RE: C128 'VIC tower' mod board

/RAS has a low time of 292ns
/RAS_FF has a low time of 375ns

It seems that /RAS has a sufficient dwell time so it must be more subtle than that. I have looked at things like TRLCH and TCLRH and they look to be within spec with either version of /RAS. I am sure there is something subtle I am missing.

Jeff

-----Original Message-----
From: Francesco Messineo <francesco.messineo_at_gmail.com> 
Sent: Sunday, July 19, 2020 2:37 AM
To: cbm-hackers_at_musoftware.de
Subject: Re: C128 'VIC tower' mod board

On Sun, Jul 19, 2020 at 5:05 AM Jeffrey Birt <birt_j_at_soigeneris.com> wrote:
>
> It seems this mod was not done on the original Rev 6 PCB but only on Rev7. The only difference I could see when comparing two Rev7 boards, is that the one with the mod board has slower, 200ns, DRAM.

that might be the reason to lengthen the /RAS pulse. According to DRAM datasheets (for example the MCM4164), the low /RAS pulse needs to be 150ns min or 200 ns minimum, depending on the DRAM speed (150ns VS 200ns grade).
Measuring the original /RAS low pulse length would give the final answer (if it's < 200ns or not).

Frank