Data/Address Bus help.

smarty

Ready Player One
vacBacker
Feedback
12 (100%)
Credits
1,305CR
Hi Guys,

I need a bit of help with a Fire Truck Board I'm trying to fix up...

The game display random graphics & text when powered up, from probing around the board I have found a couple of IC's that have had dead outputs, but nothing that has fixed the board.

I think I've found the probable issue but can't work out what is causing the faulty. The CU has good clock,& Reset and has plenty of activity on the Address bus and Data. From the address bus A11,12 & 13 feed into an LS42 decoder, on the output side of the decoder I have ROM0, ROM1 & PRAM which aren't pulsing away. These signals drive the ROM selects(ROM0,1 which are program ROMs) and some LS245's an 157's that in turn feed into RAM & ROM.(PRAM) I have taken the LS42 out of circuit and it tested fine, from what I can tell these output signal are not shorted to other signals. On my working board I can remove the ROMS and the ROM0/1 pulse away from what I can see on the schematics these don't go to any other place than the ROMs.

So my question is what would stop the outputs from triggering? Input D to the ls42 is always low even on my working board, these output signals can only be being driven from A11,12,13 which come directly from the CPU. On the Data bus side there is a LS245 transceiver, could this have an impact on the address bus data if it was not working correctly. From my logic probe everything seems to be pulsing away where expected, but I do understand using a scope may now be the way forward to actually 'look' at the signals to try and find something amiss.

Image below of the CPU section of the schematic which covers what I've talked about:

FT_CPU_Diagram.jpg


Thanks for any input
smiley20.gif
 

Mitchell Gant

Active member
vacBacker
Feedback
2 (100%)
Credits
888CR
smarty said:
So my question is what would stop the outputs from triggering? Input D to the ls42 is always low even on my working board, these output signals can only be being driven from A11,12,13 which come directly from the CPU.

Software is what could stop the outputs of the decoder triggering. Lets say that ROM3 has the startup code in it that is executed on powerup. It then sits in a loop because it has some sort of fault elsewhere, maybe the ROM is faulty, maybe the data bus is faulty, any one of many reasons. That software loop is simply going round and round only accessing ROM3. So none of the other ROMs are ever selected as their addresses are never put on the address bus. So many of the pins of the 'LS42 seem to be stuck high, even though nothing at all is wrong with that chip.

It looks to me as if Firetruck only uses the address space from 0 to 0x3FFF, and then has mirror images of this range up to 0xFFFF. So ROM3 would be where the software starts on reset, and looks like it never gets any further.

This is where a logic probe can't really help you I'm afraid.
 

smarty

Ready Player One
vacBacker
Feedback
12 (100%)
Credits
1,305CR
Cheers Neil,

Software eh! Thanks for the pointer i was really stuck in the hardware mindset, I was assuming (most likely incorrectly!) that ROM 0 would have been the star of the program code so my thinking was that the game code never started running correctly to begin with, but like you say If ROM3 is where the code begins I need to start from there. I pretty sure I swapped the complete romset with a known working set and found no change. That give me something else to work with.

I'll be taking a look at this over the weekend and will update if I get some success.

Cheers, Mart.
 

cmonkey

Active member
vacBacker
Feedback
4 (100%)
Credits
1,659CR
Reset vector for Fire Truck is 0x3e22, which is in ROM3.

Reset vector is found at address 0x3ffe (mirrored to 0xfffe, which is where the CPU looks to boot from).
 

smarty

Ready Player One
vacBacker
Feedback
12 (100%)
Credits
1,305CR
I spent some time on this today... First up I made a test rig so I could run the board from my dining room table instead of working out the back of the cab.

I confirmed on my working board that the CPU will run correctly with only ROM3 & ROM2 installed so I removed all ROMS from the faulty board except the two mentioned, and after rechecking my previous work I found that the Pin 9 of the LS367 at A1 was stuck low. I swapped the device over and now the outputs of the LS42 decoder at H2 are all pulsing away happily. But... the board is now watchdogging. I only had a 74LS367APC to replace the faulty IC which was a 74LS267AN. I'm sure before swapping the IC, the Watchdog was being reset correctly.

I'm just about to check the spec of both and order some more parts. If the APC has nothing to do with this new fault there's an LS02 which feeds into a LS42 decoder that controls the watchdog reset signal and the output of this now seems stuck low.

Thanks for the help with this one.
smiley20.gif
 
Top