2
\$\begingroup\$

Notice below part of electric scheme as designed by solar company:

PV to inverter

On the left you can see inverter used - Delta Solivia 7.6 TL (see Operation and Installation Manual for details) which supports up to 4 strings of solar panels.

Legend on the right side:

  • MP2 and MP3 are different roof surfaces (slopes). This is important for the whole story below: MP2 and MP3 surfaces are pointing towards different azimuths (so peaking energy production at different times of the day/ peaking voltage at different times);

You can see that the design has only two strings used (marked as "4" and "6" on the diagram):

  • String "6" (the upper one) has panels from both surfaces (!) - and that's why I created this question - it seems to me a big problem right here as, again, panels on roof surfaces MP2 and MP3 peak voltage at different times of the day so that panels on MP2 and MP3 would only provide voltage of lowest ones of them !?
  • String "4" has panels from surface MP2 (only) but also notice that some other solar panels from that same surface MP2 are also connected to String "6" (probably okay).

Can somebody please confirm if that's a valid concern above? I thought that was the whole selling point of microinverters (which this design doesn't have) - is that each panel can produce slightly different voltage, and so energy production can be combined efficiently. That's very strange (to say the least) that they've used panels facing different azimuths - on one string ..

This will work, but it will not be optimal. What they should have done is: use one or two strings for surface MP2 which has more panels, and a separate (third) string for surface MP3. Does this make sense?

\$\endgroup\$
1
\$\begingroup\$

I actually do the same thing on my micro inverters. I have a vertical dawn facing panel on a wall connected across a normal winter zenith facing panel on the roof.

Panels have a soggy voltage-current curve. You can parallel them. If one is in darkness, the diodes prevent reverse current. (This is debated endlessly, so I checked with an ammeter to be sure no current goes into the dark panel)

The drawback is that for each level of illumination, there is a V-I maximum power point. If two panels are at different illumination (e.g 1/3, 2/3) then they can't be running at the perfect max power point (MPP), so you won't get the max possible from them.

But it doesn't really matter. The one providing most of the power will be close to MPP,

In my case I have 550W of panels connected to a 250W micro invertor. I can't use the max possible power (only 400W due to angle between panels) anyway. The point for me is to increase the time I get any power. This is about the hostile tariff regime here, and also the relatively low cost of an extra panel relative to system cost.

\$\endgroup\$
  • \$\begingroup\$ Thanks for the answer! Yep, that's what I was thinking too - it will work, but it will not be optimal. What they should have done is: use one or two strings for surface MP2 which has more panels, and a separate (third) string for surface MP3. Does this make sense? \$\endgroup\$ – Tagar Jul 23 '18 at 15:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.