Solar panel choice advice for carport (flat) roof ?

[Bjorn Teani]

Hi Dacian, hi all !

So, after a first successful project with 12 330W solar panels (placed vertically facing south) feeding 2,4kAh battery bank (see https://groups.google.com/g/electrodacus/c/JWPr3Rd6Npg), I plan to use solar panels to cover my almost flat carport roof.

Why use solar panel on a flat roof ? To make it short : because of local legislation, a roff over 20m2 has to be of the same (double) inclination as the rest of the houses around OR to make it a "Green Roof". But, considering everything, a green roof would be as expensive as directly using solar panels to cover the roof. And legislation tend to favour more and more solar panel install… so I’m trying my luck with the permit.

My plan is to use 24 solar panels as roof for my carports. Inclination would be 5° from horizontal, facing south and 30m away from my SBMS0 and the battery.

Now I need your help for the correct choice of the solar panels as I plan to link the 24 new panels to my already working SBMS0 install in the future making the first array of 12 panels the small array and the 24 new panels the large array.  I would size up my battery bank and add a bunch of new heating elements to the diversion setup.

The solar panels I found at an interesting price are all of higher voltage an amps than the one I bought a few years ago.

So, long story short, what panels would you chose Dacian :
(I’m focusing on double glass panels only as it make sense for me to have the same material on both side… but I might be wrong to put importance on that).

1 - TrinaSolar Vertex S+ 410Wc 120 cells N-Type - Isc 12,23A / Uoc 42,4 V / Impp : 11,48 A / Umpp 35,7 V - 110€
2 - TrinaSolar Vertex S+ 435Wc 144 cells N-Type - Isc 10,64 A / Uoc 51,8 V / Impp : 9,99 A / Umpp 43,6 V - 115€
3 - AIKO SOLAR N-Type ABC Dual-glass - Full Black 445Wc - / Isc 13,68 A / Uoc 40,15 V / Impp 13.18 A / Umpp 33,79 V - 135€

The "full Black" of 3) would be a plus, but a 480€ "plus" I do not know yet. Amps would be on the higher side, but as the panels would lay almost flat, I think the 12 DSSR50 should be able to handle any of these, correct ?

Bonus question : is it a problem to have the small and large arrays on different voltage levels ?

cheers,

bjorn

[Dacian Todea (electrodacus)]

Bjorn,

Panels 1 and 3 are OK Panel 2 has open circuit voltage that is a bit to high so it will not work as efficient and it is also likely larger than the other two panels.

Based on spec and cost the first panel seems like the best choice.

Slightly different voltage is not a problem even on the same input so directly in parallel but as long as panels are on different input as each input has his own ideal diode your can even have 60 cell and 72 cell panels on same DSSR50 if each panel is connected to his own separate PV input.

[Bjorn Teani]

Thanks Dacian !

And do you think I could get away with using (10x) DSSR20 (limiting my new install to 20 panels) as I would need diversion and integration with my existing setup (DEXT16 and  6x DSSR20). In ideal conditions I would be over the 20A limit, but as the panels would be laying almost flat... ? I would reach the 16X max DSSR20 on the DEXT16 and the SBMS0 would be happy and me too.

Or do you recommend I find lower spec solar panels ?

And as far as I understand, the DSSR50 offering no "automatic" diversion unless paired with the DMPPT100, but as far as I can understand, this (DMPPT100) will not work with an SBMS0 already handling a DEXT16... Or will it ?

The last option I'm considering is switching over to a full DSSR50/DMPPT100 setup for my current install of 12 360W pannels and the 20 (or 24) new 410W panels... This solution would also permitt to add even more panels in the future.

[Dacian Todea (electrodacus)]

For DSSR20 you will need to find panels that have Isc around 10A or below so maybe the 2 panel will work but not the other two as those in some conditions could output 15A each and 30A will be to much for the DSSR20

You can use pair of two DSSR50 one for battery charging and one for diversion and they can be controlled by the DEXT16 but it will be more expensive than DSSR20's

I can not promise when the DMPPT100 will be available but that will not have any problem's working with the rest of your setup it can even work fully independent of SBMS0.

Will you have space for more panels in the future ? There is still a limit of 12x DSSR50 as that means 600A with 720A peak and 750A is the measurement limit on the SBMS0

You will need a second SBMS0 for a PV array larger than around 20kW.

You can add a second DEXT16 if that is needed limit is the same 600A STC to allow for 20% with edge of cloud effect.

[Bjorn Teani]

Hello Dacian,

So I finally bought 20 of these :  AIKO-A445-MAH54Db

STC : Voc : 40V / Isc : 13,7A

NOCT : Voc : 38V / Isc : 11A

These panels will be laying almost flat (5°) but I will follow your advice and go for the DSSR50.

I'm about to draw a update to my initial setup schematics, but (correct me if I'm wrong) I will :

- double my battery capacity to 2P8S (310Ah cells)

- the 12 panels I have today will move from 2+4 dual array configuration to a single 6 DSSR20D array on one DEXT16

- the 20 new panels will be hooked on 5+5 DSSR50 controlled for charging or heating by my other DEXT16

Right ?

And as I do not know witch one of the array will produce the highest max (20x 445W laying flat, or 12x 330W vertical) is the notion of array 1 or 2 of any importance ? Does the SBMS0 consider Array 2 should be bigger than Array 1 or does it adapt on the fly depending on output ? (I think it does, but again, I'm no Dacian ;-))

Last bonus question : my cells are actually with the terminals on the top. With another 8 cells to add in // I consider stacking them one on top of the other, on their side, with the terminals facing me. Does it have any importance on how they are laying ? Should I keep them upright ?

[Dacian Todea (electrodacus)]

Bjorn,

The SBMS expects that array 2 is the larger array (the one controlled by type 6).

So the SBMS will default on Array 1 until that array output is 50% of whatever you set as the max current limit in the DMPPT menu.

When array 1 is below 50% it will disconnect array 1 and switch ON the array 2 and if that array current is 66% less than set limit it will also switch ON the array 1 so that all panels are connected to battery charging.

I will assume that the 20 new panels will output more in most cases than the old 12 panels so the 20 panels array should likely be the array 2.

You will need to reconfigure the battery so there are groups of two parallel cells.

As for cell orientation it depends on cell type but assuming they are the common blue rectangular cells they need to either be upright or if not then on the narrow side. They can not lay on the wide side as that means some of the plates will remain dry (of electrolyte) as the capillary effect can not work in those conditions. 

[Bjorn Teani]

Thanks for the clarifications about how the SBMS0 handels the two arrays.

I assume my other assumptions are right as you havn't commented them. I'll come back with a propre drawing for your blessing ;-)