/- 20%)
However and this is the kicker and the reason I have been on this. When the 14th module is added to the string the voltage open circuit on the coldest record day in Frederick is -13C(8.4F) the voltage is over 600. It isn't till we get to -10C (14F) that the voltage drops below 600 (598 according to the SMA calculator). Now if by chance they did two strings of 7 modules on that inverter it would work and the inverter would not fry but the voltage will not be high enough to start when things get warm out due to low voltage.
So here is what can happen. During the winter if the temperature is below 14F you run the risk of frying the one inverter with the 14 modules. And here's why It doesn't take much light for the modules to start producing voltage. They will produce full voltage in dim light not much in the way of watts but full voltage. In the morning your inverters are sleeping and disconnected from the array. So the modules are in Once they wake up the voltage will drop but it is during the period that they are waking up that the problem occurs.
So as long as the temperature never falls below 14 degrees no problem.
If by chance it goes down below that then you have a problem.

If they are using the SB3000US inverter the situation is worse If the temperature falls below Well never mind they would have fried already

[websterize]

Thanks, Rich. Have not seen specs on how the array was strung, but one of the inverters does seem to be running higher than the other two, so I suspect it's 13 +13 + 14 for the 40-panel array. (All 40 panels are hot ) You can see real-time data on the array at http://tinyurl.com/7ess5fp. We've had a few 60+ kW harvests. The three inverters are in fact the SMA 3000US, with a max DC voltage of 500 V.

The electrician is scheduled to revisit the house tomorrow to add code signage around the inverters, and the electrical inspector will be out Friday. I'll be sure to raise your concerns. If all passes, we'll then submit our application to the local utility for a net meter and hope to receive by end of April.

All the best.

[Naptown]

Thanks, Rich. Have not seen specs on how the array was strung, but one of the inverters does seem to be running higher than the other two, so I suspect it's 13 +13 + 14 for the 40-panel array. (All 40 panels are hot ) You can see real-time data on the array at http://tinyurl.com/7ess5fp. We've had a few 60+ kW harvests. The three inverters are in fact the SMA 3000US, with a max DC voltage of 500 V.

The electrician is scheduled to revisit the house tomorrow to add code signage around the inverters, and the electrical inspector will be out Friday. I'll be sure to raise your concerns. If all passes, we'll then submit our application to the local utility for a net meter and hope to receive by end of April.

All the best.

In that case they could have strung 2 inverters with one string of 11 Right in the sweet spot of the inverter Max input is 2.86KW inverter is rated for 3KW no problem with cold weather.
Now the third inverter is probably strung with 18 modules in two strings of 9 This fits within the parameters for the most part In other words there is no possibility of frying the inverter on a cold morning. However the max wattage input for this inverter with 18 modules is 4.86KW. This is an input of almost a third more. It won't harm the inverter but will clip, meaning it will reduce output power to save the inverter. It will only put out 3000 watts at max power. The excess power the modules are producing are wasted. You did see a higher output on one inverter. Was it a third more than the other two. According to SMA you are losing on average 4% of harvest on this inverter. This will happen more in cooler weather than in the summer.
If anyone else has anything to say or check please feel free.

[Patafil]

Your house looks like it was made to fit that array. Looks really nice and tidy. Hopefully you can get confirmation on the concerns raised by Naptown.

[websterize]

Your house looks like it was made to fit that array. Looks really nice and tidy. Hopefully you can get confirmation on the concerns raised by Naptown.

Thanks much, Patafil. That's basically how I got my wife on board -- had to look nice and tidy. (That and she was sick of me droning about solar.) Site quality -- azimuth, pitch, no shade -- was too good to pass. Because the array had to go out front, aesthetics were a concern. We worked to get the modules as close to the edges as possible. What we got matched plan drawings. Placed on a corner lot, near a road with commuter volume, the array gets seen, and I'm glad it's not hidden out back. Many friends and neighbors seem intrigued by the notion of PV on their homes.

Rich, the electrician who was part of the installation team stopped by today to attach signs around the inverters and to make grounding tweaks. This was in advance of an electrical code inspection tomorrow. Although he couldn't confirm, the electrician said it was possible that the installation manager (on vacation this week and out of reach), who was actually on the roof plugging in the modules, left one of them unplugged. So we'd have a 10.1 kW system instead of a 10.4. I could be misreading the kW data coming off the inverters on the SMA site (They're the bottom three links at http://tinyurl.com/7ess5fp). Maybe one *isn't* running higher. There's a lot of surface up there, and I suspect not all modules are getting equal light. Again, shade isn't an issue.

If 39 panels, not 40, were plugged in, does the cold-weather problem with the 14-module inverter go away? (13 13 13 instead of 13 13 14.) Max voltage on each of the three 3000US inverters is 500 V. Each module is 260W.

[Naptown]

Not on those inverters. If it were the sb3000HFUS that are rated at 600V yes 13 is OK but the sb3000us if there are 13 on a string should have fried already.

[websterize]

The array failed electric inspection today. I've asked AtisSun folks for advice. Great guys, and I'm sure there's an explanation. Still, a major bummer. Thanks again for feedback.

After three weeks online, all seems well with the inverters. The array hit 1MWh total today (data at http://tinyurl.com/7ess5fp). Free juice for the power company, I guess. No net meter until it passes electric. More to come.

[Naptown]

The array failed electric inspection today. I've asked AtisSun folks for advice. Great guys, and I'm sure there's an explanation. Still, a major bummer. Thanks again for feedback.

After three weeks online, all seems well with the inverters. The array hit 1MWh total today (data at http://tinyurl.com/7ess5fp). Free juice for the power company, I guess. No net meter until it passes electric. More to come.

