assessing performance of skoolie rooftop solar panels

[johnim]

Hey guys I have a 48V system with 2 strings of six 240W panels for a total of 2880W potential. Obviously I'm not going to get that but I find I am only getting on average 1.2-1.4 kW at max sunlight. FYI the panels are on a skoolie parked next to a wall of trees so it takes til about 1-2P before unobstructed light hits the panels. I disconnected each string one at a time from my combiner box and found there was about a 40 watt difference between the two strings going to my inverter and my understanding is if a panel in a series is dead I shouldn't get any power past that connection but my memory is a little fuzzy on that. In any event I'm faced with disconnecting in turn each panel in a series of six one at a time and checking each with my voltmeter unless someone can offer a better idea. Any constructive ideas are welcome.

 

[Vigo]

Checking each panel individually probably IS the quickest option since you only have 12. Compared to a larger array that is not a very intimidating number although i get that the thought of doing it is irritating. Voltmeter is one thing, and if you have a dc clamp meter you can also connect the panel leads to each other and see if the current it pushes is near the ISC rating. Neither is ideal since its checking at the extreme ends of the range (open circuit and short circuit) when the panel actually operates somewhere in between, but it is still a simple way to find outliers.

 

[johnim]

Ty Vigo! it's only a pain bc I have to climb a ladder and disconnect each panel which is series connected to the next panel and temp is in the 90s all week in VA which means it's hot af. I have a Klein clamp voltmeter that I got after happening on this site and watching a ton of Will's vids but I don't like the sound of connecting the leads of a panel together that sounds not safe. I didn't even know you could do that. Looks like it's gonna be a very warm testing day tomorrow.

 

[Vigo]

Gotcha. Make sure the clamp meter will do DC as a lot of clamp meters are AC only. Shorting the panel to itself is not dangerous to the panel other than the minor arcing that could occur just as the contacts of the connectors join and separate. If its not done repeatedly its worth the diagnostic value in my opinion, other opinions may vary. But you'll never get more current than what the panel was designed to make, so it's not hurting itself in any way other than minor arcing on the connector terminals. They're your panels so if that thought bothers you you could prevent the arcing by covering the panels or waiting until shade/dark to do the connections, then removing the cover or coming back to measure when sun is high.

 

[HarryN]

Some people will cover the panel with a tarp or similar when working on them to cut the power off.

 

[SamDeleted]

skoolie

for those of us who don't speak millennial, that means a school bus converted to a camper.

(I had to Google it ?)

 

[johnim]

Haha, thank you all. I didn't even think about covering panels with a tarp ?. This morning before the sun was too high I disconnected all panels and just came in from checking the first string that's getting full sun. The number 5 panel in the first series only read 5v. However, I went on testing and came back to it and with some fiddling of meter lead placement I got a good reading similar to the other panels. So I'm thinking the panel may be okay but something is definitely going on with the connectors on that panel. Maybe I could just change them and see how that works. So now I am waiting for the sun to rise high enough to test the other side in complete sun.

 

[johnim]

Update: The second string of 6 panels all read similar to the other side and average between 31 and 32 volts. I also noted that the positive connector on the #5 panel on the other side has some corrosion and I believe I could clean or replace and probably be okay.

I went back and reviewed the panel info:
Model SS240P - 60
OC 37.2
VPmax 30.4
Current at Pmax 7.89
Max system voltage 600
fuse rating 15

So now I'm back to why am I only getting avg 1.2-1.4kW power when I have 2880kw of panels connected? I understand there will be a loss but even at 80% I should expect 2300W or even 75% is just over 2100. Could it be just because of the one panel with an intermittent connection?

Also, I should add when I climbed up to look at the combiner box yesterday I found visible burns where the negative cable going to the inverter plugged in. The wire was very loose and hot near the box and I presume it had been arcing. How it got loose I have no idea since I haven't driven the bus since I hooked up the system. In any event I tightened the nut down and checked the input to the inverter and it remained about 1.2-1.4kW. I also called Ecoworthy (where I got the combiner box)and they suggested it's not uncommon for a loose wire to arc and that I should watch it. Oh, the combiner box lists max of 10A on a single string and since my panels are 7.89 max that should be okay.

