Guy’s just took a look at the schematic. Not crazy about how the write signals (WR* etc.) are driven by open collectors + pull-ups, if I were to do it again I would use one input of a 74LS08 (the other tied high) for S100 non inverted inputs – less slew. However it should not be a problem because SEL_SECOND and DO0 are well settled before WR* goes low.
If I was going to fool around with anything I would try dropping R30 /R31 to lower (or higher) values. Would have liked to use higher powered 7407 for U17 E & F, but unfortunately U17 inputs S100 lines A0, A1, this will load the bus more that I would like. I always try and have 74LS on S100 inputs – at least in my 21 Card system.
I have to stress all these CF cards don’t behave the same way. The Kingston 4GB ones I use never have a problem switching between drives. Is that what you are using.
From: n8vem...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of yoda
It really has nothing to do with the cards as the select is driven by a flip-flop - I see the same thing on my board but have not set down yet to debug what is going on but Thomas is right there is some timing problem is going on as DO0 is clocked into the flip-flop via SEL_SECOND* and WR* being combined - there is a mis-timing of the signals as that really is the only thing that could cause this - guess others have been lucky or have not tried to use the second CF card.
It maybe necessary at some point to redo his code based on the Z80 Zeta code or other ?, that handles a large number of cards...
I'm sure its a timing issues... but it may need someone to look at it in detail...
On Sun, Jun 8, 2014 at 7:43 PM, Thomas Owen <thoma...@gmail.com> wrote: