EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: alank2 on February 16, 2023, 03:18:36 am
-
I'm studying this schematic from an old computer (NEC PC-6001). I am confused by something, AND gate #12 on the top left of it has a pin 9 input. If you follow it around it connects through two inverters as inputs (#11 pin 3 and pin 5) as well. I don't see that it is an output to anything, is that right? Does that make sense?
-
That has to be a mistake.
-
Well, it's LS family, so it'll pull high, that's something I suppose. But yeah, sounds like an error. Do you have a board to inspect, trace the circuit?
Or, if this is some bus interface glue logic, maybe trace its function back to the bus signals, and forward to /CEs and whatnot, infer what it should be? (Still only an assumption of course, but can be an educated one.)
Ed: which... https://upload.wikimedia.org/wikipedia/commons/3/39/NEC_PC-6001_Motherboard.jpg IC12 is clearly an 'LS08, sounds about right. Not a good angle to trace though, let alone a bottom view(!).
Tim
-
My guess is VRAMSW2 and a missing dot.
Upper OR is enabling _RAS and lower something not visible, _CAS maybe.
With a dot both inverter operations are also pretty equal.
-
I had redrawn it and to me it looks like pin 11 of #25 is DISABLE for input pin 10 of #12 - some 1 bit memory. Since to reset the whole circuit, the both inputs need to be zero.
-
The pin that this all drives it a /RD pin they call /V.RD for external memory on a cartridge. I found another place in the documentation about this pin and it says 0x8000-0xBFFF so I'm guessing the gates up to it are selection logic for that. Thanks everyone for working on the mystery!
Page 5:
https://archive.org/details/pc6001schematic/page/n3/mode/2up
What do the 74157's do under the dynamic RAM? I can see they are selecting between A and B, but why?
-
No, it does not make sense.
And it is a perfect illustration of why I do not like the style of schematic that does not use the "bridge" symbol where two wires cross but do not connect. They probably omitted a dot, indicating a connection, somewhere on that wire.
IMHO, a PROPER schematic uses both the dot for a connection and the bridge for no connection. And two wires that just cross indicates a MISTAKE.
LAZY ENGINEERS!
-
Well, there's an easier way to do that... 4-way junctions just never connect, only make 3-ways. The excerpt shown doesn't have any obvious 4-way candidates so their drawing style/policy is inconclusive from here, but it may well be drawn in this way. Which makes the hanging wire all the more perplexing -- a drafting error seems likely.
Which, this probably just predates schematic capture? I mean for broad use at the price point this would've been designed with. Engineering price point that is. (Yes they've had it around for a long time, but it doesn't count if you're talking IBM internal tools, right...) The layout might've been automated? But it might've been driven from a netlist rather than schematic capture; and even if so, the documentation might be done separately just because the schematic output wasn't suitable or compatible with their preferred documentation style. Anyway, it's whatever.
Tim