Seems like repair threads have been a bit sparse lately. Good to see a lot of the usual suspects here.
I do not have one of those magnifiers, but am increasingly feeling the need.
Looking over the sparkle and RGB circuits, I noticed that Atari seems to have missed the fact that BD0-BD3 make an appearance on sheet 9A. That's how MH (and Tempest) can load colors into the color RAM and then rapidly change the colors of existing vector by changing what's in color RAM. It's write-only since the RAM used is an 82S25 which has separate input and output lines, and the outputs go only onward to the final RGB output stages.
Note: Tempest and Quantum also have color RAM in the RGB output section, but only a single chip. MH has two color RAM chips in parallel with the enable line run through an inverter at 9J so when one is selected the other is off and vice-versa. Anyone know why?
There does seem to be COLA0-COLA3 going into both of the RAM chips, as well as activity on the ingoing BD0-BD3 data lines and the outgoing D1-D3 lines. When not in sparkle mode, all colors look good. According to TM-252 the Color Intensity Test Display (screen 7) of self-test is meant to troubleshoot color RAM and uses every location in color RAM, and it looks just fine.
I do not have one of those magnifiers, but am increasingly feeling the need.
Looking over the sparkle and RGB circuits, I noticed that Atari seems to have missed the fact that BD0-BD3 make an appearance on sheet 9A. That's how MH (and Tempest) can load colors into the color RAM and then rapidly change the colors of existing vector by changing what's in color RAM. It's write-only since the RAM used is an 82S25 which has separate input and output lines, and the outputs go only onward to the final RGB output stages.
Note: Tempest and Quantum also have color RAM in the RGB output section, but only a single chip. MH has two color RAM chips in parallel with the enable line run through an inverter at 9J so when one is selected the other is off and vice-versa. Anyone know why?
There does seem to be COLA0-COLA3 going into both of the RAM chips, as well as activity on the ingoing BD0-BD3 data lines and the outgoing D1-D3 lines. When not in sparkle mode, all colors look good. According to TM-252 the Color Intensity Test Display (screen 7) of self-test is meant to troubleshoot color RAM and uses every location in color RAM, and it looks just fine.

