PDP-11/05 debugging tips?

Hi all -- As I mentioned a few days back, I recently picked up an 11/05 that's in need of some troubleshooting. This weekend I've started getting more acquainted with the details of the CPU set (reading through the schematics and the various manuals (which are quite thorough, fortunately.)) Before I start getting too deep with this, I figured I'd see if maybe I'm overlooking something simple -- I'm relatively green with Unibus systems. So here's what I'm seeing: - At power-up, the system comes up in a halted state and I can load addresses from the front panel switches into the address register, and these are correctly loaded -- no stuck bits, etc. - Toggling "Start" causes the "Run" light to go on and the system seems to be doing... something (the address/data lamps flicker). Attempting to halt the machine at this point does not work, but if I have the "Enable/Halt" switch set to "Halt" and hit "Start" again the machine does stop and the "Run" light goes out. - Hitting "Exam" while halted has no effect (no data is loaded, addresses are not incremented), and seems to lock the machine up -- I can no longer load addresses via the front panel. Hitting "Start" still puts the machine in the "Run" state outlined above, and once halted again, "Load Addr" works again. - Hitting "Cont" while halted puts the machine back in the "Run" state regardless of the state of the "Enable/Halt" switch. So it definitely looks like the Run state control logic should be looked at... The machine's configuration is very basic: the CPU set (M7260/M7261), an 8KW core board set (G110, G231, H215), and an M903C Unibus terminator. I've verified that these are all in the correct slots. Bus grant cards are in the D slot of all four Unibus slots (and inserted in the correct orientation, I hate those things...) Thus far I've tested the power supply and voltages are spot on with acceptable ripple. AC LO L and DC LO L are correctly going high after power on, so that shouldn't be causing issues. I've gone through and cleaned everything including the edge connectors. (I left the core plane undisturbed, it's already been banged around enough...) I double-checked the backplane for bent pins that might be touching other pins. I've checked the processor clock, and it's running properly and at the correct rate (310ns). I've run the machine with just the processor set with the same behavior. Any other simple things I should check before I get seriously involved with this thing? Fortunately I have a set of KM-11 replicas I got from Guy Sotomayor a few years back so that will help immensely in debugging... Thanks as always, Josh