proper string inverter sizing

[charles2]

I think the problem is 13 panels are on one roof face, and 14 are on the other. 60 cell panels would have more flexibility, but may not use the available space on the roof as efficiently.

You are exactly correct. So what's the best I can do?

What if I drop the odd panel, so that I'll have 2 strings of 6 panels each on one slope and 2 strings of 7 panels each on the other slope?

[charles2]

2 strings of 6 panels each on one slope and 2 strings of 7 panels each on the other slope?

Does anyone agree or disagree that what I've proposed is the best solution?

[sensij]

Does anyone agree or disagree that what I've proposed is the best solution?

Did you read my previous response? Min string length is 8 for that inverter. Options I see:

1) Use SolarEdge or microinverters

2) Use two SMA inverters (SB4000TL-US-22), which have a 175 V MPPT minimum and would be fine with strings of 6. In that case, you would have 4 mppt inputs. On inverter 1, input 1, you could have 6 panels, inverter 1 input 2 7 panels, inverter 2 input 1 7 panels, inverter 2 input 2 7 panels.

3) Use a Fronius Primo 7.6 inverter. It has a min MPPT of 150 V, or maybe as low as 80 V. Then split into two strings of 6 into one input, and two strings of 7 into the other input, dropping the odd panel as you suggested.

4) Use a different panel, preferably 60 cell, that operates at a lower voltage as has more flexibility.

[charles2]

Did you read my previous response?

I did but guess I'm having trouble paying attention. Thanks for the other options.

[solarix]

I guess I don't understand why SMA even makes a 7.7kW inverter if it can only handle 22 panels?

I'm doing a SB7700 right now that has 12 260W panels facing south and 2 strings of 8 facing west and another string of 8 facing west. It takes a DC combiner to put all three strings of 8 in parallel on one inverter input, but the alternative was 2 inverters. Putting the east and west strings on the same input is a good trick because they will lose very little productivity and still allow doing this roof with one inverter. My quote won out over the competition because of this.
However, your point is well taken that using large panels on string inverters is again becoming pretty restrictive. On the older inverters, with narrow voltage operating ranges, there was a time when the panels sizes got up around 200W that strings had to be 10, 11, or 12 panels - period. Now with the >300W panels, we're running into these problems again... We find that the mainstream panels sizes of ~260W are better in that they have less string sizing problems, give us more flexibility fitting them onto most roofs, are about the limit that can be handled by one installer carrying them, and are better value. Of a course the high efficiency panels get around some of these problems but cost way more.

[sensij]

I'm doing a SB7700 right now that has 12 260W panels facing south and 2 strings of 8 facing west and another string of 8 facing west. It takes a DC combiner to put all three strings of 8 in parallel on one inverter input, but the alternative was 2 inverters. Putting the east and west strings on the same input is a good trick because they will lose very little productivity and still allow doing this roof with one inverter. My quote won out over the competition because of this.

Interesting. Clearly the Isc of the three 8 panel strings will exceed the 19 A Isc limit, and even with E/W orientation and knowing the two faces won't peak simultaneously, I'd be worried that three strings would exceed the 18 A input limit. Are you just counting on the inverter moving away from MPP and limiting the power in that case? If you don't mind sharing, what zip, tilt, and azimuth? I'm curious to see what PVWatts/SAM would show for this.

[solarix]

This is a hot Phoenix site with 6/12 pitch and 60-cell panels with Vmp= 8.53Vdc so there is never a time when both the East and West sides peak out. There may me instances when the inverter will have to throttle back a bit, but this design comes straight from the advice of a SMA app. engr and it checks out fine on their string sizer.

[sensij]

This is a hot Phoenix site with 6/12 pitch and 60-cell panels with Vmp= 8.53Vdc so there is never a time when both the East and West sides peak out. There may me instances when the inverter will have to throttle back a bit, but this design comes straight from the advice of a SMA app. engr and it checks out fine on their string sizer.

Thanks for the response. The 6:12 pitch helps. I'm glad to hear your ability to understand the system and propose a creative design was actually rewarded with some business!

