[M_J_B]

I would like to use 2 x 100W and 2 x 50W panels on my roof. They meet a specific need that I have and would like to work out possible options before moving to other alternatives.

My question is related to the advice to avoid mixing panel wattage. I think that everyone ends up recommending two separate charge controllers but my brain is spinning with all the other items that I need to address for the van and I am not looking to become a solar expert for a single project. Don't get me wrong, I would like to know everything there is about the setup that I end up with but right now I just need help with choosing the best path.

So the panels in question:
- 100w -

• Maximum power (Pmax) 100W
• Voltage at Pmax (Vmp) 19.23V
• Current at Pmax (Imp) 5.21A
• Open-circuit voltage (Voc) 22.12V
• Short-circuit current (Isc) 5.50A

- 50w -

• Maximum power (Pmax) 50W
• Voltage at Pmax (Vmp) 19.06V
• Current at Pmax (Imp) 2.63A
• Open-circuit voltage (Voc) 21.82V
• Short-circuit current (Isc) 2.77A

I am looking at if I can get away with a single charge controller without hampering the system too greatly or if I get two charge controllers which volt/amp size do I need for each?

I will eventually be dealing with the B2B and shore power parts of this system but if there is a solution that fits the entire scope that would be great. These panels are supplemental to the charging system and will not be relied upon for my primary charge source.

EDIT: added "solved" to title.

 

[MaggieMarty]

I like Nate’s aka Explorist videos. Perhaps this helps.

 

[larry barello]

They have essentially the same Vmp, so just run then in parallel. You could series the 100w panels, and the 50w panels and then parallel the two. If you series an 100 and 50-watt panel you would effectively limit the 100 watt panel to 50 watts.

There is a lot of discussion on parallel vs series. Parallel wired are tolerant of partial shading and theoretically series perform better on low light situations. My experience is that low voltage parallel works just fine in low light.

MPPT charge controllers (mostly) work as "buck" voltage converters and the less down conversion the more efficient they become. It is a myth, or just bad/old design, that they only work when the input voltage exceeds the battery voltage by 5v.

 

[MaggieMarty]

They have essentially the same Vmp, so just run then in parallel. You could series the 100w panels, and the 50w panels and then parallel the two.

There is a lot of discussion on parallel vs series. Series wired are tolerant of shading and theoretically series perform better on low light situations. My experience is that low voltage parallel works just fine in low light.

Speaking of low light… I’m in Southern California and we are at the tail end of the latest Pineapple Express. Yesterday it was heavily overcast, drizzling and foggy. My 400W panels wired 2 series/2 parallel still produced 5 amps to offset my freezer’s use in the van, and even recharged my battery from 75% to full by early afternoon.

 

[dog_house]

IMO even if your setup is in theory a few percent less efficient, at the end of the day you are getting more power in your batteries because you are getting more panel on the roof with the "mismatched" panels.

 

[Chance]

They have essentially the same Vmp, so just run then in parallel. You could series the 100w panels, and the 50w panels and then parallel the two. If you series an 100 and 50-watt panel you would effectively limit the 100 watt panel to 50 watts.

……cut…..

To avoid limiting and therefore wasting capacity, he can either wire all in parallel as you mentioned since voltages are near identical, or use combination of parallel and series in slightly different way.

Combining the two 50 Watt panels in parallel yields essentially the same profile as the 100 Watt panels, and then can be wired in series to provide +/- 58 Volts. Open circuit would be a little high in that it would be over 60 Volts, but not by much. Both options should yield equivalent of 300 Watt panel capacity, but at very different voltages.

 

[larry barello]

Good point, I hadn't considered the (50|50)+100+100 case.

I still advocate parallel for all. 20v at 15a is safer than 60v @5a. From an electrical shock point of view double breaker might actually make sense for 60v and above.

 

[dog_house]

Good point, I hadn't considered the (50|50)+100+100 case.

I still advocate parallel for all. 20v at 15a is safer than 60v @5a. From an electrical shock point of view double breaker might actually make sense for 60v and above.

☝Above 50 volts DC is where most people draw the line as far as it being lethal. Once you get above 48v you need to start treating it like “high voltage” and things like exposed terminals become a big deal.

 

[Chance]

☝Above 50 volts DC is where most people draw the line as far as it being lethal. Once you get above 48v you need to start treating it like “high voltage” and things like exposed terminals become a big deal.

I don’t know who most people are, but when European auto companies were looking at increasing voltage beyond 12 VDC to improve efficiency, they conducted safety tests to see how high they could go without excessive danger. Test results showed up to 60 VDC was not too dangerous, so working backwards in 12-Volt increments they agreed on 48 VDC nominal as the new standard to eventually replace 12 VDC. Just like 12V charges at 14~14.5V, then 48VDC will charge at 56~58 Volts.

Interesting enough, since lithium cells have higher voltage than lead, the 48V standard is now often 51+ Volts, which often charge at 58V, or even higher occasionally.

Anyway, as I noted previously, “open circuit” voltage if all in series would exceed 60V; by approximately 6 Volts (66 V total). On other hand, V (max power) would be 57.5 Volts. If not fully comfortable with electricity, parallel would be a better choice regardless. I’m not personally concerned working with 66 Volts, so would consider both.

P.S. — Before European auto industry agreed on 48V nominal voltage, the US (GM) had use 36V that charged at 42 Volts. That voltage was never too popular. Solar houses and even golf carts were already at 48V. Today even higher voltages are seen on some yard equipment, electric bikes, etc. Granted, it’s more dangerous, but manageable with care.

