Thanks Mark,1024MAK wrote: ↑Sun Aug 09, 2020 11:09 am Hi Andy
You say that it seems to be coming out of reset okay, have you tested any of the Z80s signals (e.g. /MREQ etc...)?
And have you done a visual check, it’s very easy to miss a solder joint on these boards because of the large number of joints required.
The transistor used for the master clock oscillator and for the clock signal to the Z80 have to operate at just the right point in their linear region. Otherwise either the oscillator will not oscillate (or will not be stable) or the transistor for the Z80 clock will not turn off fast enough (due to stored charge in it’s base junction). The result being the clock signal on pin 6 of the Z80 being continuously low.
Mark
Yes the master clock is outputting a stable 6.5Mhz and I traced this through to the output of pin3 of Z9A (cnt0). This was then going into the 2369 transistor, but I had fitted an PH2369 and this resulting in a weak sawtooth output. I replaced it with a PN2369 and this looked much better. Pin6 of the Z80 is getting the resultant 3.25Mhz (actually reading 3.33 on my scope). I haven't looked at /MREQ but I watched reset transition okay and observed the address and data lines seemed to be toggling. In fact I accidentally shorted a couple of the high address pins by accident when probing and the screen filled with A characters.
The composite video transistor issue was that I was using was a BC548AC but the output seemed to weak. Replacing it with a 2N3904 seemed to give a much stronger signal.
Next job as you say is to check for dry joints. My guess is that there is something iffy with the part of the circuit that deals with video memory reads and writes, my hunch on this is when typing on the keyboard I do seem to get random characters appear through the 'snow', and mashing the keys and pressing return does affect the random patterns being output on the screen.