• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

I'm arguing with my solar installer. Please fact check me!

Shinzul

Solar Enthusiast
Joined
Jul 6, 2022
Messages
179
Hey everyone!

My wife and I decided to get solar installed on our primary home finally, and we decided that it wasn't worth the time and effort for me to do the work myself, since it would involve climbing up on a 14/12 pitch roof (NO THANK YOU) to do the install work.

Anyway, long story short - I found a reputable solar installation company and signed a contract for them to do the work.

I decided to get a solark 15k inverter, and 45 panels (460w) - here is the datasheet.

The company sent me their engineering plans, and for some reason, their engineer thinks that the solark 15k cannot support all 45 panels. I'm 100% certain that their engineer is just plain wrong, and I'm arguing back and forth over email with them.

Can you kind folks please fact check me to make sure I'm not being an idiot somehow with this? Here is the email I sent them most recently, and I'm still waiting to hear back their reply. By the way, apparently they opened a support case with Solark to find out if I'm right?!? (which I'm really annoyed by, because isn't this literally the reason why I'm PAYING SOMEONE ELSE TO DO THIS FOR ME?!). I'm super frustrated right now tbh. Anyway, here's the email:

Let me clarify what I'm saying, since I'm not sure this is coming through clearly.

There are a few main and important points:

1. You CAN over-amperage MPPT inputs safely. There is no damage if you have too many amps going into MPPT inputs, the inverter will just self-limit the incoming wattage to its theoretical max of 6.5kw per MPPT.
Per the solark 15k manual:
"Max DC solar input = 19.5 kW (± 5%) | Max input power per MPPT = 6.5 kW | Max recommended input voltage per MPPT = 425 VOC, Max input current per MPPT = 26A (self-limiting)."
2. You CANNOT over-voltage MPPT inputs, it will fry the electronics.
3. When you put panels in series, it increases the voltage, and when you put panels in parallel, it increases the amperage.

So the solark 15k has 3 separate MPPTs, where each one can support a max (+/-5%) of 26 amps (6,500w), and as long as the power stays below 44a per MPPT, it can self limit the power safely.

5 panels in series would mean a max VOC well within the 500v limit. Then you would parallel 5 and 5 more panels to increase the amperage of those panels. This would be a single string of 5x3 panels, for a total of 15 panels per string.

The max power output of these 15 panels would be 6,900w, which is just barely above the max 6,500w input power per MPPT, but again, this is fine because the MPPT will self limit the power input.

The Solark can accept 6 total strings as inputs, but it's only THREE MPPTs in total. PV1 has 2 string inputs, and then the solark parallels them for you if you decide to configure it this way (which we wouldn't). From the solark manual:
This would be a similar configuration that I'm talking about, except we would have 5 panels in each "row" (series) with 3 rows (parallel), each using a Y connector to connect the rows together. In total this would be a single string.

Then we would do this 2 more times for the same total power output per MPPT.

Does this make more sense now?
 
Last edited:
The issue is not whether you are right or wrong. They being a business are more worried about liabilities, and will fight you tooth and nail on this issue. You will not win, and you won't find a reputable company that will allow you to exceed the max current rating.

If doing it your way is important to you, Joe Schmoe Solar would gladly have your business.
 
Hi. Just in case you have not seen it. https://www.sol-ark.com/resources/sizing-tools-calculators/solar-panel-sizing-tool/

Solark has a panel sizing application that you can enter your panel stats (if not already in the system). It also takes in to consideration min\max temperatures, and wire sizing. It's a good tool.

That is a high voltage panel, and how many you can have in series will depend on your record minimum temperatures for your area.
 
Last edited:
I don't see anything wrong with your plan.

Ask them if they know what over paneling is...

Point them to the Max Short Circuit Input current per MPPT on the data sheet - you're well under that:

1755291376071.png

If they don't understand that "current is pulled, not pushed", and that the Max Operating Input current is simply what the inverter will use, then I would have someone else do the work. An analogy is that a grid connection has a whole lot of current on tap, but it's not being forced (pushed) into my appliances - they use (or pull) whatever they need.

Or just find someone to install the panels only, and do the rest yourself.

That panel has a high VOC that appears to limit series strings to 7 panels. Assuming two parallel series strings of 7 on each of the 3 mppt's that's 42 panels.
He's looking at doing 5s3p on each MPPT, and I'd be concerned about voltage going too high in cold weather with 7s
 
The panel calculator that Sol-Ark updates regularly says Additional string -not recommended, and says NO on the wattage for the panels on each MPPT if configured in 5s3p. So, the manufacturer's tool supports the installers in this case.

I am guessing they are trying to stay within all of the spec sheet parameters, meaning listed wattage, voltage and amperage. Which for a company that will be expected to work with the manufacturer and support the product is a decent plan.

--------------------

Not directly related to this, but what I have found is that there are ideal panels for each situation. If have a given array location distance from the inverter, have a certain inverter voltage/MPPT input limits, and weather info, then you can often find a panel that is in the sweet spot. Picking things independently from each other, then trying to make them work is less than ideal.
 
Aren't they basically saying in their reply that you can't bring 9 strings into the 15K?

But then says they can do 45 by saying:
"
This would be a similar configuration that I'm talking about, except we would have 5 panels in each "row" (series) with 3 rows (parallel), each using a Y connector to connect the rows together. In total this would be a single string.

Then we would do this 2 more times for the same total power output per MPPT.

Does this make more sense now?"
 
Last edited:
If the installer won't do what you want, get a new installer.
They aren't going to do anything that they aren't comfortable covering under their warranty.
 
