The are many possible ways to damage a solar panel. The most common seems to be broken glass, from rocks, bullets, a fall (loose bolts on the mount!) or large hail. If you are able to remove all the broken glass, it might be possible to replace it with a new piece -- but the hard part will keeping water from condensing inside and fogging the new glass. It's also possible the panel will still work even with the broken glass in place, though with reduced output from the 'shade' from the broken glass edges. So don't just just throw the broken panel away -- at least check it for output first!
Another common problem is loose solder connections--some connections are intermittent and cut in and out as the panels heat and cool. You may notice cells cutting in and out if you sharply rap the panels with your hand. It may be necessary to cut through the soft silicone in which the cells are embedded to access the backs of the cells for repair. Windy Dankoff's report, reprinted below, shows how to diagnose which cells have bad connections that must be resoldered.Individual solar cells are somewhat difficult to solder to. Self-adhesive metal stained-glass tape makes for the easiest soldering job--the tape holds itself in place while you solder. Use electronics solder that contains 2% silver. One side of the cell is (+), the other (-). Cells are connected in various combinations of series and parallel, depending on the intended use--our broken panels are made to be connected with 3 or 4 panels in series to charge 12 volt batteries. Another product that may be useful is conductive epoxy. It's expensive (about $20 for a small tube) but may do the trick for connecting to the back side of solar cells in a broken panel. It's available from Digikey Electronics. We will be experimenting with this product soon and will post the results on our site. An update -- one of the members of discussion board says conductive epoxy does NOT work --
Posted By: Opa
Date: Tuesday, 4 March 2003, at 12:16 p.m.
Otherpower.com mentioned using conductive epoxy on their page about solar panel repair. I tried two different (expensive)tubes and they just don't conduct when the expoxy is dry, defeating the purpose of the product. Secondly, after it hardened it is very hard to remove. What DOES work is so-called 'SilverPrint'. That worked right away. Conductive epoxy may have a purpose somewhere but surely not repairing solar panels.
Another update -- this one says it DOES work ---
I just wanted you to know that I had read your article dealing with solar cell repair using electrically conductive epoxy. You had one contributor who claimed it would not work. I can categorically assure you that it DOES work. I used a silver impregnated epoxy made by Chemtronics. It is called CircuitWorks Conductive Epoxy and I used it to attach the positive wire to the metallic film of my photo panel and it's electrical conductivity is equal to a normal solder joint, but it works where solder would not. It's strong and waterproof. Doug Austin, TX
Here's some advice from an expert, reprinted with his permission:
Troubleshooting a PV ArrayCopyright © 1999-2001 by Dankoff Solar Products Photovoltaic modules are so reliable that we forget that things can go wrong! The real world imposes temperature extremes, lightning and static electricity, moisture and wind stresses, as well as imperfect manufacturing. Here are some suggestions for testing and troubleshooting.
Selective shading test - If the array is in a parallel or series-parallel configuration, this trick will help you locate a fault without disconnecting any wiring. Find an object that is large enough to shade at least 4 cells. (A cowboy hat will do.) Shading just a few cells will drop the module's output to less than half. With the array connected and working, monitor the current (or in the case of a nearby solar pump, just listen to it). Now, shade a portion of one module. You should see the current should drop noticeably (or the pump should slow down). If the current does NOT drop, then the module that you are shading is out of the circuit. Look for a fault in the wiring of that module, or of another module that is wired in series with it.
Fading in the heat
Occasionally somebody complains of reduced array output when the sun is hottest. Heat fade shows up most severely in battery systems. If the difference between the array voltage and the battery voltage approaches zero, then current flow can drop nearly to zero. This can also cause a solar pump to produce less than it should.
The voltage of a PV module normally decreases with temperature rise. PV manufacturers document this by showing several lines on the IV curve (the graph of amps vs. volts), or by stating it in volts per degree of deviation from 25°C (77°F). Nominal "12 volt" PV modules are designed to sustain good current flow all the way to 17 or 18V at 25°C. This allows for voltage drop at higher temperatures. If heat fade is severe, it MAY be caused by weak PV modules or by any other weak links in the power chain, including undersized wiring, poor connections and controller losses. Here are some tests to isolate these factors.
First, you can confirm heat fading by cooling the array with water while the system is operating. Monitor the current. Does it rise to normal? If so, you need to determine where the voltage drop is severe. Connect a voltmeter directly to the PV array (or it's combiner box). Disconnect the array from the controller, in order to read the open circuit voltage. If it is less than 18V (relative to a 12V configuration), then part or all of the PV array may be defective. The selective shading test (above) can help you locate weaker modules in an array.
Next, reconnect the array to the system. Under good sunlight, test for voltage drop in the wiring by measuring the voltage at the array, and then again at the controller input. Note that voltage drop in wiring will increase in proportion to the current flow. Next, test for drop in the controller by measuring the voltage at its PV input, and then at its battery terminals. Remember, if the battery is fully charged, the controller SHOULD drop the voltage. If that is the case, you can bring down the battery voltage by turning loads on. When the battery is at less than 13.5V (relative to a 12V system), the controller should allow full current to flow.
If voltage drop occurs at a single point (at a connector or within the controller) then concentrated heat will result. You may feel it, or see signs of heat damage. If voltage drop is evident at the loads (dimming lights, low voltage disconnection when batteries are not low) then check for corroded battery connections (see "Batteries: How to Keep Them Alive" in SunPaper 1, or at our website).
Years of temperature cycling will occasionally cause a screw to loosen, or metal to distort. This can be caused by poor workmanship and/or inferior materials. Add a touch of oxidation and corrosion, and you get electrical resistance. Now, keep the current flowing and you get even more heat. When you repair overheated connections, replace all metal parts that have been severely oxidized. In worst cases, an electric arc will jump a gap, melting metal and burning insulation to a char. Charred terminals on PV modules can be bypassed by soldering a wire directly to the metal strip that leads to the PV cells.
Most PV modules have bypass diodes in the junction boxes, to protect cells from overheating if there is a sustained partial shade on them. On rare occasions a diode will fail, usually as a result of lightning. Most often, it will short out and reduce the module's voltage drastically. (A shorted diode will read near-zero ohms in both directions.) If the module is in a 12V array, there is no need for the bypass diode so you can remove it. In a 24V array that is unlikely to experience sustained partial shading, you can remove it. In any other case, replace it with a silicon diode with an amps rating at or above the module's maximum current, and with a voltage rating of 400V or more.This article on diagnosing broken solar panels was used by Otherpower.com with permission from: Windy Dankoff
Dankoff Solar Products, Inc