diy solar

diy solar

Used 2kW Grid-tie, very low wattage output

printbarron

New Member
Joined
Dec 14, 2023
Messages
3
Location
Steuben, WI
A friend of our was grappling with whether or not to replace her 2kW grid tie system. She didn't want to see it go in the landfill. I told we would take it, as any PV system was better than what we had (which was nothing). A local retired installer told me to test the voltage of all panels. At the time I didn't know I should also test the amperage. The panels are Sharp 165W and the inverter, a Sunny Boy 3000 (2019).

I finished mounting and connecting our 12 used panels and got everything approved by our power cooperative. The lineman tested our output and it was very low, only 0.8A AC at the time. Indeed, the inverter also shows very low wattage produced, though that number doesn’t align with the amps AC. At 0.8A X 240VAC, I would expect to see ~192W output reading from the inverter(?).

The inverter shows ~42-108 W on a sunny morning – about 5% of what the system “should" be. I was expecting at least 50-60% of what it would have been new. I’ve tested every panel as another installer has instructed, and aside from one dud, all look fairly consistent (see diagram below).

Even if one string is low (string A), does it make sense that the total output watts would be compromised by one string (the inverter accepts two separate strings)? I might try swapping two higher watt panels from string A, with two low panels from string B to see if anything improves. In your experience is there another thing I should be checking? The inverter seems happy, with no errors, but I know nothing beyond what the readout is showing.

Any help is appreciated!

(Thinking about swapping poor for decent, to bolster string B, making string A all poor performers.)
231214_Solar_Panel_Test.png
PV_FINAL_1635.jpegPV_FINAL_1637.jpegPV_FINAL_1636.jpeg
SHARP_165W_tag.jpeg
 

Attachments

  • SB_3000US_Inverter_tag.jpeg
    SB_3000US_Inverter_tag.jpeg
    262 KB · Views: 7
There are any number of things that can completely destroy array performance in a way that doesn't make sense. Now is not the time to randomly swap thing around hoping for improvement.

After sundown recommend you disconnect all panels from each other - completely dismantle strings A & B. Assuming each string is 6S, and they are in parallel.

Tomorrow, as close to noon as possible, assuming full sunshine, go check the voltage and short circuit current of each individual panel (a harbor freight multimeter has a 10A setting current setting). Document all values and report back.

You may find that one or more panels need to be completely removed from the array.
 
Sharp 165W are known to degrade.
It may or may not be apparent from Voc and Isc measurements. Some defects only show up at a voltage and current operating point between those two.

I located and replaced 5 panels out of 24, boosting performance.

Because your inverter is transformer type, we won't blame it on "PID".


New and good used panels are so economical ($0.20/W more or less), you may want to upgrade and put about 4kW of panels on the 3kW inverter. For an investment in the $800 range.
 
Thank you both for your replies. We finally have a nice sunny day and I was able to test all of the panels around 11:45AM. Attached are the readings I recorded, which is very surprising to me. How am I getting 200W? Indeed, three look significantly lower. Thoughts on the best way to proceed? Any ideas are much appreciated!

231218_Solar_Panel_Test.png
 
Remove ( bypass ) the suspect panels and reconfigure the remaining 9 panels to a single series string to the sunnyboy. Even if you had working 6 panel strings parallel the voltage would very close to the minimum of the sunnyboy ( 200V )
 
Thank you both for your replies. We finally have a nice sunny day and I was able to test all of the panels around 11:45AM. Attached are the readings I recorded, which is very surprising to me. How am I getting 200W? Indeed, three look significantly lower. Thoughts on the best way to proceed? Any ideas are much appreciated!

View attachment 183738
Unless you had the panels actually hooked up to a load and 2x meters (one for voltage, the other for current), you measured the open circuit voltage (VOC), and the short circuit amperage (Isc). Panels never produce at the power represented by VOC x Isc. There is a curve. MPPT (maximum power point tracker) chargers try to find the maximum power point on the curve.

 
If the original configuration was 6 series, 2 parallel the vmp would be right at the inverters minimum voltage, a loss of even one panel in each string would make the system generate very little energy. Going to a single string and removing the marginal panels should get things working, likely even better than when initially installed, even with fewer panels
 
eliminate panels 2, 9 and 11 and place the remaining 9 in series. Given the measured Voc, should be good down to -31°C before you blow 500V.

re-check a couple of panels for Voc only first thing in the morning when the sun first hits them.
 
Back
Top