Not clear what the OP's situation was that might be similar, but last year I accidentally fried my solar controller, where it stayed on line ok, but just did not control the voltage anymore.
So here is a WAG on what might have happened to the OP.
It turned out that mine must have had an "open FET (power transistor)"
where if the FET is shorted, the controller will not charge a battery at all, but one with an open FET will overcharge a battery.
Just how I fried it is unclear :( ,but I think I managed to break one of the rules which was "do not connect the array to the battery when the array is connected to the controller at the same time. This will damage the controller when the battery voltage reaches the Charge Termination set- point."
(I sometimes go battery- direct from array to do an overcharge on purpose. I spotted the trouble soon when the Trimetric showed things were not behaving right, so no damage done to the batteries etc.)
OK, so the OP has some fuses gone and whatever, so perhaps in the middle of all that, he fried the controller with an "open FET" situation, allowing the voltage to reach 19v, and that fried the batteries.