Robert
R7 & R8 (both 4.7k) looks like a standard divider network. I've built a multitester for the opto-isolators and found a couple of those bad, also.
No tester for the TOP224 yet. I would like to have an actual schematic, instead of the many UC3845 / UC3842 partials comparison. Still waiting for 3845's and 3842's didn't work. Not the timing but the voltage turnon difference...