Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!oliveb!pyramid!athertn!paul
From: paul@athertn.Atherton.COM (Paul Sander)
Newsgroups: comp.sys.apple
Subject: Re: Apple 2e doesn't see my 6522's !!
Summary: Need 2MHz parts
Message-ID: <12355@athertn.Atherton.COM>
Date: 6 Sep 89 22:48:13 GMT
References: <8909060935.aa25446@SMOKE.BRL.MIL>
Organization: Atherton Technology, Sunnyvale, CA
Lines: 54

In article <8909060935.aa25446@SMOKE.BRL.MIL>, ROBBEL%NKIVXH.SURFNET@HASARA5.BITNET writes:
> Hi,
> 
> I'm having this trouble with a parallel interface I'm building for my //e.
>
> [Stuff omitted]
>
> The design is pretty straightforward:
> 
>         2 VIA's and one 74LS138 3-to-8-decoder for address-decoding:
> 
>         A4 - A5 - A6 connected to input lines,
>         A7 to first enable (active low),
>         _________
>         IO-SELECT to second enable (active low),
>         third enable to +5V (active high).
>         Output 1 of the LS138 is connected to VIA-1, output 2 to VIA-2.
> 
>         A0 to A3 are connected to R(-egister)S(-elect)0 to RS3 of both VIA's.

So far so good.

>         As a clock I'm just using Phi-0 (I also tried inverted Phi-1 but that
>         didn't work either). No patching done to the clock.

When I built my 6522 peripheral card, this is where I had problems.  In my
Apple ][ Plus (and I assume also its successors) the -IOSEL line is low only
during the second half of the clock cycle.  Because of this, the VIA's setup
times are not met.

> The interface should be in the address range $C700 to $C7FF (if I put it in
> slot 7; VIA 1 : $C700-$C70F,  VIA 2: $C710-$C71F) but there's nothing there !!

Yep, this is what'll happen.

I never was able to repair the problem with 1 MHz VIAs without adding a
jumper to my motherboard which forced -IOSEL to widen to the entire clock
cycle.  This patch is not recommended because it probably will cause
glitches elsewhere, such as causing two card to compete for the $C800 ROM
space because a glitch enabled something accidently.

The way I solved the problem was to buy a 2 MHz VIA, wire it as you did,
but clock it off the 2 MHz asymmetric clock (I forget its name here).  This
card has worked flawlessly for me for years.

> [stuff omitted]
> 
> Robert Belleman,        ROBBEL%NKIVXH.SURFNET@HASARA5.BITNET
>                         ROBBEL%NKIVXH.SURFNET@SARA.NL
>                         PSI%02041291011::ROBBEL
-- 
Paul Sander        (408) 734-9822       | If a machine is powerful enough
paul@Atherton.COM                       | to have a DWIM button, why bother
{decwrl,sun,pyramid}!athertn!paul       | with the button?  -- Eric Black