Weighing Parallel vs Series Connection of Solar Panels

[Jon R]

Hi All,

I understand the basics of this question regarding how to wire it, doubling of current versus voltage, and wire sizing considerations. I'm trying to weigh the actual charging benefits and overall system efficiency effects in a real world camper installation.

I have a new Grandby arriving soon and have already purchased two 200 watt Renogy rigid monocrystalline solar panels (open circuit voltage 25VDC and short circuit current 10.5 amps, roughly). The panels will be mounted flat on the roof and I don't plan right now to attempt a tilting installation. I couldn't find it in the specs, but measuring the OCV of one of the panels in direct sunlight while successively blocking more cells one by one seems to indicate the panels have bypass diodes on each individual cell.

I also have Victron Smart MPPT 100/30 and Orion 12/12-30 chargers. My 200Ah LiFePO battery is on backorder, so I'm bench testing the solar system with a large truck FLA battery.

I'm in the Pacific Northwest West of the Cascades, so much of my local camping is going to be in clouds and/or tree shading. (Everywhere else in sun I expect I'll have excess charging capability relative to my usage.) My measurements seem to show that, in cloudy conditions, if I connect in parallel I will not generate enough voltage to start charging (5 volts over battery voltage) until the sky gets fairly bright. However, with them connected in series, I was charging, at least at a low rate, by 8am on today's cloudy morning. The bypass diodes seem to largely eliminate the disadvantage of running in series when the panels are partially shaded on a sunny day, plus I expect I will have excess charging capability when camping in sunny conditions.

It seems obvious to me that, for my usage, series wiring will give me more total energy input on a given cloudy day due to being able to exceed the charger voltage requirement earlier and later in the day, even though the watts generated at those times will be fairly low. It will also have the slight advantage of less wire loss using the 10AWG factory solar panel wiring. Am I missing anything I should be considering?

Eventually I could just try it both ways, but I'm hoping some has already done a similar comparison with their equipment. Thanks for any advice you can offer.

 

[Vic Harder]

Nicely researched post Jon R. I'd go serial myself for all the reasons you stated. As for more charging in the PNW, have you considered adding another 200W as a portable with 100' of cable? Should be able to find a sunnier/angled spot with that.

 

[ri-f]

Hi Jon, I live on Whidbey Island, just to the west of you. Today, it is cloudy here, no sun breaks, and my two 160W panels are seeing 29v coming in now in the early afternoon. The panels are rated at 31.7v, so they are putting out in these conditions. I have a Lithium battery and it is sitting at 14.2v at the moment. Earlier this morning with heavy, dark cloud cover, I was seeing 13.6v, rising to 14.2 as the sky got a little lighter. If the sky is bright, but cloudy, it will put out current. Of course I still need direct sun for it to be putting out significant current. I configured my panels in parallel, because 320w (in my case) wasn't an issue with 10 AWG wire from the roof to the controller. I used 6 AWG from controller to battery on the output side of the Victron 100/30. So, the wire size was not a limitation. The partial shading/obstruction advantage of parallel (with regards to partial shading) was why I went parallel vs. series. I don't believe that you will see a significant advantage under darkish, heavy cloud cover simply by configuring in series. It's a tradeoff, I know, but I do think that your panels are similar to mine in their max voltage output (29v-31v) and because of that, you will have a lot of wiggle room during low-light times (early morning/late afternoon. Your MPPT controller will do a great job over optimizing the extra voltage-to-current conversion during those times of day. Yes, there is an advantage to series, as you are aware, but in your case, I still believe that parallel is preferable - at least it has been for me. You will need another controller if you add an auxilary panel , down below, iwith your roof panels in series. I do think that 400+ watts is on the edge of what 10 AWG can do without efficiency losses through the wire in a parallel config, but again, with a VoC of 29vdc, you got voltage to spare and losses through the wire probably will have a mimimal effect, ultimately, on your output current. Just going off my own real-world results.

Rich

 

[rando]

I don't think there will be much of a difference either way. You will have slightly lower resistive losses with series panels, but better performance under partial shade with parallel panels. The earlier charging of series panels is a bit of a red herring - the output voltage is far less sensitive to the amount of light than the output current, so even if you can get the voltage above the threshold to turn on the controller, there will be almost no power as the current will be so low. You can prove this to yourself by playing around with a solar simulator: https://www.pveducation.org/pvcdrom/solar-cell-operation/effect-of-light-intensity

It sounds like you have enough panels that when you have good sun, you will harvest more than enough power regardless of resistive losses, so your design criteria are more focused on partial shade and low light, in which case parallel has a slight advantage. Unless you have an unusual (and expensive) panel, I highly doubt that you have bypass diodes on each cell, most likely 2 - 4 per panel, 1 or 2 for each string.

 

[rando]

One other comment - while I am not familiar with your specific panel, most non-residential panels have two strings of cells in parallel, so measuring Voc won't tell you much about the diode configuration. Unless you shade cells in both strings, the voltage should stay constant (as it will be provided by the other string) as you shade a few cells, but Isc will drop.

 

[Jon R]

Thanks for all the thoughtful and helpful comments. I’ll call Renogy on Monday and see if I can get an answer on the panel string arrangement and bypass diode configuration and post what I find out.

 

[rando]

One other comment - while I am not familiar with your specific panel, most non-residential panels have two strings of cells in parallel, so measuring Voc won't tell you much about the diode configuration. Unless you shade cells in both strings, the voltage should stay constant (as it will be provided by the other string) as you shade a few cells, but Isc will drop.

Well, I stand corrected. If these are the panels you have, they are only 1 string (40 cells x 0.6Voc = 24Voc), and have 2 diodes. I am fairly sure those diodes are rated at 10A, so I would hope they are in parallel in this application as the Isc is > 10A. You can probably tell if you take a look in the junction box, in which case there is 1 bypass diode per panel.

 

[Jon R]

Thanks, Rando. Yes, those are the panels I bought. I didn't think to download the spec sheet because I assumed it had the same specs shown on the web page, which didn't mention the diode configuration. It looks like parallel is the initial configuration I should try.

FYI for anybody who has a recent Grandby front dinette, two of these particular panels fit between the two roof vents with a couple of inches to spare. It was the most watts I could find to fit in that space. There is room for a third matching panel behind the rear vent as well, but I don't think I'll need that much power output. In addition, the distance between the center points between the pairs of mounting holes on the long sides is exactly 58 inches, which is the distance between the factory installed Yakima tracks. That makes it easy to install them with a small bracket at each corner without having to build an additional frame. I'll show my mounting method in late May after I have them installed