Solar station. DIY thermal imaging and electroluminescence testing

An overview of the typical solar problems and what to look out for before it's too late. Based on an analysis of 50+ home solar stations of various ages.

Introduction

The main testing methods which may show no qualitative changes in the work of both the solar station and the panels are thermal imaging diagnostics and testing using electroluminescence. 

Quantitative measurements include measuring the volt-ampere characteristic of the panel (IV Curve), but about that another time.

This spring, finally, all the work was completed, my https://habr.com/ru/post/465133/ stations were completed and it was understood that it would be nice to check that everything that was acquired by back-breaking labor was built.

Having heard enough about possible variants of defects in solar modules, I decided to delve deeper into the topic and share my experience.

Thermal imaging survey of solar stations

Thermal imaging was chosen as the simplest, though not the most budgetary solution. There is a lot about thermal imagers on HabrΓ©, and a lot has already been written on the Internet, but after reading it, including

https://habr.com/ru/post/457808/

https://habr.com/ru/company/lamptest/blog/402071/

my choice fell on the FLIR ONE Pro Gen3 (Thermal resolution 160 Γ— 120, Optical resolution 1440 Γ— 1080), which was more than enough for my purposes. I bought it slightly in a second-hand condition, for $ 400.

Running a little ahead of myself, I will say that I could not even dream of the best for this money, I recommend it.

My stations were quite far away, and given the quarantines and other "amenities" of the year, there was no way to go there quickly, and the desire to try something other than cats

The Sphinx can be seen from afar
The Sphinx can be seen from afar

. , , . 

, 50 , 5- ,  . 

, , "". 

, -

punched Schottky diodes

punched Schottky diodes
punched Schottky diodes

.

poorly compressed input to the inverter
hot interpanel contact
diode box is heated
late with the test for a week
poorly compressed input to the inverter

, . , , . (junction box), , .

modern diode box

.

autopsy revealed that the patient died from the autopsy
,

, ( 1/3), β€œβ€ . , , . 33% , . .

, , , .  , , , (, , ). 

, - , .

, , .

every spot is panel damage, no shading, station less than 6 months
- , , 6
the same panels, close-up
,
the station is one month old

- , ( ), . , , .

, , . , . , -  .

.

, , .

(EL imaging solar cells), EL-, .

, (Near Infrared), . 

, , , . , , , .. 

, , :

-  / . , .  , , - 80V 10A, ~80$.

2 punched diodes
2

, , 1000, . , , .

- , , (NIR).

, , Raspberry PI, . - InGaAs sensor.

CCD . ~950 . - , . 

, , .

Olympus  Β΅Digital 800, . , . , .

, , . , .

CCD - , , , , . β€œ ”, , 10- , (WIFI, pc live view, ..)

- BSI-CMOS. , . ,   CCD CMOS, . 

.

, , . https://www.dpreview.com/products/search/cameras#!

, , . β€œβ€ Samsung NX mini, BSI CMOS 1β€³, Wi-Fi, RAW,  110$ . Wi-Fi , . 10 , .

, , 2 , 0,5 . . , .

, β€œ ” - , " " . , , , , , . ,  , 2 , .

, , , , , .

, .

home test
this is how punched diodes look like
this is how punched diodes look like

, .

cell damage
so the same area is seen by a thermal imager
everything is ok here
everything is ok here

, .

"snail trails"
"snail trails"
peeling off the backing
there was only one step before the fire

, - , , , - , .

- , - (IV Curve). 1000$, , - https://www.instructables.com/IV-Swinger-2-a-50-IV-Curve-Tracer/, . , .

, , . , . , " ".

, , .

, / ,

https://iea-pvps.org/wp-content/uploads/2020/01/IEA-PVPST13-012014ReviewofFailuresofPhotovoltaicModulesFinal.pdf

https://iea-pvps.org/wp-content/uploads/2020/01/Review on IR and EL Imaging for PV Field Applications by Task_13.pdf




All Articles