diy solar

diy solar

Power performance warranty - how would you know?

mciholas

Solar Enthusiast
Joined
Feb 18, 2024
Messages
239
Location
Indiana
I'm shopping for panels (450W class, know any deals?) and was putting a lot of weight on the power performance warranty percentage and duration. Some are as bad as 20 years and 80%, some are as good as 30 years and above 90%.

But then I wondered, how would you know? Or more to the point, under what conditions would the vendor have to make good on a warranty claim?

Most warranties are about 25 years and 85% or so. A 15% drop in panel power output is probably something you can notice over a long span of collecting data, but there is no way you can verify that on a short time span like a day or week. A drop in power could be for all sorts of reasons other than the panel performance, like dust. How exactly would you know if your power has dropped due to panel performance?

I am kind of thinking these warranties are mostly useless for anything but gross power loss, like 50% or worse. You would be able to tell by comparison with a neighbor panel. But if the entire batch is aging uniformly and worse than warranty limits, you would be hard pressed to figure that out, and even harder to prove it. The warranty language seems to require testing in lab settings to verify, which is obviously not worth the effort for an aged panel.

Another factor is if these companies even exist and will provide warranty service 20+ years from now?

So do the power performance warranties actually have any value?

For reference, here is the warranty language from CanadianSolar:

TWENTY-FIVE (25) YEAR LIMITED PERFORMANCE WARRANTY

CSI Solar warrants that for a period of twenty-five years from the Warranty Start Date, the Products listed above will maintain a level of performance as set forth below:

• During the first year, CSI Solar warrants the actual power output of the Products will be no less than 98% of the labeled power output.

• From year 2 to year 25, the actual annual power decline will be no more than 0.55%; by the end of year 25,the actual power output will be no less than 84.8% of the labeled power output

The actual power output of the Products shall be determined for verification using Standard Testing Conditions only. Testing measurement uncertainty shall be taken into account and applied to all actual power output measurements.

In respect of the Twenty-Five (25) Year Limited Performance Warranty, if CSI Solar verifies in its reasonable judgment that the Products fail to conform to the terms of the Limited Performance Warranty set forth herein, CSI Solar, at its option, will provide one of the following remedies: 1) repair the Products; 2) replace the Products with new products whose labeled power wattages equal to or exceed the Warranted Wattages of replaced Products; 3) provide additional Products to make up the wattage difference between the actual measured power output wattages at the time of claim and the Warranted Wattages; or 4) provide a refund of the fair market value of the wattage difference between the actual measured power output wattages at the time of claim and the Warranted Wattages.

Mike C.
 
If you have several panels from same manufacturer it is easier to establish some sort of "baseline" and compare the output to that. If one panel is 20% lower than others it looks quite suspect.

Other than that you would need solar irradiance meter like https://fi.rsdelivers.com/product/f...irr1-sol-solar-power-meter-fluke-irr1/2107935 and solar panel MPPT tester https://holdpeak-store.com/holdpeak...ge-troubleshooting-tool-for-solar-pv-testing/
1. measure solar irradiance
2. measure panel temperature
3. measure panel output
4. calculate theoretical output to 1000W/m2 irradiance level with temperature compensation.

My best quess is that you wont get much better than 10% accuracy and with 0.55% degradation per year you won't notice other than gross failures for first 20 years.
 
I don't put much stock into Warranties and especially ones that are more than a couple of years. Few companies last long enough to honor such promises. Solar panels with 20+ year Warranties for amount of degradation likely would never be honored on the retail level. Perhaps commercial.

On a slight tangent, with prices of new panels getting so low these days it make used panels practically worthless.
 
If you have several panels from same manufacturer it is easier to establish some sort of "baseline" and compare the output to that. If one panel is 20% lower than others it looks quite suspect.
You can find an outlier panel that has a fault, but if all the panels age the same, and age worse than the warranty, comparing panels to each other won't find that.

My best quess is that you wont get much better than 10% accuracy and with 0.55% degradation per year you won't notice other than gross failures for first 20 years.
You might get to 5%. But even if you could get to 0.001%, I doubt that would be enough evidence to trigger a successful warranty claim.

In other words, vendor A claiming 92% after 25 years and vendor B claiming 85% after 25 years, is there really any practical difference in the warranty?

Maybe one can take the warranty terms as a "spec" of how good the panels are when it comes to aging, but I don't see anyway to actually hold the vendor to the terms if the panel don't actually meet that spec.

Maybe at the solar farm level it would be worth dismounting an array and subjecting it to a full STC test to prove the array is out of spec, but at the residential scale, that's not feasible.

Mike C.
 
Back
Top