[charles2]

Use two SMA inverters (SB4000TL-US-22), which have a 175 V MPPT minimum and would be fine with strings of 6. In that case, you would have 4 mppt inputs. On inverter 1, input 1, you could have 6 panels, inverter 1 input 2 7 panels, inverter 2 input 1 7 panels, inverter 2 input 2 7 panels.

Since 325*14 = 4550, wouldn't you want to go with 5000TL-US-22 rather than 4000TL-US-22?

Also, I'm very confused by SMA's string sizing spreadsheet at SunnyBoyTLUS-22_QDR_v1.10.xlsx. It seems to indicate that the modules per string can vary from 5 to 13.

[sensij]

I don't think the spreadsheet is accounting for the difference in MPPT range between the different models of inverter.

4550 / 4000 = 1.14... this is a good loading ratio. The panels will typically produce less than their rating, so you won't really be missing out on any power.

You might actually be better off going for a pair of the SB3800TL, with a 1.20 ratio. You can use a 40 A breaker to protect two of those, which in some cases will help avoid the expense of a panel upgrade, in addition to the savings from going with the smaller inverter.

You can use PVWatts and vary the DC-AC ratio in the Advanced Parameters to get a feel for how much energy you might miss out on by undersizing the inverter. Until the ratio gets to 1.25 or more, with a good array orientation, it usually isn't much.

[charles2]

You might actually be better off going for a pair of the SB3800TL, with a 1.20 ratio. You can use a 40 A breaker to protect two of those, which in some cases will help avoid the expense of a panel upgrade, in addition to the savings from going with the smaller inverter.

Agreed. But I'm still wondering if it's possible to make a single 7.7kw work well. How do you like dropping the odd panel to reach 26, then have 2 strings of 9 panels and 1 string of 8 panels? I guess I'm asking, how much do the different panel orientations mess things up?

[charles2]

I think the problem is 13 panels are on one roof face, and 14 are on the other. Splitting them into parallel strings doesn't work great. A string of 6 * 37.0 Vmp = 222 Vmp, which is too low for this inverter (low MPPT range is 270 V).

sensij, are you sure you are looking at this correctly? Isn't the Rated MPPT voltage range the sum total of both strings on the MPPT? For example, suppose I have 2 strings and each is running at 150V. They both meet the start up Voltage requirements (125V/150V) per string. They combine for 300V total, putting them above the 270V MPPT range minimum. Can you confirm or disconfirm?

[sensij]

Agreed. But I'm still wondering if it's possible to make a single 7.7kw work well. How do you like dropping the odd panel to reach 26, then have 2 strings of 9 panels and 1 string of 8 panels? I guess I'm asking, how much do the different panel orientations mess things up?

Different panel orientations on a single string will mess things up. The current will be limited to that of the lowest performing panels, unless they underperforms by enough to just get bypassed (through their diodes) completely.

sensij, are you sure you are looking at this correctly? Isn't the Rated MPPT voltage range the sum total of both strings on the MPPT? For example, suppose I have 2 strings and each is running at 150V. They both meet the start up Voltage requirements (125V/150V) per string. They combine for 300V total, putting them above the 270V MPPT range minimum. Can you confirm or disconfirm?

Strings in parallel do not sum voltage. They only sum current.

[charles2]

sensij, SMA tech support addressed my question today and they say the single 7.7kw inverter will only suffer a slight (1%) efficiency loss if operated below the Rated MPPT voltage range lower limit of 270V. In other words, not enough to justify the expense and trouble of 2 inverters.

As support, they cited data obtained from http://www.gosolarcalifornia.org/equ...-22_240Vac.pdf . I've attached the actual graph they sent me.

Your thoughts?

[sensij]

sensij, SMA tech support addressed my question today and they say the single 7.7kw inverter will only suffer a slight (1%) efficiency loss if operated below the Rated MPPT voltage range lower limit of 270V. In other words, not enough to justify the expense and trouble of 2 inverters.

As support, they cited data obtained from http://www.gosolarcalifornia.org/equ...-22_240Vac.pdf . I've attached the actual graph they sent me.

Your thoughts?

If the SMA tech so goes for it, who am I to say otherwise? The attached CSI data doesn't show below 270 V, but I'll take your word for it that the jpg shows data from this inverter.