diy solar

diy solar

Does series/parallel need to perfectly factor number of panels?

slipperysam

New Member
Joined
Feb 2, 2021
Messages
22
Hi, I can't seem to figure this out after reading the beginner resources. I see that everyone seems to factor the number of panels N into, say, S series x P parallel. For example if you have 6 panels, people consider either 2S x 3P or 3S x 2P.

Well, I'm putting solar panels on top of a 34' bus, and my space constraints limit me to 7 panels. Of course 7 is a prime number. So, from my initial reading it looks like I need to decide whether or not to wire all 7 panels in series (pretty much a non-starter with partial shading issues), or 7 panels in parallel which will likely require some really thick gauge wires. It's also possible that I just install 6 panels if I must, but I'd rather not miss out on an extra 400W.

(Each panel will probably be a residential 400W panel, e.g. Voc 48.55, ISC 10.64, Vmpp 40.77, Impp 9.81 - haven't picked particular panels but this is a good example of what we've been looking at.)

So, am I understanding this correctly? If so, can someone explain to me why I can't mix different volts & amps, for example running 7 panels as 4 parallel arrays of 2s, 2s, 2s, 1s?
 
Welcome to the forum.

All 7 in series or all 7 in parallel.

7 in series pushes Voc too high for your typical charge controller. 7 in parallel may result in excessive wiring losses - 37V @ 70A is pretty stout.

Your battery voltage also matters. Your series Vmp needs to be at least as high as 1.5X nominal battery voltage, i.e. 36V for a 24V battery, 72V for a 48V battery.

The 1S panel in parallel with the 2S panels will pull the 2S panel voltage down to 1S Voc. Performance of the 2S strings would be substantially less than 800W - far more loss on the 2400W panels than you'd lose by just omitting the 7th 400W panel. Your TOTAL output on a 2S3P + 1S panel configuration would likely be less than 2100W. Without the 7th panel, your 2S3P array has a potential of 2400W.

If you really want the 7th panel, just get another charge controller with only that panel connected. Connect the other six in a 2S3P array to the first charge controller.

What is your battery voltage?
 
Hi, I can't seem to figure this out after reading the beginner resources. I see that everyone seems to factor the number of panels N into, say, S series x P parallel. For example if you have 6 panels, people consider either 2S x 3P or 3S x 2P.

Well, I'm putting solar panels on top of a 34' bus, and my space constraints limit me to 7 panels. Of course 7 is a prime number. So, from my initial reading it looks like I need to decide whether or not to wire all 7 panels in series (pretty much a non-starter with partial shading issues), or 7 panels in parallel which will likely require some really thick gauge wires. It's also possible that I just install 6 panels if I must, but I'd rather not miss out on an extra 400W.

(Each panel will probably be a residential 400W panel, e.g. Voc 48.55, ISC 10.64, Vmpp 40.77, Impp 9.81 - haven't picked particular panels but this is a good example of what we've been looking at.)

So, am I understanding this correctly? If so, can someone explain to me why I can't mix different volts & amps, for example running 7 panels as 4 parallel arrays of 2s, 2s, 2s, 1s?
The issue is not the panels themselves but the way the solar controller works. Your best option is multiple controllers instead of one larger controller. Multiple controllers can feed a battery for charging. Think of fit this way, once you start paralleling it is just like putting battery cells in parallel you can't have one panel at 96 V (2S configuration) paralleled to a single panel at 48V.
 
Okay, thanks for helping me understand. This is all quite new to me and I haven't had to mess with electric wiring theory since I was an undergraduate! It sounds like I should have a 6-panel array, and optionally add another panel with a separate controller, if we deem the extra wattage worth the expense of another controller. This might also be good for the future if I wanted some ground-deployable solar panels to hook up to the second controller.
7 in parallel may result in excessive wiring losses - 37V @ 70A is pretty stout.
In this application, we may be able to have the wiring from the panels (after they are wired together) to the charge controller be as short as ~4' or so. Perhaps that makes 7 in parallel more feasible?
What is your battery voltage?
24V battery bank
 
Thinking outside the box a little, why not 2 different arrays? Perhaps an array of 4 panels and an array of 3 panels which feed 2 different, smaller SCC's?
Of course you could do 6 panels and 1 panel as well. You get the idea.

