On Tue, Feb 20, 2018 at 9:06 PM, Gerrit Heitsch <gerrit@laosinh.s.bawue.de> wrote: > On 02/20/2018 08:48 PM, Francesco Messineo wrote: >> >> >> yes, refresh counter was one of the first things that I checked, since >> the faults seemed to be related to the >> activity I was doing on the machine (they're not related actually). > > > You also verified that the refresh addresses reach the RAM and /RAS goes low > when they do? nope, I didn't verify what happens when /RAS is low in general, that's the next thing. The fault is really intermittent, as the PET can live for hours idling at basic prompt sometimes without a glitch and then suddendly dropping into TIM, and other times it fails during memory test at power on. So refresh addresses do reach the RAM most of the times. I think currently my only hope is finding some address changing right after /RAS edge. If it's only one or two from the same '153, then it must be that 153. Should I find all addresses moving at the wrong time (sometimes), then it must be something on page 6 of the schematic. > > > Gerrit > > > Message was sent through the cbm-hackers mailing list Message was sent through the cbm-hackers mailing listReceived on 2018-02-20 21:01:05
Archive generated by hypermail 2.2.0.