Series vs parallel/panel buying decision

[sepiot]

I am going to buy 3 panels, and need to make a decision: do I buy the higher spec panels which can only be connected in parallel? Or do I buy slightly smaller panels that allow me either series or parallel configuration, with the bonus of a higher voltage reading (when connected in series)?

These are my AIO inverter/charger specs:

Hybrid inverter specifications:

• Nominal battery voltage: 24V
• Continuous power output: 3000W / 3000VA
• Surge power: 6000VA
• Output voltage: 230V AC
• Output frequency: 50Hz /60 Hz
• Output waveform: Pure sine wave
• Maximum solar input power: 1500W
• Maximum open circuit solar input voltage: 145V
• Maximum solar charging current: 60A
• MPPT operating voltage range: 30V – 115V
• Maximum mains AC charging current: 60A
• Maximum combined charging current: 120A

Panel A (higher specs): can't connect them in series as voltage is out of range. Parallel results in 49.89v; 31a; 1547W
Peak power 425W
Vmpp 41.5V
Voc 49.9V
Impp 10.24A
Isc 10.74A

Panel B: when connected in series results in 112v;13.12a; 1470W
Peak power 410W
Vmpp31.25V
Voc37.25V
Impp13.12A
Isc13.88A

The panels are to go on top of a south facing shed, in the Uk, with partial shading in the early morning and the evening, though in the winter a neighboring house/chimney also gets partially in the way. It's almost impossible to guess, but I'd say the shed's roof is around 70% of the time under direct light, and 30% in partial shade (maybe more in winter, maybe less in summer); the shaded area covers anywhere between 10% to 35% of the roof within that 30% of the time that is partially shaded.

Obviously, 'Panel A in parallel' wattage is higher than 'Panel B in series', so this would indicate it is the better option? However, many people suggest that the higher voltage of panels in series is generally a better idea to get the batteries charging for longer etc.

There is also the consideration of buying a slightly more expensive DC breaker if going for Panel A (32A) vs Panel B (16A). And having to buy a couple of '3 to 1' MC4 adapters for the parallel configuration. I have also read that panels connected in parallel may need fuse/s as well as the breaker. Is this correct?

I am assuming that the solar cables would be same thickness (4mm square), but I am just wondering if the 77 extra watts justify the higher expense (not by much, but still...) and slightly higher complexity of the parallel set up. I don't know if there are any other variables to consider (safety?).

Any thoughts?

 

[Rednecktek]

For shading issues the parallel would be the better option slightly, but causes wiring issues you mention above. I think if you go parallel, since you need fuses now, the best bet would be to get a real combiner box so you can use thinner wires from panel to box, then a nice heavy wire from the combiner to the AIO.

You'll only need a 40a breaker which are pretty easy to source, but your combiner box will usually have one in there anyways.

Being in the UK, you're going to want every watt possible. In a perfect world you'd get 4 panels and rig them in a 2s2p config that would give you the best of both worlds. If physics says 3 though, then 3 it is.

Other than that you're looking good so far.

 

[sunshine_eggo]

I've recently changed my tune on parallel vs. series for shading based on the way shaded parallel strings behave.

I recommend panel "B" in series. Here's why:

3P array:

Panel 1 is partially shaded cutting off 1/3 of the panel output. Panels 2 and 3 are in full sun.

Panel 1 MUST now operate at a lower voltage to deliver it's max current through the bypass diodes, BUT it's in parallel with two higher voltage panels at higher output, so it can't flow a single microamp. In this case, the entire production of the panel will be lost UNLESS the MPPT can figure out there's more current available at a lower voltage; however, even if it does, it's going to pull down the voltage of the other two panels to the shaded panels... thus all panels are operating at max current, but 2/3 of Vmp... meaning you've still lost the equivalent of 1 of 3 panels.

In series, the offending section of panel(s) will simply be bypassed resulting in a Vmp drop but maintaining the full amps. In the above shading scenario, only 1/9th of the array would be lost vs. 1/3.

 

[Pi Curio]

I am going to buy 3 panels, and need to make a decision: do I buy the higher spec panels which can only be connected in parallel? Or do I buy slightly smaller panels that allow me either series or parallel configuration, with the bonus of a higher voltage reading (when connected in series)?

These are my AIO inverter/charger specs:

Hybrid inverter specifications:

• Nominal battery voltage: 24V
• Continuous power output: 3000W / 3000VA
• Surge power: 6000VA
• Output voltage: 230V AC
• Output frequency: 50Hz /60 Hz
• Output waveform: Pure sine wave
• Maximum solar input power: 1500W
• Maximum open circuit solar input voltage: 145V
• Maximum solar charging current: 60A
• MPPT operating voltage range: 30V – 115V
• Maximum mains AC charging current: 60A
• Maximum combined charging current: 120A

Panel A (higher specs): can't connect them in series as voltage is out of range. Parallel results in 49.89v; 31a; 1547W
Peak power 425W
Vmpp 41.5V
Voc 49.9V
Impp 10.24A
Isc 10.74A

Panel B: when connected in series results in 112v;13.12a; 1470W
Peak power 410W
Vmpp31.25V
Voc37.25V
Impp13.12A
Isc13.88A

The panels are to go on top of a south facing shed, in the Uk, with partial shading in the early morning and the evening, though in the winter a neighboring house/chimney also gets partially in the way. It's almost impossible to guess, but I'd say the shed's roof is around 70% of the time under direct light, and 30% in partial shade (maybe more in winter, maybe less in summer); the shaded area covers anywhere between 10% to 35% of the roof within that 30% of the time that is partially shaded.

Obviously, 'Panel A in parallel' wattage is higher than 'Panel B in series', so this would indicate it is the better option? However, many people suggest that the higher voltage of panels in series is generally a better idea to get the batteries charging for longer etc.

There is also the consideration of buying a slightly more expensive DC breaker if going for Panel A (32A) vs Panel B (16A). And having to buy a couple of '3 to 1' MC4 adapters for the parallel configuration. I have also read that panels connected in parallel may need fuse/s as well as the breaker. Is this correct?

I am assuming that the solar cables would be same thickness (4mm square), but I am just wondering if the 77 extra watts justify the higher expense (not by much, but still...) and slightly higher complexity of the parallel set up. I don't know if there are any other variables to consider (safety?).

Any thoughts?

If I had to pick one option out of these two;

Since it's a 24V system, with presumably an available installation footprint of 3x400W panels, Option B, 410w in series.

 

[sepiot]

Thank you all for the info. I think I'll go for Panel B in series: slightly cheaper, slightly simpler, and arguably more productive.
Yes, 2s2p would be ideal, and I may yet do it in future, but I will squeezing the 3 panels on the roof as it is, so the 4th one will have to be on the ground, with much less light exposure, and... getting in the way of us people, basically. Perhaps some sort of raised platform somewhere, but can't see it happening at the moment. But it's ok, this is just a secondary offgrid system to run the shed, and as a back up in case of need.

 

[sunshine_eggo]

Thank you all for the info. I think I'll go for Panel B in series: slightly cheaper, slightly simpler, and arguably more productive.
Yes, 2s2p would be ideal, and I may yet do it in future, but I will squeezing the 3 panels on the roof as it is, so the 4th one will have to be on the ground, with much less light exposure, and... getting in the way of us people, basically. Perhaps some sort of raised platform somewhere, but can't see it happening at the moment. But it's ok, this is just a secondary offgrid system to run the shed, and as a back up in case of need.

I would argue the 4th panel may actually be counter productive in the described conditions - i.e., 2S2P with a 4th panel may produce less than 3S.