And as the OP mentioned, provisions for hooking up a ground-deployable array cold be made as part of this scheme.
 
Look at going with two smaller charger controllers- one set of 2s2p and one set of 3s.
 
2400W/24V = 100A - you need a beastly SCC just for the 6.

A second SCC for a deployable ground array could really enhance harvest. Flat panels on a roof always disappoint. 12V panels would also be an option. 2nd SCC with the 1 400W roof panel attached. You could then parallel 2SXP 12V panels as a lightweight deployable ground array. There would be a very small performance penalty due to the slight Vmp mismatch, but you're talking 5% or so. More than offset by the capability to optimally face and tilt the ground array.

Ground arrays are also nice if you want to keep the bus in the shade.

@Forbisher called it. 3S 72 cell panels on any typical 150V controller is a no-no unless you never see below 50°F. 3S would limit you to 60 cells panels.
 
72 cell panels on any typical 150V controller is a no-no unless you never see below 50°F. 3S would limit you to 60 cells panels
Hm. Can you point me to a resource that would explain why the number of cells comes into play here? I just noticed the panels I was looking at were 144 cell, but the manufacturer (Trina) has some similar spec-ed panels in 72 cell.

Thanks everyone for your ideas! I'm thinking a 2s2p array to a 150/60 MPPT controller and then a 3p array to a 150/45 MPPT controller is the new plan. I plugged in the specs of these panel arrays to the Victron calculator, and it suggested these controller sizes to me, so hopefully I'm on the right track.
 
# of cells refers to the # of series cells that dictate Voc/Vmp. Vmp is about 0.5V per cell. If the 144 cells panels are Vmp ~36V, the panel is really a 72S2P panel.

Victron manual explicitly addresses that 72 cell panels run a risk of over-volt in 3S where 60 cell panels can be safely wired in 3S. I made this mistake with my Victron 150/100, and due to laziness and the need for a second charge controller anyway, I left my array 3S and upgraded to the 250/100, which can handle my full 6kW array.
 
2400W/24V = 100A - you need a beastly SCC just for the 6.
So, we've been re-thinking the possibility of just doing all 7 in parallel. It will take a bit of research to figure out just how thick our wires need to be, but as far as the SCC goes, the Victron calculator tells me that the MPPT 150V/100A will work in this scenario. Which makes sense to me, since 1 * 50V < 150V and 7 * 10A = 70A < 100A.
But your comment calculated the amps differently... Taking the total panel wattage and diving by the battery voltage, for 7 panels, would be 2800W / 24V = 117A.

Which calculation is actually relevant here for choosing the SCC?
 
The panel current and the battery charge current are two separate factors. The 150/100A MPPT is limited to 100A of CHARGE current to the batteries. It is also limited to 70A of PV input current per the data sheet. Both limitations are important.

The 2800/24 = the maximum charge current needed for a battery at low SoC.
The 150/100 has an absolute peak power output of 2900W. 2800W/100A = 28V, i.e., you will only get peak 2800W @ 28V and higher.

With your 150/100 and 2800W of solar, your charge power is limited by your voltage, so at 24V, you'll get 2400W, 25V 2500W, etc. This is the 100A limitation of the charge controller.

Your Isc is 10.64. 7 * 10.64 = 74.48A. Victron explicitly states the 70A input limit should not be violated. This prohibits you from using those 7 panels in parallel. I missed this in your initial post. Sorry for dangling the possibility in front of your nose.

IMHO, 2S3P on a 150/100 SCC and anything else you decide on a separate SCC.
 
Back
Top