FYI the combiner box is positioned about midway on roof of a 37' bus so each connecting cable running to the box is about 20-25' long. I got it from FG cable on Amazon and it's 2PfG 1169 PV1-F 1x6.0mm sq with rated current of 70 and rated voltage of 1000 so for each string the current of 7.89 and voltage of roughly 223 are well within limits.

Also the wire running from the box to the inverter is the same as that connecting each string to the combiner box and my inverter is an EG4 3000W. The loose wire didn't seem to make a difference in what is getting to the inverter. And when I checked each string separately I got about 40W less on the side with the faulty panel/connector. But nothing yet explains why my wattage is so far below expectations.

What am I missing?

 

[Vigo]

So, GOOD thing that you found the loose connection. Sounds like it could potentially have escalated into greater damage.

Just to clarify, when you are getting 1200-1400w, what is the condition of the rest of the system? Like, are the batteries charging, what is your max charge current, are there AC loads, how many watts?

If for example you are not running AC loads and your max charging current equates to ~1400w, that is all you're going to get because that's all you can make use of under those conditions.

Assuming its truly the panel strings causing the issues, one thing you could do is make sure your batteries are not full, set a certain charge current that either string individually should be able to support, and disconnect one string at a time and see what the remaining string's PV volts and amps do on your inverter screen. Having slightly lower OC voltage on one string is not nothing but it also doesn't correlate to 1kw+ of missing power. Running each string against the same 'load' and then observing differences in voltage and possibly current under that load condition would probably let you narrow down which string was underproducing in a more direct/reliable way vs just OCV. If you didn't mind being on the roof at that point you could also check for uneven temps between all the MC4s on that string. If one was causing a bad connection and a bunch of voltage drop it would be getting hotter than the rest as well. Could also check the 'junction boxes' where leads come off panels for uneven heat level between panels. Also viewable with a sweet IR camera if you happen to have one.

An IR camera may even be able to spot a bad section of a solar panel, but that's just a thought exercise in my head with no real world experience to back it up. Plus, not many people have an ir camera of any sort to begin with. Just a fun thought.

 

[HarryN]

Sometimes photos help us visualize things that we can't interpret completely from text.

Tell us about your charge controller, battery pack, the wire that runs from the combiner box to the solar charge controller, etc.

Is your battery pack depleted enough to pull in that much power ?

 

[HarryN]

Your solar panel testing so far is a good basic "it turns on" test but not a really good "it is working well" test.

To really load test a panel, sometimes this requires either connected a healthy resistor across the terminals and measuring current with the clamp meter - or connecting each one to a battery pack to see what is happening.

For instance in your case the Vmp is ~ 30 volts, so you could make up a 24 volt battery pack that is at 50% SOC and just connect one panel at a time directly to the 24 volt pack. This will bring the panel voltage down to V bat and then you can measure the current using the clamp meter.

You don't need to be in full sun to get some basic measurements and just some MC4 cables run up to the roof from a battery pack on a wagon (fused) can tell you a lot fairly quickly.

Well not super fast - it is going to take quite a while to test them all but at least it is simple enough.

The current will be no more than Isc so maybe 10 amps under ideal conditions and more likely close to Imp.

Keep in mind that your location is more or less coastal / sea level. There is a lot less sunlight and high energy photons reaching your solar panels than for instance on top of a plateau in NM where there is negligible humidity in the air and it is nearly 5K ft altitude. (official test conditions )

 

[johnim]

Thanks to both of you. So I was on phone with Signature Solar (sells my EG4 batts and inverter). Hold time sucked but staff was actually great and knowledgeable staff. Something I hadn't considered goes to your point Vigo is that my charge current default was set at 30A which is the low end. (100A max but recommended 30-50A). And just to be clear the 1.2-1.5kw is under no load with exception of the inverter obviously but under a 12k mini-split running a load of 1-1.2kW I may get 3-400W going to the battery. Harry, some of what you're saying is a bit over my head and knowledge base but I will keep it in mind and revisit once I reconnect everything. Now that I've bumped up max charging amps to 50 I am waiting til tomorrow to reconnect everything and then see the results of that change. Hopefully, the charging amperage is the major component of the my dilemma but if I don't see a boost than yea I guess I need to consider how the panels are reacting under load. They're the only used component of my system. I couldn't see paying a premium so I bought a bulk set of 20 from one of the sites Will recommended, I can't remember the name off the top of my head. Anyway gentlemen thanks again and I'll update tomorrow.

