Series vs parallel

Status
Not open for further replies.
H

hillbilly1

Senior Member
Location
North Georgia mountains
Occupation
Owner/electrical contractor
Doing an install for a friend on a motor home, he has four 100 watt panels, and wants two wired series, and the other two parallel, all going to a single 40 amp mppt controller. He read on the internet that wiring two in parallel gave more charge in low sunlight conditions. I would think this would not work because of mismatched voltages? Or does the controller don’t care? All four panels are the same model.
 
The panels in parallel will have the full voltage across them and so they'll be the ones setting the MPPT (where the tradeoff of the current dropping off vs. voltage at the knee of the I/V curve is optimized for best power out). Therefore the currents from each of the two panels in parallel will be a little lower than that from each of the ones in series but they'll still be relatively close to one another.
However each of the ones in series will have half of the voltage across them, and so the power they deliver will be down by almost 50% from the ones in parallel.
 
Whatever is paralleled needs to be the same. E.g. two strings of two or four individual panels. Any large discrepancy between inputs will confuse the mppt algorithm as synchro described, and most likely not get the most power out of any of it. The controller's voltage range has to support the voltage that you want to put into it. So without knowing the model we can't tell you which of those examples will even work, let alone which is best.
 
Also what he 'read on the internet', if true of any equipment, would be dependent on that particular equipment, especially the controller and how its algorithms handle that voltage range.
 
Yeah, he also wants to run the A/C! He has ordered a 3000 watt pure sine inverter, but that’s going to suck down his four batteries pretty quick with that load. Even though it’s not near enough, I got him talked into eight hundred watt panels, two sets of four in series, then paralleling the two sets into controller. It accepts up to 100 volts. The controller is a Renovy Rover
 
hillbilly1 said:
Doing an install for a friend on a motor home, he has four 100 watt panels, and wants two wired series, and the other two parallel, all going to a single 40 amp mppt controller. He read on the internet that wiring two in parallel gave more charge in low sunlight conditions. I would think this would not work because of mismatched voltages? Or does the controller don’t care? All four panels are the same model.
Click to expand...

If you want to combine sources in series, they all must produce the same current.
If you want to combine sources in parallel, they all must produce the same voltage.

Draw out the series/parallel arrangements, such that series strings are horizontal rows, and each module within the string is a vertical column. Or vice-versa if you prefer. If you directly combine panels in series and parallel, the whole electrical zone must be a rectangle on a drawing like this. Not necessarily a rectangle in the physical layout, but rather a rectangle in the logical layout. By electrical zone, I mean the group of panels associated with an individual power processing device, in this case a charge controller.

In the event that you deliberately mismatch the strings, the MPPT tracker and panels will have to find a compromise point on the respective current-voltage curves of the panels, in order to force current to be the same in series, and force voltage to be the same in parallel. It could be far away from the sweet-spot on the curve, that the MPPT tracker aims to achieve, which means all panels involved in the electrical zone lose performance.
 
Each set of four will be in series, then the two sets will be parallel. Unless I parallel all, the voltage would be too high to series all. The kit he got with the four panels, has only enough cable to series them. He is buying four more of the same panel.
 
LarryFine said:
Won't the panels in series "set" the voltage?
Click to expand...
The p-n junctions in solar cells behave as current sources up to a certain voltage , and the amount of output current is proportional to the amount of light that they receive. The limit on voltage is caused when the "diffusion" current (which is the current that flows into a normal diode when you forward bias it) becomes significant compared to the photocurrent produced by the p-n junction that we want to flow out of it. Effectively this effect is stealing away current we want. When you open circuit the solar cell the voltage rises to the point where this diffusion current is the same as the photocurrent, and so there is no net current out of the solar cell. This sets the maximum possible voltage that can be developed by the solar cell but it is useless at this point. And so at the the MPP the voltage is backed off enough from the open circuit voltage that this diffusion current does not draw off too much of the photocurrent away, and this operating point is near the "knee" of the I/V curve I mentioned above.
The panels in parallel will reach this voltage limit first, but the ones in series will only have half the voltage and so they'll have way more voltage margin than they need.
 
Connect cells in series to create the best (efficient?) voltage, then connect groups of those series strings in parallel.

In the OP, it sounded like you wanted to connect two cells in series, another two in parallel, then combine them.
 
Generally with PV you want to series your modules because it's a free lunch as far as losses go . However you are limited by the max voltage of your controller - Some are 150V, some are less, and there are two that do 600V - so unless you have only a few modules, you will have multiple strings connected in parallel. . Don't forget to temp adjust the VOC, about 1.2 of STC VOC here in the Northeast.

I recall some conversations back when I used to live off grid and was more into it about what voltage was most efficient for the charge controller. There were some reports on some of the forums about xyz controller running hotter/ more losses with this or that input vs output voltage, but my take was always it wasn't worth worrying about.
 
LarryFine said:
Connect cells in series to create the best (efficient?) voltage, then connect groups of those series strings in parallel.

In the OP, it sounded like you wanted to connect two cells in series, another two in parallel, then combine them.
Click to expand...
That’s what the customer was wanting to do, not me. I didn’t think it was a good idea, but everyone believes you can’t get bad info off the internet! LOL!
 
electrofelon said:
Generally with PV you want to series your modules because it's a free lunch as far as losses go . However you are limited by the max voltage of your controller - Some are 150V, some are less, and there are two that do 600V - so unless you have only a few modules, you will have multiple strings connected in parallel. . Don't forget to temp adjust the VOC, about 1.2 of STC VOC here in the Northeast.

I recall some conversations back when I used to live off grid and was more into it about what voltage was most efficient for the charge controller. There were some reports on some of the forums about xyz controller running hotter/ more losses with this or that input vs output voltage, but my take was always it wasn't worth worrying about.
Click to expand...
That’s why I don’t want to series all of them, it would exceed the rating of the controller.
 
hillbilly1 said:
That’s why I don’t want to series all of them, it would exceed the rating of the controller.
Click to expand...
Be careful about operating even close to the rating of the controller. You need to add the open circuit voltages, Voc, not the nominal voltages or the sweet spot, Vmp, voltages. Then add up to 25% to allow for increased voltage output in coldest weather.
Instead of 125% you can use the actual panel temperature coefficient as ND actual lowest winter temperature.
On a cloudless night the panel temperature at dawn may be lower than the air temperature at the same time.

Sent from my Pixel 4a using Tapatalk
 
Status
Not open for further replies.
Top