I've performed the resistance checks for both pairs of TIP-50 and 2N2222A (see picture). I've measure the TIP-50 using the pins on the top side, and the 2N2222A via the bottom of the PCB. The TIP50 readings are similar for the pair; but not the 2N2222A. In fact, on the "bad" side, one pair of the legs for 2N2222A is showing 17 ohms, which is unusual given that the resistance on the same pair of legs on the "good" side shows rapidly rising resistance to OL. I'm using a Fluke 87V.
Does this mean the 2N2222A on the "bad" side is broken, hence causing the resistor to break as well? If so, looks like we might found the problem.