On Fri, Sep 29, 2017 at 7:19 PM, Ethan Dicks <firstname.lastname@example.org> wrote: > On Fri, Sep 29, 2017 at 4:20 AM, Francesco Messineo > <email@example.com> wrote: >>> I will also change the timing resistor and tantalum capacitor on the >>> 555 reset circuit, I've noticed the reset signal isn't very well >>> behaved (it resets randomly and sometimes doesn't release the reset >>> line). I think the tantalum capacitor is shorting randomly to ground >>> or the 1M resistor is intermittent. > > It _should_ be a ceramic cap there (not all hard-shell shiny caps are > tantalums) but if it's a large value and electrolytic, it's probably a > tantalum. I'd be less worried about the resistor failing, but that's > an easy test. it was a 1uF tantalum capacitor and it's now a ceramic one. Reset worked flawlessly since the substitution. The 1M resistor was good enough and I left it alone. > > Either way 555-based reset circuits do age poorly. > >> update: changing the UD8 socket, did bring the system to the basic >> prompt, with correct bytes free count. > > Nice. Yes. Multiple bad sockets - not surprising. > >> I then restored the two CAS resistors (I had lifted them to swap RAM >> banks) and now I have a completely black screen, not even the garbage >> screen if I remove the UD9 ROM :( > > Hmm... that's really odd. indeed. > >> Looks like I now have to debug the video signal path. > > Yep. > > Check Φ2. I am pretty sure you don't get video if the CPU is not > installed and cranking Φ2 out. Maybe you also have a bad CPU socket? > You don't have to be running good code - which is why you get random > chars when the KERNEL ROM us pulled, but you _do_ have to have > something generating Φ2 to feed into the video circuit. It's also > possible something failed in the primary timing circuit? Rather than > bad/missing output from CPU, the input clock is bad? I've checked the main clock input to the CPU and it's there. I will check the clock output and the video and sync signals going to the monitor too. I just need to find the next free time to do it. Frank Message was sent through the cbm-hackers mailing listReceived on 2017-09-29 18:03:09
Archive generated by hypermail 2.2.0.