NEO-MVH MV4 Boot problem related to A15

ack

Ninja Combat Warrior
15 Year Member
Joined
Apr 9, 2009
Posts
538
I would go ahead and swap the 245 on the lower pcb just so you can rule it out.

I dont think slot 4 is passing the check since you are getting the crosshatch instead of it loading the game. The graphics used to render the memory viewer comes from the S1 rom on the active slot. If the board isnt able to get at the S1 rom for whatever reason is common you will get a mostly blank screen, since the char data is all 0xff.

A. That would be true if you disconnected the input pin on the 245 coming from the slot board. If you disconnect the pin on the cpu side, it will normally take on the last state the data line was in (from being used to read other parts of memory)

B. Cant you just do the probing with it powered normally?
 

mikeDB

n00b
Joined
Jul 10, 2021
Posts
17
Yes, thanks for bringing me back to my senses. I keep thinking the slot 4 is working because compared to the other slots it has some semblance of life. But, of course the mere fact that it doesn't actually start the game from the cartridge means it clearly isn't fully functioning.

I'll try replacing that other 245 on the CPU PCB.

B. If I power it up just the slot PCB without the CPU PCB, then I can exercise full control - I can pulse inputs and check resulting outputs of individual ICs without any interference from the traffic generated by the CPU PCB.
 

mikeDB

n00b
Joined
Jul 10, 2021
Posts
17
ack,

Finally the '61' and slots 1-3 issues have been fixed!!
One remedy fixed both problems.

First, let me confirm that isolating select legs of the 245 on the top PCB does result in a high value for those associated bits as indicated by the results observed in the Memory Map. At least that was the case for the 245 at P4 which was indeed the lower byte of the word. For instance isolating pin 5 changed memory map results as follows:
@100 6145 612D changed to 6155 613D.

Doing the same for the 245 at P3 did not result in any change.
But now that I confirmed that the 245 at P# was the high byte, I knew there was no need to replace the 245 @B7 on the lower PCB which would have been associated with the lower byte; and some time ago I had already changed the 245 @ B8 that is related to the higher byte.

So then I isolated and grounded a number of the 245 input pins @ B8 which by rights should have changed the '61' value to '00'. But, that still didn't change the '61'.

At that point I traced the CE pin of both the B7&B8 245's which navigated as follows:
B8-245,Pin19 wired to C7-74AS08,Pin6
B7-245,Pin19 wired to C7-74AS08,Pin3
The 7408 is a quad And gate.
Pin6 output is dependent on input at Pin4 & Pin5.
Pin3 output is dependent on input at Pin1 & Pin2.
But Pin1 & Pin4 are wired together, so I traced Pin5 and Pin2.

C7-7408,Pin5 input is wired to C11-244,Pin7 output, whose input is Pin13.
C7-7408,Pin2 input is wired to C11-244,Pin9 output, whose input is Pin11.

C11-244, Pin13 is wired to PRO-C0, Pin11 (ROM0EL).
C11-244, Pin09 should be wired to PRO-C0, Pin3 (ROM0EU) - But there was no continuity between the two. Following the traces through some vias, I found an open trace. Installing a wire between the two vias on either side of the open trace restored that connection.

That one fix resolved both the '61' issue and the slot 1 thru 3 issue.
Now all 4 slots work and no crosshatch is encountered.

Thanks so much for all your invaluable help,
Mike
 
  • Like
Reactions: ack
Top