Oh btw, SamG no offense taken but I am in no way, shape, or form mellinial ?. I didn't know what a skoolie was either until I started researching bus conversions.

 

[HarryN]

Thanks to both of you. So I was on phone with Signature Solar (sells my EG4 batts and inverter). Hold time sucked but staff was actually great and knowledgeable staff. Something I hadn't considered goes to your point Vigo is that my charge current default was set at 30A which is the low end. (100A max but recommended 30-50A). And just to be clear the 1.2-1.5kw is under no load with exception of the inverter obviously but under a 12k mini-split running a load of 1-1.2kW I may get 3-400W going to the battery. Harry, some of what you're saying is a bit over my head and knowledge base but I will keep it in mind and revisit once I reconnect everything. Now that I've bumped up max charging amps to 50 I am waiting til tomorrow to reconnect everything and then see the results of that change. Hopefully, the charging amperage is the major component of the my dilemma but if I don't see a boost than yea I guess I need to consider how the panels are reacting under load. They're the only used component of my system. I couldn't see paying a premium so I bought a bulk set of 20 from one of the sites Will recommended, I can't remember the name off the top of my head. Anyway gentlemen thanks again and I'll update tomorrow.

Oh btw, SamG no offense taken but I am in no way, shape, or form mellinial ?. I didn't know what a skoolie was either until I started researching bus conversions.

I had no idea that you were using a smart battery / controller combo like that which had the ability to limit the charge current.

I am more experienced with conventional charge controllers and battery packs that will run as fast as they are rated for.

Good example of a reason to have some photos of the rest of your equipment to help with the diagnosis.

Good info - please keep us informed.

 

[johnim]

My bad Harry, didn't consider that. So cleaning the #5 panel connector I got a consistent 31-32V reading so I think the panel is fine--or at least as good as the other panels. My max now is 1.85kW and 35A going to charge the battery in full sunlight with only a100W fan running and of course the inverter, which also runs at about 100W so that puts me theoretically at almost 2kW. Out of 2880W of panels that's about 69.4-70% efficiency and I believe they're supposed to be good for 25 years of 80% output power so I'm still missing 10% somewhere. I'm thinking it could be soot from the trees causing some of the deficit. I did clean them a couple days ago with a car brush and dawn soap but I noticed it didn't get all of the soot completely off. I could try something that cleans better and test again but I don't have a battery pack and I'm not sure how to do the resistor thing. The other thing you mentioned, at max sunlight the sun is not directly overhead, more like only 45 degrees to the surface of the panels, so I guess that is playing a role as well. Wow, so many things to consider with real world solar vs lab values ?. Anyway, I think this finishes me out for this issue. Thanks for everyone's help!

 

[rustythorn]

My bad Harry, didn't consider that. So cleaning the #5 panel connector I got a consistent 31-32V reading so I think the panel is fine--or at least as good as the other panels. My max now is 1.85kW and 35A going to charge the battery in full sunlight with only a100W fan running and of course the inverter, which also runs at about 100W so that puts me theoretically at almost 2kW. Out of 2880W of panels that's about 69.4-70% efficiency and I believe they're supposed to be good for 25 years of 80% output power so I'm still missing 10% somewhere. I'm thinking it could be soot from the trees causing some of the deficit. I did clean them a couple days ago with a car brush and dawn soap but I noticed it didn't get all of the soot completely off. I could try something that cleans better and test again but I don't have a battery pack and I'm not sure how to do the resistor thing. The other thing you mentioned, at max sunlight the sun is not directly overhead, more like only 45 degrees to the surface of the panels, so I guess that is playing a role as well. Wow, so many things to consider with real world solar vs lab values ?. Anyway, I think this finishes me out for this issue. Thanks for everyone's help!

70% for flat panels, especially if they have low air flow is not unreasonable. also the farther north the less power you will get from flat orientation

 

[johnim]

good to know..thx rusty