UNISYS U6000/50 Help needed! :)

Hi all,

here at our computer museum in Palazzolo Acreide, Italy, we're trying to restore this ancient UNISYS server: (some picture here)

wiki DOT museo DOT freaknet DOT org/?title=UNISYS_U6000/50

(sorry I can't post any URL)

We have casual memory errors but we tested every single SIMM and they are OK. We tried everything; no ripple on power supply, RIFA capacitors changed on power supply, other electrolytic capacitors checked and tested, hot / fresh air on some part of the boards... but sometimes the system run smooth, and sometime it gave us weird errors and lockups. :wall:

We removed every unnecessary board; changed the VGA board, clean everything and look for contacts, bad joints, cables and whatsoever. We also made a complete dump of our 650 Mb SCSI hard disk to save data.

Our goal is to restore this UNIX System V. Release 3.2 Version 3.0 back running again! :slight_smile:

The meaning of our Museum "Informatica Funzionante" is to restore pieces and to let people have fun(*) using them. So we're restoring it :slight_smile:

(*) under our control, obviously! :slight_smile:

We have no spare memory board to swap and make other tests, so here we are asking if somebody can have some boards around for this kind of machine.

Another clue can be to create a similar machine to boot this O.S. but it seem very weird - the SCSI HD image show that there's no boot record so we assume system boot via some hardcoded program into the BIOS. Board start as a Phoenix 386 bios but seem hardly different from any known pc from that era. Also, the Western Digital controller chip seem peculiar and I've find no SCSI controller with this same chip - and the kernel, I presume, is compiled for this controller... :confused:

Maybe can anyone help us? Any spare board, any old UNISYS similar server in the junkyard, or any kind of hint?

Thank you all :slight_smile:

Gabriele

Check the cas and ras timings against spec, maybe it expects too fast an access. Also, check for adequately frequent CAS refresh across all CAS addresses.

Some early minis had hand/paper tape pn console entered bootloaders, but usually a SCSI era machine had a rom bootloader. Some DOS's booted from tape or card for a long time.