I got this with a very low bid on EBay. The device was reported as “powering up but display stays black”. It is worse than that really: there is no activity at all from the CPU board. The device came with all usable slots filled:

• BN9305/00.26 Protocol Analyzer module, 2x (but slightly different boards)

• BN9305/00.35 16MB FDDI Analysis module, a 2-slot part

• BN9305/00.17 2Mbps WAN Analyzer module

• BN9305/00.08 Ethernet 802.3 interface

It also has marked the options:

• BN9315/90.05

• BN9315/90.08

Attempted repair

Opening up the device showed its interior. It consists of a device specific backplane board in which boards are plugged:

(tbd)

It is very well constructed but a terrible amount of work to get things to a state where I can actually measure what is happening; I needed to completely remove everything from the frame so that in the end I’m left with this:

The power supply seemed to be in good condition; no leaking capacitors and all voltages seemed fine. It outputs +/- 5V and +/- 12V.

Next thing to check is the PC board:

According to the only datasheet I could find ( ) this board should contain:

• A 486SX/33 processor

• 8MB of memory

(although the sheet is confusing; the German portion mumbles about a 200MHz Pentium, and 32MB of RAM).

The RAM is the first oddity. There are 16x TMS44400DJ RAMs which should be 1MBx4bit RAMs, making for 8MB. But there are also 8 81C1000A RAMs which are 1MB x 1bit RAMs. Perhaps this is video RAM.

The board further contains:

• FDC37C651QFP Super IO controller (Floppy, IDE controller, parallel printer port and 2x serial ports)

• 2 resonators:

  • 14.31818 MHz which is not resonating despite pin1 being +5V. This one seems dead. Datasheet:

  • 24.0000 MHz which resonates just fine.

• VL82C486: a single chip 486 controller chip which contains most logic needed for a 486 level PC:

  • 2 82C37A DMA controllers

  • 2 82C59A Interrupt controllers

  • 82C54 Timer

  • 82284 Clock generator and ready interface

  • 82288 Bus controller

  • Memory refresh controller

  • Port B and NMI logic

  • Bus steering logic

  • Turbo Mode control logic

  • Parity check and generation logic

  • Datasheet here:

• VL82C113A: SCAMP Combination I/O

  • 146818A compatible real time clock

  • 64 bytes of battery-backed CMOS RAM

  • AT compatible keyboard controller and PS/2 mouse

  • SCAMP compatible processor to ISA bus address latches and buffers

  • Datasheet here:

• AV9155-01CW20 Frequency Generator

  • Generates clocks for 486 systems from a 14.318MHz crystal - bingo, this thing provides clocks for the CPU, sigh.

  • This is the 01 variant for 16MHz BUS

  • Datasheet here:

Probing the board shows that everything is completely idle; signals on the RAM chips are mostly all high, the buffer chips visible on the board (ABT2244, ABT16245) show no moving signals - it’s dead, Dave.

The clock is clearly not there, and one reason for that is the +5V on pin 1 of the resonator. So challenge is to find out where that comes from…