 

[M_J_B]

So I found a calculator online that provides the previously mentioned options. Upon entering the specs I end up with these possible outputs:

All panels wired in SERIES:

• Max power output: 151.28W
• Max power voltage (Vmp): 57.52V
• Max power current (Imp): 2.63A
• Loss: 39.61%

All panels wired in PARALLEL:

• Max power output: 248.73W
• Max power voltage (Vmp): 19.06V
• Max power current (Imp): 13.05A
• Loss: 0.71%

All identical panels wired in SERIES, then all series strings wired in PARALLEL:

• Max power output: 149.43W
• Max power voltage (Vmp): 19.06V
• Max power current (Imp): 7.84A
• Loss: 40.35%

All identical panels wired in PARALLEL, then all parallel strings wired in SERIES:

• Max power output: 100.7W
• Max power voltage (Vmp): 38.29V
• Max power current (Imp): 2.63A
• Loss: 59.8%

So it looks that "all in parallel" wins out according to this calculator. BUT... I am further confused; this other calculator provides different results (same panel specs entered):

This is why I am so confused.

I am OK with the "all parallel" option unless I am missing something.

 

[larry barello]

The second calculator is correct.

In series, the lowest current panel rules them all. The 100w panels will try to push current through the 50w panels but that excess current will just be turned into heat. So the effective string (100 + 50) will just produce ~100w

In parallel the lowest voltage panel rules. Similar logic to series in that the higher Vmp will try to boost the voltage of the lower panel and the difference will just be heat in the lower voltage panel.

Parallel wins (IMO) because in partial shading the lowest voltage panel (or string of cells in a panel) will "drop out" due to the built in blocking-diodes (a twist you probably didn't expect) and the rest will continue to contribute. The different in Vmp of your mixed panels is too small to be affected by the blocking diodes, hence the small loss due to the dominance of the lower voltage panels.

In large arrays series are used to keep the total current to a manageable level, again, all tradeoffs based on site, size and what you are trying to do. In your case 20v @ 13A are both low numbers easily handled with the standard #10 solar pigtails and combiners available for cheap on Amazon, etc.

 

[RV8R]

So I found a calculator online that provides the previously mentioned options. Upon entering the specs I end up with these possible outputs:

All panels wired in SERIES:

• Max power output: 151.28W
• Max power voltage (Vmp): 57.52V
• Max power current (Imp): 2.63A
• Loss: 39.61%

All panels wired in PARALLEL:

• Max power output: 248.73W
• Max power voltage (Vmp): 19.06V
• Max power current (Imp): 13.05A
• Loss: 0.71%

All identical panels wired in SERIES, then all series strings wired in PARALLEL:

• Max power output: 149.43W
• Max power voltage (Vmp): 19.06V
• Max power current (Imp): 7.84A
• Loss: 40.35%

All identical panels wired in PARALLEL, then all parallel strings wired in SERIES:

• Max power output: 100.7W
• Max power voltage (Vmp): 38.29V
• Max power current (Imp): 2.63A
• Loss: 59.8%

So it looks that "all in parallel" wins out according to this calculator. BUT... I am further confused; this other calculator provides different results (same panel specs entered):
View attachment 94770

This is why I am so confused.

I am OK with the "all parallel" option unless I am missing something.

Hi @M_J_B

I like much of what @larry barello writes on electrical & solar ,,, I think he knows what he’s talking about in general.

My suggestion is you “wire” your panels for worse case scenario for “size” or gauge. For this you are basically concerned about “amps” and you will get the most “amps” out of any given Wattage from “Parallel”. Thus wire to your roof with the largest gauge wire you think you will need for “Parallel”.

If you do this, you can “reconfigure” your array “series” vs “parallel” and experiment, providing your solar charger can handle the various configurations you might have.

I have a couple of Victron 100 | 50 solar chargers. Victron “clearly” states battery voltage + 5 volts before any charging will start. Once started that requirement drops. So I think that “myth” really depends upon the equipment & software parameters. My comment around the higher voltage not being as effective as these units are basically “bucks”; I think it depends more on “The Sun” and the equipment. Here is a “perfect score” @ my Cabin “North of the 49th” with my 3S/2P that is 600W @ 60 volt @ 10 amps;

I think “Forum Advice” as priceless as it is, is not a replacement for “Real World Experience”. Further if you size your wires for worse case scenario then you have the options of reconfig & learning what works best for you with your equipment & installation.

 

[M_J_B]

Thank you everyone for your well thoughtout and detailed replies. It has been really helpful. I have decided to go with the 2x100w and 2x50w panels and will be wiring them in parallel.

 

[HarryN]

Thank you everyone for your well thoughtout and detailed replies. It has been really helpful. I have decided to go with the 2x100w and 2x50w panels and will be wiring them in parallel.

If you are going to do that, strongly consider to use a PWM controller such as a bogart engineering unit rather than an MPPT type.

If your plan is to use an MPPT controller, then look again at what @MaggieMarty posted in #4.

 

[Chance]

Yeah, the 38V series/parallel set up is even better and doesn’t waste capacity. Avoids high voltage also. That makes three possible efficient choices.
.

 

[sjbird]

On series vs parallel and partial shading - great video I came about recently.
No theory, just down to earth experiment.
Video is long, can be shortened to about a minute, but still worth seeing.

Spoiler alert - doesn't matter how you wire it. Partial shading kills a panel. But if that panel is in series, the bypass diode saves the day without killing whole string.