SolarEdge minimum number of active panels to produce?

[inetdog]

... but have a hard time believing the optimizer can "boost" the voltage if the panel is not receiving enough sunlight to activate a normal Voc.

That statement does not seem to make any sense.
If the available I from the panel is not enough to operate the electronics of the optimzer at the input voltage the optimzer requires, the optimizer cannot run.
What that has to do with "activating a normal Voc" is beyond me.

[adoublee]

Just from the data shown it looks like an optimizer can go as low as 8V ouput without shutting down and as high as 58V.
Do you have better information on what the top and bottom limits are?

I would think that a partially shaded panel in an otherwise sunny string would require the optimizer output to go below 8V to keep that panel producing. I wonder just where the threshold is.

Spec sheets indicate maximum output voltage. 60V for most, 85V for the P405 for thin film panels.

I believe the output voltage can drop to any value needed to maintain (panel's MPPT wattage) / (inverter's desired string voltage), but have a hard time believing the optimizer can "boost" the voltage if the panel is not receiving enough sunlight to activate a normal Voc.

[inetdog]

Just from the data shown it looks like an optimizer can go as low as 8V ouput without shutting down and as high as 58V.
Do you have better information on what the top and bottom limits are?

I would think that a partially shaded panel in an otherwise sunny string would require the optimizer output to go below 8V to keep that panel producing. I wonder just where the threshold is.

[adoublee]

Thanks for the data. String voltage appears very much as expected. The dip in the panel DC voltage to around 7V at 10:30am supersizes me a little - especially if every panel in the string did the same. It's hard to believe the MPPT caused that or that everything went dark. Current did not spike during this voltage dip, so not sure how power was maintained and rising through this either as shown on PVoutput. Have you noted/seen if string voltage can exceed the total of all panel voltages (a voltage boost included in optimizers)?

Also, the talk about a "self-sustaining power supply" a la SMA's secure power supply got me wondering what the smallest number of panels could be used to support the 1500W at 120VAC from the smallest HDwave inverter. Specs don't say anything other than needing a minimum 8 panels/optimizers, but you seem to know they don't lock out operation with only 6 connected.

[ButchDeal]

It is nice that everyone has ideas on what happens and how it works etc. They do start up with low light rather easily. Here is a minimal system of 8 modules (Canadian Solar 280M) in NY (Bronx) with a 30 degree tilt to SW (240 degrees).
This particular system actually started up on January 1/1 at 7:30 am with just 4.31 watts total but the photo is from 8:00 am with 35.58watts Layout.jpg




you can also see it on pvoutput: http://pvoutput.org/intraday.jsp?id=...13&dt=20170101


just to make it a little more clear here is a graph of the same day (1/1/2017) showing ONE module (1.0.6) with green being panel voltage and red being optimizer voltage, Blue is Current: Chart.jpg


and just because everyone asks about the DC voltage this one is nice and smooth around 372V with AC at 245V

Chart2.jpg

[adoublee]

Actually it is going to depend on whether the optimizer can get up enough input current to run itself and have some to spare for creating output current.
The output voltage could be as low as 1V, it think, even with 40V in if the input current is limited.
The sunny panel is going to be able to deliver both current and voltage depending on what is needed, but it may not be able to step up the voltage far enough to meet the inverter input minimum on its own. And the shaded panels will not be able to help much.

No, the inverter is not going to produce power when the optimizers put out their "I'm here" 1V. The optimizers change their output voltage as required to create a string of approximately 350VDC for a 240V residential inverter. When doing that, current is what it is which is why their is a low end number of optimizers (have to get the voltage built up) and a top end (have to limit the current that could be created. In a similar way a 480VAC inverter is going to want to see even higher voltages, and requires a minimum of 18 optimizers in a string to build up the voltage.

[inetdog]

In the end it's going to depend on if the optimizer can get enough voltage to put out voltages that add up sufficiently for inverter to produce grid voltage. It takes very little sunlight for a panel to get up to voltage, with current available from panel changing as sunlight intensity changes. They optimizer converts the voltage and current, but needs a certain level of initial voltage to build up the sufficient string voltage.

