Extremely perplexing solar problem

[tanglewood]

Originally Posted by Rohan

Replacing MC4 connectors looks like a huge pain in the a** but I'll try it if nothing else works.

Right now, semi-update, the system is working usually, but randomly drops to 0 amps for awhile, then comes back like nothing is wrong.

This is the same system where the panels are overloading the charge controller? I think the first step is to rebuild with appropriately sized components.

[Captain Bill]

Rohan, when you measured the "charge" at the disconnected wires to the controller were you measuring voltage or amperage when you said you were getting a strong charge? To test the panels you must measure both open circuit voltage and short circuit amperage. If you test both at the controller end of the wires you will test the entire wiring system and the panels, but it is best to test the panels individually as well. If all you measured was open circuit voltage it really doesn't tell you anything if you get an acceptable reading. A high resistance connection can show excellent voltage with no load but not be able to pass any current. I once was helping a friend with a problem similar to yours that had a connection that only failed once the panel got hot, later in the morning. It would seem to charge normally for a couple of hours in the morning until the panel got hot and expanded a bit at which time a connection failed. It would pass the open circuit test, but fail the short circuit test but only if the panel was hot. It's not exactly the problem you are describing, but an internally cracked wire can show similar symptoms and would explain why it sometimes charges for a while then stops. Having 12.5v from the panels on a sunny day really sounds like a high resistance connection and it seems unlikely that both panels would do this at the same time unless there was something external like a nearby lightning strike. I had a friend whose boat was hit by lightning and both of his panels failed like this except that they never charged again. They would show normal open circuit voltage but could not supply any current under load. Before anyone comments yes this was after the controller had been replaced. I on the other hand had a nearby lightning strike that took out my solar controller, battery combiner, and LED anchor light, but the panels worked fine after replacing the controller. I would think that the problem you are having is in the common wiring since both panels output seems to be affected. I am somewhat a assuming that you have a setup similar to mine where I have an MC4 Y adapters connected to a single positive and negative wires running to the controller. If only one panel was failing or one leg of a splitter, output would only drop by half, not to zero. My guess would be the connection at the splitter on either leg. If that is ok I would suspect anywhere a wire passes through a bulkhead or makes a short radius bend around something. The easy way to diagnose this might be to make a temporary wire with an MC4 connector on one end and connect it in place of the positive wire. If the problem goes away you have Identified the problem wire, if not repeat with the negative wire. If the problem still exists you could connect to a panel and bypass the splitters. If the problem goes away then you will have identified the problem splitter. Mdenize's recommendations on the DC clamp-on meter and an infrared thermometer/thermal imager are ideal tools for diagnosing his type of issue, but a simple bypass cable can tell you a lot if you don't have access to these tools.

[Rohan]

Well, mystery solved!

It turns out the problem was a relatively simple one, and several of the comments here headed in that direction.

Although the outside looked fine, the metal pin inside one of my MC4 connectors had snapped, so it was only making sporadic contact. That explains why it was working randomly. The part I don't understand - if the metal pin snapped in half, how is that possible when the plastic housing was totally intact??

But anyway, I cut the wire, spliced on a new MC4 connector and it works fine now.

[sailorboy1]

Big cheer goes out to those that said it was the MC4 connector.

Right?

Taking those damn things apart I can see easily breaking a pin inside.

[GreenWave]

Originally Posted by Rohan

Well, mystery solved!

It turns out the problem was a relatively simple one, and several of the comments here headed in that direction.

Although the outside looked fine, the metal pin inside one of my MC4 connectors had snapped, so it was only making sporadic contact. That explains why it was working randomly. The part I don't understand - if the metal pin snapped in half, how is that possible when the plastic housing was totally intact??

But anyway, I cut the wire, spliced on a new MC4 connector and it works fine now.

Good to hear you solved it. In my experience 90% of electrical issues can be solved by looking at the connections - either corrosion, broken, etc.