Some of my bits not setting turned out to be bad parts on the
front panel! 15 years ago I installed LEDs/resistors in place of
all the (mostly burned out) incandescent lamps. One had a cracked
composition resistor, and the other an open (base-emitter) driver
transistor.
As I posted earlier, the problem with the core not reading/writing
bits 5-11 was indeed the MEM ENABLE signal from the M617 in slot
A09. But the 7440 was good for a change (not one of the bad batch
with 7005 date code). Turns out the wire from that pin on the
backplane was installed too tightly at the factory and had shorted
through the insulation to a ground pin. Now the core works, and
CPU executes various test programs; SR, AC, MB, MA all working
fine... EXCEPT it won't read/write on any of the 128 bytes of page
0! Going to try to figure that one out tonight...
Another strange find - the PROTECT key did not, in fact, protect
the upper page (7600-7777) of core. I found a factory-appearing
wire on the backplane, jumpering the output of an inverter
directly to ground! Some idiot probably was bothered by noise
giving false PROT errors or couldn't figure out that the switch is
off when in the down position, and just shorted the signal out.
Now that works, too. Didn't fix the page-0 problem either.
The DF and IF switches do nothing (i.e. flipping them does not
bomb a running program, or prevent reading/writing from the front
panel switches), but I think that's normal without the extra 4K
installed.
-Charles
Show replies by date