Perhaps the following might help Gary…
In my travels is working with the 80486-40 I noticed that with one (of my two) SMB-V2’s it would not switch back to the Z80 using the Output a 1 to bit 0 of port 0EEH.
However if I jumpered K3 2-3 so it used only 8 bit port addressing and that immediately corrected the problem.
When I set K3 to 1-2 (16 bit port addressing) the problem returned. I suspected the problem is U31B is not getting the address lines up in time on its pin 5 to clock the correct data into U33. If I use a 74F138 for U27 and a 74F10 for U18 the problem disappeared. I.e. I could always switch back to the Z80.
Could you try these in your system.
From: n8vem...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of Gary Kaufman
Sent: Saturday, February 7, 2015 9:27 AM
Subject: Re: [N8VEM-S100:6234] Re: 80386 CPU Board - Correction of Correction
Ok, see the Z80 pull ups now...
Interestingly when I was confirming that I had the correct resistors on the 80386 board I noticed that "in circuit" R1 and R8 measure 500 ohms. This makes sense for R1 with P36 jumpered 1-2 and 3-4. I haven't thought thru R8 (master reset) yet.
I tried the "external port" approach with a simple switch, but I think I was getting too much switching bounce to draw any conclusions. I'll wire up the port hopefully tonight.
On 2/7/2015 11:57 AM, John Monahan wrote:
Early morning here, coffee has not kicked in yet! Should read:-
Yes, on the Z80 board (RR5) Gary. Remember only one, 1K pull-up per S100 bus OC line. You can get away with two but you are overloading the gate (now 500 Ohms), more, then you will have severe signal slew, more, then unstable signals.
Any luck with the “external port” approach?
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.