Hoping to get some advice regarding an electrical fault I had this morning. Riding to work the gauge cluster started behaving oddly. Lights coming on and off, gauges reseting. The engine continued to operate as expected with no interruption.
Eventually the odd behavior stopped and things returned to normal, except the speedometer was not responsive. The trip meter is also not recording distance. On startup the speedometer needle still briefly pings to the max value as expected, and all lights on the gauge work as expected, just no speed.
Looking at the main connector on the left side of the fairing, there appears to be a short. This is a 2004 DL-650. It's been with me for ~70k miles now. This is the first real fault I've ever had with the bike and it still runs beautifully - pretty amazing engineering.
As notacop said, it's an overly "loose" connection. Tin plated connections, as found in that and other cheap connectors, need to have significant pressure between contacting surfaces to impede growth of tin oxide into the space that one would like to remain conductive. That connector might have enough pressure, but between temperature cycling and that particular pin pair carrying a lot of current, there is regular movement between the mating surfaces, allowing gradual growth of the oxide inward into that contacting area. As that happens, the resistance goes up, hence the temperature excursions increase, promoting even faster oxide growth until there is so much contact resistance that the pins get hot enough to char the housing, (as you can see.)
A good way to preclude or at least forestall this process is to get new pins to replace the now-damaged ones, (one in each half of the mated connector pair), then apply a small amount of DeoxIT
where it can wick into that contacting area. These substances greatly slow that oxide growth. (The DeoxIT manufacturer much overstates the virtues of their product, with claims far beyond reality, but it does perform the oxygen blocking function. They are selling into the audiophile market where snake-oil-style marketing is the norm.)
I would be tempted, when repairing this failure, to put another pair of pins into one of the unused positions in the housings, then parallel the wire going though the fixed connection with another going through the newly added pin pair. One contributor to this failure is that Suzuki is asking that connection to pass more current than it can reliably handle in the long term. By paralleling connections, you will halve the current in each pin pair and reduce heating (in each pair) by a factor of 4. We see that failure often enough in this forum that I would be willing to bet that the current rating is being exceeded, even though I have not looked up the connector specification or measured what current passes in that position.