MicroVAX debugging

[jmdhuse]

Hello All,

I am starting a journey with a microvax... I have a KA650 CPU and two 16MB memory boards installed in the first three slots of a H9278-A backplane. I have the CPU patch panel and have wired the console output to my laptop via an RS-232-to-USB device. Upon power-up, the console displays a header message, seems to run a few tests, then hangs. A couple of things confuse me - it does not always hang at the same place, and the hex display in the patch panel is not always the same. Here are three example power-on console message:

KA650-B V1.2/0123
Performing normal system tests.
30..29..28..27..26..25..
This one hangs with "A" in the patch panel hex display.

KA650-B V1.2/0123
Performing normal system tests.
30..29..28..27..26..25..
This one hangs with a "C" in the patch panel hex display.


KA650-B V1.2/0123
Performing normal system tests.
30..29..28..27..26..25..

?04.49 2 12 FE 60 0000
20080108 0000000B 000000FF 00000010 00000002
00000000 00000000 00000000 00000000 2004A7F4
00000000 0080017C 00000000 00800200 00000000
00000000 00800000 20080144 00000000 80028180

24..
This one hangs with an "F" in the patch panel hex display.

Can someone suggest where to turn to start to figure out the meaning of these messages? Are the console messages and the patch panel hex display related to each other, or do they present independent information about the state of the system? Does it matter that I don't have a battery installed in the patch panel? I've looked at the firmware section of the KA650 Technical Manual, but it does not help much.

Thanks in advance for any help.

Cheers, Jon.

 

[m_thompson]

See section 4.1.2 of the KA6S0 CPU Module Technical Manual. There is a copy here.
Section 4.4 has lots of details on diagnostics.

"A" in the patch panel hex display: FPA Tests
"C" in the patch panel hex display: SSC and ROM tests
"F" in the patch panel hex display: Initial state on power-up, no code has executed

 

[jmdhuse]

So here's the confusion: If "F" means no code has executed, then how, in the 3rd example, does it get all the way to test 24? Obviously, code has been executed. Of course, the confusion may just reflect the defect in the CPU, and therefore it may not make sense.