EG4 18kPV : Parallel 2 panels into series to combat shade

[inowhavesolar]

Per title, I have an EG4 18kPV and 8 x 370W panels. (Vmp 34.2V, Imp 10.82A, Voc 41.0V). The spec sheet on the EG4 says the following (it has 3 MPPTs):

* Nominal Input Current : 25/15/15A
* Max Short Circuit Input Current : 31/19/19A
* DC input voltage range: 100-600 VDC

Right now I have the panels in one long series. This is working fine, until the sun starts setting behind a tree which casts shade panel by panel. One evening, as an experiment, I disconnected the currently shaded panel, having 7 in the series. The incoming watts to the inverter increased by almost 30%. I'm assuming this is what optimizers do: monitor each panel and if a single panel is outputting below the average of the others, bypass it.

I read that panels in parallel effectively function independently, reducing the impact of shaded panels to the overall system. Given the restrictions on my inverter, I can't parallel all 8 panels, as that would be over 86A vs the inverter limit of 25A on MPPT #1.

What I was thinking then instead to do is parallel every 2 panels (21.64A) and then series each group together. This should keep me under the inverter limit on Amps.

Will this help maximize my output? Are there other options?

 

[EG4_Eric]

With 4 panels in series you would be very close to the operating voltage of the MPPT of 140VDC-500VDC. I would recommend using optimizers or removing the shading.

 

[inowhavesolar]

With 4 panels in series you would be very close to the operating voltage of the MPPT of 140VDC-500VDC. I would recommend using optimizers or removing the shading.

How are you getting that? Wiring in series keeps AMPs same, but adds VOLTAGE. 8 panels * 34.2V = 273.6V, which is well below the EG4s max input of 600 VDC. 4 panels in series would be 1/2 that. Can you please clarify your statement?

 

[EG4_Eric]

What I was thinking then instead to do is parallel every 2 panels (21.64A) and then series each group together. This should keep me under the inverter limit on Amps.

From reading this it seemed like you were going to do 4 panels in series with 2 parallel strings. Would you be able to show a wiring diagram on how you're planning on connecting these.

 

[inowhavesolar]

Yes, I said "parallel every 2 panels and then series each group together":


But now that I think about it, this won't help with the shade because when the right-most panel gets shaded, it's still in series with the panels before it and therefor will bring down overall output. And I can't series every 2, then parallel those 4 groups because that'll be too much AMPs for my inverter. Darn. Are there other options besides optimizers, or cutting down the 30 y.o. tree?

 

[Buff]

Another configuration would be to series the left most 4 and series the right most 4 then parallel the two "halves" together. As the sun starts shading panel by panel, the one "half" would be reduced but the non shaded "half" would not. At least until more than half the array is shaded. This would probably be the best configuration without optimizers or cutting the tree down.

I'm in the same boat but only have 6 panels and 3 in series do not have the volts to meet the 120v min.

 

[y0bailey]

I'm currently running this same experiment.

4s2p vs. 8s for my 8 panels with East to West shading (but my Vmp is right around 40v so I have more wiggle room than you).


I have the RIGHT 4 panels together and the LEFT 4 panels together so one set is in full sun while the other is dealing with shading. I would highly advise this vs. your every other panel configuration.

To easily swap between series and parallel I just have these Yconnectors: (https://www.amazon.com/gp/product/B07H7J9PNL/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1)

I've run some extra wiring so it's just a few disconnects and I can swap between the two configurations, but I can still use my one run of cabling into the house/inverter (make sure you have the proper wire gauge).


So far the 4s2p setup is performing better than the 8s, and my voltage hasn't dropped below the cutoff for the inverter. I'm tracking it now that the temps are increasing, but so far so good.