SolArk 15k with 42x ~490 W panels. 6 strings, 7 panels in each string. 4/12 pitch roof, but the peak runs N-S, so 2 strings on the east side, 3 strings on the west side and 1 string split. 20 kW of panels but I've only ever seen 18 out of it, with 14 being much more common.
 
I am not a solar power engineer. I am a retired telephone office power engineer specializing in 48 volt power plants. I can state with 100% certainty that no power engineer worth the name will design a system that exceeds manufacturer recommended system specs. Your proposed design exceeds manufacturer recommended specs which are provided by Sol-Ark.

What can be done to fix it? Look closely at the panels being provided. There is a very high probability you can find an alternative panel that will do what you want. The panels currently named have a problem with the way they can be stacked together in series/parallel strings without exceeding voltage and without exceeding recommended amperage.

Panel thermal efficiency is -.24% per degree C which means nominal temperature of 44C gives 65.6 VOC which goes up to 79 VOC at -40C. That exceeds MPPT voltage rating with 553 volts for a string of 7 panels where the MPPT is rated for 500 volts max.

with a 3 X 5 arrangement you wind up with rated panel amperage exceeding MPPT rating of 26 amps.
 
Last edited:
If your luck is similar to mine then ditch the solark and look for other options. There are inverters out there with similar or better specs (like 4 more, etc) that do not carry the "brand premium" of solark, which in my experience is not worth the electricity it takes to display their logo on an LCD screen.
Then again, that's just my experience.
 
I am not a solar power engineer. I am a retired telephone office power engineer specializing in 48 volt power plants. I can state with 100% certainty that no power engineer worth the name will design a system that exceeds manufacturer recommended system specs. Your proposed design exceeds manufacturer recommended specs which are provided by Sol-Ark.

What can be done to fix it? Look closely at the panels being provided. There is a very high probability you can find an alternative panel that will do what you want. The panels currently named have a problem with the way they can be stacked together in series/parallel strings without exceeding voltage and without exceeding recommended amperage.

Panel thermal efficiency is -.24% per degree C which means nominal temperature of 44C gives 65.6 VOC which goes up to 79 VOC at -40F. That exceeds MPPT voltage rating with 553 volts for a string of 7 panels where the MPPT is rated for 500 volts max.

with a 3 X 5 arrangement you wind up with rated panel amperage exceeding MPPT rating of 26 amps.
the amps from 3 panels in parallel are going to be 8.29 x 3 = 24.9A, still within MPPT input limit of 26A, where the specs are exceeded? Voc of 5 panel string is also well within specs. Is the total input power a concern with 15 x 450W = 6,750W per MPPT input exceeding 6,500W? But then again it shouldn't be a problem, just not all that power will be converted. In reality panels never produce 100% of their rating so this will not be even over-paneling.
 
This is odd, but I ran a regression to show how many panels of what voltage and amperage could be stacked to optimize power supplied to a 15K without exceeding Sol-Ark's recommended voltage and amperage ratings. The optimum size panels to max out power for 6500 watt MPPT's is between 720 watts per panel and 810 watts per panel. This presumes 48.1 volts output which is pretty close to standard for oversize panels.

With 810 watt panels, you could make 3 strings of 8 panels giving a nominal total power of 6480 watts per string with input voltage per string of 385 volts (well within safety range of 500 volt MPPT max) and 16.8 amps per string. This would be inside of all required Sol-Ark requirements yet would total 19,440 watts going into the 3 MPPT's. Getting late and I'm tired so somebody check my math. Somebody else can figure the optimum for other panel voltages.

Good luck finding 810 watt solar panels rated 48.1 volts! JA Solar makes 810's but VOC is way out of line.
 
Last edited:
In reality panels never produce 100% of their rating
I've been measuring output from my 705 watt Canadian Solar panels for the last 2 weeks. Even when facing full on direct at the sun, they do not reach 705 watts. The best I have gotten out of them was about 650 watts per panel.

Also, 8.29 amps is the rating for 450 watt panels. OP specified 460's.
 
Last edited:
OP, here is a suggestion that they might be willing to go with. Change to the 450 watt REC panels. The specs are just barely close enough that they may not burp about them.


They will definitely meet or exceed their rated output, if the environment is right.
I'm probably dealing with high haze levels of moisture in the air reducing sunlight on the panels just enough to drop the output. Given that temps were lower and the air was drier they would produce a good bit more. However, over time, all panels lose efficiency so after 10 years there will be measurably lower production.
 
However, over time, all panels lose efficiency so after 10 years there will be measurably lower production.
Definitely
1% degradation per year is expected.
So at 10 years, you should be around 90% of nameplate, on the best day.
 
the amps from 3 panels in parallel are going to be 8.29 x 3 = 24.9A, still within MPPT input limit of 26A, where the specs are exceeded? Voc of 5 panel string is also well within specs. Is the total input power a concern with 15 x 450W = 6,750W per MPPT input exceeding 6,500W? But then again it shouldn't be a problem, just not all that power will be converted. In reality panels never produce 100% of their rating so this will not be even over-paneling.
Way to pick and choose which number to quote. If you're going to go with Imp, might as well go with Vmp too, then? Why are you quoting Imp in 1 scenario and Voc in another?
 
Last edited:
For whatever reason, the Sol-Ark documentation says that you can only tie two strings to one MPPT. They don't give a reason, but that's their limit. Otherwise, it seems like Voc and Isc would be acceptable. I'm pretty sure the NEC code will also require installation according to limits of the manufacturer, so I think you're out of luck with any reputable installer.
 

diy solar

diy solar
Back
Top