Reanimation of the "yellow" Scorpio

Hello, All The car came to me in “as is” condition - it is unknown how long it had been in last time operational and what happened to it. The following is observed: 1) There are spurious sawtooth pulses on the /RES line. Amplitude and the repetition period change when desoldering/soldering the capacitance by 10 µF to the total. 2) When the container is soldered, the test ROM is not performed, the “mattress” is correct, but unstable - apparently due to the fact that /RES twitches quite intensely. 3) When the container is sealed, /RES twitches slightly less and rolls higher log 1. - the “mattress” is stable, with great difficulty I manage to launch the test ROM. Tests for 48k they pass without errors, tests for 128k complain about the configuration port - Apparently, it loses content due to the same “buggy” /RES. 4) I tried to disconnect D48, D49 and D59 from /RES one by one - the line behavior did not changes. Like this. What could it be? I don't have much experience with repairs. Spectrum-compatible computers - I would be very grateful for any useful information information on the topic :o

Hello, m.d. 1. Check the power supply. Maybe it itself is sawtooth-shaped? Is the power supply working? 2. Replace the container. 3. What else is hanging on the reset? Turn off outputs one by one. Unfortunately I can't say anything more without a diagram.

Hello Sonic Scheme: http://zx.pk.ru/attachment.php?attachmentid=519

Hello, m.d. As it turns out, the source of the disturbance is the CPU. Blocked it carefully power supply (1000 µF and 0.01 µF in parallel directly to the 11th and 29th legs), 26th leg disconnected from the circuit and hung a 12k resistor + +5V diode and a 10uF capacitance on it to general - jumps (in the range of 2-3 V), without capacitance after a while (2-5 c) starts to make a lot of noise. Most likely the transition died on the 26th leg... Am I right? Percent change?

Hello, m.d. md> Most likely the transition on the 26th leg died... Am I right? Percent change? that is, in the complete absence of anything on the 26th leg, is there noise? then definitely change the percentage. why, just like that, the input (not the output, look at the circuit design) makes noise. Although, I’m practically sure that there is snot somewhere that gives a tip.

Hello, Mikka_A Well, in general, in order for the TTL input to “jump,” the snot must be powered. And the food is on The process is smooth (checked). The average voltage on the 26th leg is 2 V (before the way it starts to make noise), and the frequency of “jumping” directly depends on the capacity.

Hello, m.d. I changed the processor... it worked: v2_yahoo; Possible; the cause of destruction on the 26th leg is the lack of a blocking diode in initial reset circuits. Every time the power was turned off, the voltage at /RES dropped very slow compared to the feeder... everyone endured it, endured it, endured it... Z80 was the first to get bored :) Next up is replacing the crib for ROM - the one that now works normally only in Even lunar days to the sounds of a shaman's tambourine...