Actually it is going to depend on whether the optimizer can get up enough input current to run itself and have some to spare for creating output current.
The output voltage could be as low as 1V, it think, even with 40V in if the input current is limited.
The sunny panel is going to be able to deliver both current and voltage depending on what is needed, but it may not be able to step up the voltage far enough to meet the inverter input minimum on its own. And the shaded panels will not be able to help much.

[adoublee]

I have a buddy who is skeptical if you are claiming that SE will produce power if 7 panels are in complete shade and 1 is in full sunlight. Anyone else with SE experience can comment?

In the end it's going to depend on if the optimizer can get enough voltage to put out voltages that add up sufficiently for inverter to produce grid voltage. It takes very little sunlight for a panel to get up to voltage, with current available from panel changing as sunlight intensity changes. They optimizer converts the voltage and current, but needs a certain level of initial voltage to build up the sufficient string voltage.

[foo1bar]

I have a buddy who is skeptical if you are claiming that SE will produce power if 7 panels are in complete shade and 1 is in full sunlight. Anyone else with SE experience can comment?

I have 2 strings of 16.
I regularly have 5 of them getting direct (but poor angle) sunlight in the evening with the other 11 getting no direct sunlight.
and the 5 in the light generate 50W or 100W (280W modules) and the other 11 are at 10-20W.
With that string I have 5 west facing, 4 south facing and 7 east facing. (Actually about 255, 165 and 75 degrees)

The other string in late evening I get no direct sun on all but 4 - and partial shade (2nd floor shadow) on 2 of those 4 and it still looks like I can get production (ie. ~10W each on the ones in shadow, 30W on the partial shade and 50W on the ones in sun)

[dennis461]

..... I am very roof limited and have some major shade issues I am trying to work around.

You really should use the website i listed below to get an idea how much energy your system will produce.
It may not be cost effective, regardless of the system you choose.
While the power optimizers in Solar edge keep power going out on a rainy day (major shade), it is very low, 1/10 or less of normal output.

Try and figure how much money you save each year then determine what year the system is paid off. The compare to

http://pvwatts.nrel.gov/

Dennis
Camden County, NJ, USA
19 SW285 panels
SE5000 inverter
grid tied

[ButchDeal]

I have a buddy who is skeptical if you are claiming that SE will produce power if 7 panels are in complete shade and 1 is in full sunlight. Anyone else with SE experience can comment?

Define complete shade? 7 modules facing east in the evening is diffuse light and it would still produce.

7 modules under foot of snow and one exposed, not going to produce.

[ButchDeal]

I am sure that ButchDeal was talking about diffuse or partial shading of the other panels. And enough current/voltage for the inverter to start is a lower bar to reach than producing useful power.

I agree that total darkness on 7 of 8 panels is a non-starter. [pun intended]

Yes there would have to be diffuse light such as an array of 7 modules facing east with one module facing west, in evening sun. The east would have diffuse light.
If you were to cover 7 modules with cardboard the 8th would not be producing either.

[inetdog]

I have a buddy who is skeptical if you are claiming that SE will produce power if 7 panels are in complete shade and 1 is in full sunlight. Anyone else with SE experience can comment?

I am sure that ButchDeal was talking about diffuse or partial shading of the other panels. And enough current/voltage for the inverter to start is a lower bar to reach than producing useful power.

I agree that total darkness on 7 of 8 panels is a non-starter. [pun intended]

[sigfreund]

I have a buddy who is skeptical if you are claiming that SE will produce power if 7 panels are in complete shade and 1 is in full sunlight. Anyone else with SE experience can comment?

[ButchDeal]

How hard would it be to add some panels later to a Solaredge system? If I went with a 7.6kw inverter but only initially went with 18 x 280w panels (5kw) would it be difficult to add another 8 panels. Would it be better to go with a 5kw inverter since 24 x 280w would be less than the recommended "oversize" limit for the inverter?

It is trivial as long as each string is at least 8 modules and under the max wattage. I would go with the SE7600 as it is the most capable and future upgradable to storedge.