What did the inspector fail you for?
70% of the time it is lack of knowledge on the inspectors part.
The other 30% is usually for dumb things.
Did the inspector ask for string calculations?

[Mike90250]

If it failed, why is it still on line ? You are only allowed to power it for testing till it's approved.

[Naptown]

If it failed, why is it still on line ? You are only allowed to power it for testing till it's approved.

he has been previously told.
he could also be running his electric bill up running it. Many of the meters that look analog are actually reading a magnetic impulse from the wheel. Wheel spins backwards meter still counts up.
his house may just be new enough to have one of these meters.

[websterize]

What did the inspector fail you for?
70% of the time it is lack of knowledge on the inspectors part.
The other 30% is usually for dumb things.

Rich, he appeared to echo what you have said about the inverter math not making sense, and he mentioned the cold-weather volt inverter issue -- problems could occur once the temperature drops to 30 degrees F, higher than the 14 F you said would fry the inverters. He wrote three comments on the electric report:

"1- String size over 500 Vdc per inverter" [He noticed the placard reads 600Vdc, but the inverter nameplate says 500 V max.]
"2- 3,380 watt per string is more than 3,000W max inverter rating" [I think he's assuming 13 x 260W; one string could be 14 x 260. There are 40 x 260W modules total.]
"3- 50 amp backed is more than 20 percent bus rating." [No idea what this means, but it sounds ominous.}

Did the inspector ask for string calculations?

No, but I said there were most likely 13 or 14 modules per string. I have not seen the stringing calculations from the installer.

[websterize]

he has been previously told.
he could also be running his electric bill up running it. Many of the meters that look analog are actually reading a magnetic impulse from the wheel. Wheel spins backwards meter still counts up.
his house may just be new enough to have one of these meters.

I pulled last month's electric bill, which showed a 30kWh daily average. Since then, with the array online while we await electric inspection and installation of the net meter, our daily kWh average has nearly doubled. You were right, Rich. Array is dark until we get the net meter.

[Naptown]

Rich, he appeared to echo what you have said about the inverter math not making sense, and he mentioned the cold-weather volt inverter issue -- problems could occur once the temperature drops to 30 degrees F, higher than the 14 F you said would fry the inverters. He wrote three comments on the electric report:

"1- String size over 500 Vdc per inverter" [He noticed the placard reads 600Vdc, but the inverter nameplate says 500 V max.]
"2- 3,380 watt per string is more than 3,000W max inverter rating" [I think he's assuming 13 x 260W; one string could be 14 x 260. There are 40 x 260W modules total.]
"3- 50 amp backed is more than 20 percent bus rating." [No idea what this means, but it sounds ominous.}


No, but I said there were most likely 13 or 14 modules per string. I have not seen the stringing calculations from the installer.

To address comments by Inspector
1) This is what I have been telling you Too much voltage on the inverters The max I would go would be 11 modules on one string on each inverter.

2) Well if this is the case It really is not a problem. Inverters should be sized +/- 20% meaning a 3000 Watt inverter is good from 2400W to 3600W In your inverter's case The max STC input is 3750W (from SMA spec sheet)
The reason is because once you get outside the 20% you start to lose harvest. Above the rating and the inverter will clip meaning it will throttle back the input to protect itself. Below the 20% and the inverters work at a lower efficiency. They like to be right up on the peak power for the highest efficiency. Both will result in losses in harvest.

3) As far as the circuit breaker going into the panel goes
You most likely have a 200A service panel
As the code reads the sum of all overcurrent devices feeding a panel shall not exceed 120% of the rating of the buss bars. In addition the feed from the solar array must be located at the opposite end of the buss bars from the main breaker.
In your case there are really three options.

A) reduce the circuit breaker for the solar in the panel to a 40A. The combined output of the three inverters is 39A. This in my opinion is cutting things a bit too close. There is a potential of nuisance tripping.

B) Replace the main circuit breaker with a 150A breaker. This will meet the code requirement however it could also cause the main to trip under certain conditions. (Cold and cloudy heatpump and backup heat running water heater runnning etc)
This solution may also violate the load calculations on the house.

C) Really the only viable option is to do a line tap connection. This is where the solar is tied into the wires between the meter and the main breaker in the panel. Now if per chance they wired the three inverters together into a breaker box on the exterior of the house and then fed to the main panel and tied into the 50 A breaker. (This is most likely what was done as this will also serve as the utility disconnects which the utilities like to see near the meter.) If that is the case then they will have to add another disconnect between the main panel and the outside box. The reason is that this is considered as another service and all service disconnects must be grouped together.

Keep working on getting the string wiring You have a serious problem.

The only solution I could come up with to make this work without having to tear out the inverters that are in place is to add one more and put the array on 4 inverters in strings of 10 modules.
This is slightly below the sweet spot of these inverters and will cause minor harvest losses but it will be a safe installation meaning nothing will fry. With 10 modules per string and inverter the inverter is just beyond the 20% at 21%. Not ideal but not entirely out of the ballpark either.

Now this bring up the next issue which is at that point you will have 4 inverters to replace down the road. My recollection (We don't do many of these small inverters) is that they go for about $1800 each. Which mean down the road you are looking at about an $8000.00 Plus maintenance cost at about year 13. I surely hope they told you about this.
The good news is that SMA does offer extended warranties. They cost almost as much as an inverter but they are available. I would insist your installer provide you with the extended warranties preferably at no additional cost to you.

[websterize]

[QUOTE=Naptown;43583]I had a bit of time today to run some numbers on your inverters and panels.