Gregory's Serversomeone else has a card and desoldered all the chips. they also used a multimeter to ohm out all the connections.
|
11 comments
   it's tough to reverse engineer this visually because traces can run underneath the pads! as long as they don't touch the plated via barrel, they're not connected. I'm glad I've got the netlist. I've already created a schematic from that netlist. it helps understand how the card works. it pretty much just buffers the address and data lines to the cable, although it also has a state machine for adding wait states to slow down the ISA bus.  @tubetime the CM-2 had a similar clear soldermask, the nearly transparent laminate layers, and an almost uniform grid of through-holes (although there are some exceptions, and there are traces on all layers)  @tubetime As a young lad my uncle gave me some boards from an IBM 360 mainframe. The PCB was like this (and unlike anything I had seen before). Full of square metal cans. Alien tech! (Unfortunately I lost it) |
this card is fascinating to me because the layout is unusual but very much the way certain divisions at IBM liked to do things. all the pads and vias are on a strict 0.1" grid. the traces are quite thin: only 5 mils! there are no traces on the top and bottom of the board: everything you see here is on layer 2.