Thanks everyone for the input. When this happened, all that was on was the radio and intercom so no big load. However, up until this day, I have done mostly touch and goes, stalls, etc. so the engine never ran for more than 10-15 minutes or so a full throttle. This day I was out messing around a river so I was at 5500 rpm for over 20-30 minutes so the alternator was putting out full power longer than any other time. I am kind of wondering if this gave the bad connection enough time to to get enough resistance to melt the insulation?

I think my game plan now is to re-do the connector. I think I am going to solder them to make sure the connection is good. This part of the electrical was done by another builder so I didn't have any input other than checking it to make sure it was right according to the manual.

The one thing I noticed now is it seems there should be a switch to turn the output from the alternator off. Kind of like a "crow bar" but something I could turn off in case this happens. I always thought that when the master is turned off, it shut all the electrical off but now I realize this is not the case since the alternator is still powering the regulator. After I saw the smoke, I shut the master off and reduced my rpm. The smoke subsided and most blew out during my 2000 fpm slip to the runway. The crazy thing is, in theory, turning the master off did not stop the alternator output so the wire should have continued to burn. I am wondering if reducing the throttle thus reducing the alternator output is what stopped the smoke? Has anyone put a switch in? If that is not a option, is the crow bar just a circuit breaker that I could buy at an auto parts store? I guess a panel mounted C/B would essentially be a switch. If this ever happened again, I could kill the master and pull the breaker. This might be a better option?