MicroVAX 2000 with Network Interfaces problems.

[Peter Coghlan]

Mattis Lind wrote:
>
>I have a MicroVAX 2000 that has suffered from a leaking battery. It has
>been cleaned and seems to be mostly working except for the network
>interface.
>

(Mine is a Vaxstation 2000 and the serial interfaces suffered instead)

>
>The startup reports
>
>?? NI       0011.700E  V1.1
>
>
>for the network interface.
>
>
>The switch is set for thin ethernet, and a 50 ohm T is in the connector
>with two terminators.
>
>
>What does the failure code mean? I cannot find any document identifying
>different failure codes. It seems that the software finds the version so at
>least something is working.
>
>
>Anyone that has an idea what is wrong?
>

My VAXSTATION 2000 and MICROVAX MAINTENANCE GUIDE EK-VSTAB-MG says that the
error code is of the format 00XXYYYY where XX is the number of retries over
the ethernet cable and YYYY is one of a number of codes listed in table E-17.
Table E-17 says 700E means "RX failed".

If you also have other error codes, I can look them up too.

>
>I will try to examine the content of registers at 200E0000 and 200E0004 and
>also if the ROM is accessible properly at 20100000 - 2011FFFF. But other
>than that, what can be done?
>

Try the AUI interface in case only the thinwire interface is broken.  It may
require a cable with a specific plug that can get into the space available - if
you don't have a suitable one a short piece of ribbon cable with a male and
female DA15 crimped onto it might be helpful.

I'd also suggest checking the switch - on mine the slot it protrudes through is
only just long enough and it has to be pushed quite hard to one end for it to
make contact properly.  Also, the contacts could be corroded from the battery
leakage, as could the contacts on the connectors on the little ribbon which
goes between the AUI connector and switch on the case and the main board. (I
really hate when I've struggled to put the machine back together and forgotten
to plug this one back in beforehand...)

Finally, it might be worth putting a scope on the thinwire connection while
still terminated and running the tests to see if the interface is managing to
drive the network.  If not, the failure would seem likely to be in the
circuitry common to both transmit and receive paths close (electrically if not
also physically) to the thinwire network connector.

Regards,
Peter Coghlan.