OJ’s projects…usually breaking stuff.

[olivier.jan]

Actually there’s a video signal… I scoped the RGB at R115, R116 and R118 and I can clearly see the signal there. Using diag cart menu I can see variations of the signals depending on the screen background colour.
But these signals seems way to high, with a black value of 0.84v and a white close to 2v. Need to check the video circuit.

[olivier.jan]

After seeing this video signal, I installed a Video DAC upgrade from Exxos, and again black screen. The signal is obviously much better in term of range and voltage, but still black screen. So, video circuit isn't responsible for the issue, which is a good thing as the Video DAC upgrade doesn't fit really well under the keyboard.
Will check H-Sync, V-Sync and Blank to check if it is properly in sync tomorrow.

[olivier.jan]

Took me a while to do the tests, but I can now confirm I have H-Sync and V-Sync working fine.
I did some comparisons with a working 520STM and I have similar Sync signals and similar Mono signal (using mono makes it simpler to scope).
Information gathered so far:
- Machine displays a full black screen (borders and picture are black).
- Produces sync signals with the proper frequency when using Mono or RGB.
- Video circuit is not causing the issue as using Exxos video DAC gives a similar black screen
- Shifter is OK (tested in a different machine).
- CPU is working OK as the Diag Cart (through serial) works fine.
- RAM Tests with the Diagnostic Cart gives no error.


I will try different GLUE and MMU tomorrow

[exxos]

Blanking signal transistor ?

[olivier.jan]

Actually there is no transistor for /BLANK, at least on this revision. But you’re right, /BLANK is a suspect, I will check it. I was thinking of checking the 74LS244 and 74LS373 which are involved in getting memory access for the shifter too.

[olivier.jan]

As planned, a few tests today:
- tested /BLANK, HSYNC and VSYNC, the three are OK on video circuit and on GLUE. (Compared with a working STM).
- I swapped GLUE with the STM, no change.
- I swapped MMU with the STM, machine doesn’t even boot.
- Installed MMU and GLUE from dead machine in the working STM: works OK.
- Swapped MMU with a working STF, works OK.
- Swapped shifter with a working STF, works OK.

I checked the 74LS373 AND 74LS244, nothing suspicious there neither.

I would like to remove the memory extension (piggyback) as I suspect this might be causing the display issue, even though Diag Cart doesn’t report any memory problem. Is there a way to disable it without removing all the extra memory modules?

[exxos]

Likely just lift RAS/CAS signal piggyback RAM and tie to 5V.

[olivier.jan]

So, today I tried disabling the piggyback RAM, but it doesn’t change a thing : still full black screen. ( actually it changes the amount of RAM obviously)

I’m trying to understand what could cause such symptoms : full black screen (even the borders) but with proper sync signals.

[exxos]

Sync is controlled by GLUE...

Possible some ram chip blown ? Disabling ras cas can't disable a blown chip.

[olivier.jan]

Yes, I guess next step will be to remove RAM chips and replace them with known good chips.
Not sure I will have time today though.
Anyway this is going to be a long repair.