After a couple weeks of less-than-usual power generation from my solar panel array, I decided to pull out the ladder and start testing panels. Turned out, the Coleman string was dead.
Voltage read fine, but a current test showed 0 amps.
I narrowed it down to 1 of the 2 Coleman panels, unmounted it, and lugged it down the ladder to take a closer look, thinking “it’s either the diode or a toasted cell…. please be the diode…”.
Turns out, the diode was dead. Really dead. Massive-crack-running-through-it-dead.
You can click for a larger image.
The good news:
- The catastrophic diode failure was nice and visually obvious. No guesswork here, and no need to actually test the diode. If only all components failed so spectacularly!
- It’s a fairly simple fix.
You can see the replacement diode installed above in the black box located at the rear of the solar panel (4 small philips screws to open).
If you have a replacement diode and are handy with a soldering iron, it’s a pretty straightforward fix. Pull the 2 nuts, slip off the wires, slip off the diode assembly, de-solder the old diode from the ring connectors, and solder the new one in. Be mindful of the diode orientation (white/grey strip on the diode goes to the positive wire here).
If you have a replacement diode but are not handy with a soldering iron (or are lazy), after you have the old diode out, the leads on the new diode are probably long enough to curl around the threaded shaft, so if you’re careful you can probably get it in place and tighten the nuts enough to keep things together. Again, be mindful of the diode orientation (white/grey strip on the diode goes to the positive wire here).
If you don’t have a replacement diode, you could skip the diode to get the panel working for the short term. Either use a jumper wire where the old one was (to connect left to center), or move the “tabbed” wire from the center peg to the left peg (connected directly to the red positive wire). The down side here is that without a diode, your battery/etc may discharge into the solar panel at night which isn’t so good… if you’re using an array of panels you could run into other issues here too. Really, a diode (known as a “blocking diode” here) is installed to make sure power only flows out of the panel and not into it! So at the very least, if you’re skipping the diode, disconnect the panel from whatever it’s hooked up to in the evenings until you get a diode.
As for what diode to use, most of these panels seem to come with standard run-of-the-mill 10A 1000V diodes (10A10 diode in this one), However, if you want to increase the panel voltage slightly (up to half a volt or so), and reduce power wasted in the diode, you can look at using a “Schottky” diode. They’re more efficient but you can’t get them in really high voltages. I used a 10A 45V Shottky, but as long as whatever you get has a minimum voltage of 6A or so (realistically you’ll find a 10A), and a minimum voltage above the peak your panel will see (say… 25V if you’re just using single panels on a battery, 45-50V if 2 panels in series with only the final panel w/diode, etc) you should be good. If you’re doing something really exotic you might need something with a higher voltage, but for most of the common use cases you shouldn’t.
The final result:
Over 5A short-circuit current with some decent sun.
All was now well! Panel went back up, and power is back to where it should be.