Data General 8000-series logic in my Nova 3?!
Mark J. Blair
nf6x at nf6x.net
Wed Jun 22 02:42:15 CDT 2016
The testing that I've been doing so far to get the 6045 hard drive working on my Nova 3 suggests that the interface card receives commands over the IO channel (i.e., I can command seeks and get the expected clunking sounds from the drive). But the interface card does not appear to be responding back to the CPU so far, since attempts to read the three IO registers or the busy/done flags always return zeros. So, I'll need to move on to component-level debugging of the interface card now.
I'll need to have access to the interface card, of course. The first step was to swap the positions of my Nova rack and my VAX-11/730 to get the right side of the Nova away from the wall. This wasn't easy in the tiny, cluttered room that they live in.
Next, I lowered the Nova 2 rack units, because it was in the top rack position and I couldn't get access to all of the top cover screws to get the top cover off. Damn, that thing is heavy! I pressed my hydraulic lift hand truck into service. There was a 2U filler panel under the Nova that can now live at the top of the rack, so there will be no need to raise the Nova back up later.
With the top of the Nova accessible, I removed the quad serial mux in slot 12 to expose the component side of the disk interface card in slot 11. There are 6 empty slots under the interface card, so I have good access to both sides of the interface card, as well as the backplane.
Now I should be able to do things such as running a tight loop reading or writing a controller card register while I probe the logic. Should be simple, right? Well, it would be if I had a schematic diagram of the interface card. So, I'm doing this in hard mode. I decided to do a little preliminary trace identification on the card before going to bed tonight, and that's when I discovered that this game is in very hard mode: Most of the ICs on the controller card are marked with a DG logo and an 8000-series number, and I have no documentation about those chips yet. The busy/done flags come out of a DG 8109, but what the heck is that? I hope that they'll end up being rebadged 7400 series chips or something like that so I'll have some chance of finding replacements, but I'll need to figure out how to identify these DG chips before I can make much progress debugging the card.
I've been looking through the documentation that I have, as well as looking in documents on Bitsavers for DG gear other than the Nova 3 in hopes of finding anything identifying these 8000-series chips. I haven't found the decoder key so far. If anybody out there in cctalk land knows about DG-marked 8000-series logic chips, I would appreciate any help very much!
Mark J. Blair, NF6X <nf6x at nf6x.net>
More information about the cctalk