Tech: Help needed with Donkey Kong Cocktail

[Larry]

Alright.

I've posted about this unit a couple of times before, and I'm at wit's
end, so I hope that someone can help me.

This is a 4 board Donkey Kong cocktail that was working perfectly up
until four weeks ago. A neighbor kid was playing, and when they were
done, the machine wasn't working anymore.

What I get when plugging her in, is
http://www.behavioralert.com/art/dk1.jpg
for about a quarter second, then
http://www.behavioralert.com/art/dk2.jpg
for about a full second, then
http://www.behavioralert.com/art/dk3.jpg

for the remainder.

I do get the jump sound when I first plug it in, as well.

Based on the second and third screens, I thought it was a bad 5E (CPU
5F on my set) after comparing it to the screens on Brasington's site.

I ordered a set of the program ROMS, and replaced them.

Same thing - I get the exact same results. Therefore, I conclude it's
not the EPROMS.

I removed, recleaned, and reseated every socketed chip on all boards.
No luck - same screens.

I have looked at all boards, and can't see any broken traces, and a
moderate spot check on several pins all over the board shows pulsing
and oscillation on my oscilloscope.

I am frustrated, and could really use the help of anyone who can.

I don't have deep expertise in fixing boards, just the easy stuff.
So if anyone can help, I will greatly appreciate it!

Thanks!

Larry Ziegler

[Scott Caldwell]

Larry, did you check the cables to make sure they were
firmly seated? Remember, the 4 boardset has more cables
than a typical 2-board. Plus, there is a power board
that goes to that inside PCB (I noticed it on one of
mine I'm trying to sell).

Scott C.

[Larry]

> > Larry Ziegler- Hide quoted text -
>
> - Show quoted text -

Hi Scott,

Yep, this is with all of the 475 connectors connected (who thought
that was a good idea?), and I have disconnected and reconnected the
IDE-like cables jumping between the 3 big boards, and all the
connectors appear to be firmly seated on the cables from the sound
board to the CPU board.

Your questiont does bring up a further question - which f the
5,875,325 connectors being disconnected might cause these symptoms?
Perhaps I have a problem further up the chain?

Again, this unit was literally working one minute, then with no other
changes except letting the neighbor kid play (I was not watching to
see or hear the point of it not working) the machine was in this state
when I got back downstairs.

Thanks again in advance for help!

Larry

[Scott Caldwell]

Larry, I'm battling a similar issue on a 2-board set. From
what I can tell, it's a failure in the video section (yeh,
I know that's a big duh!). However, there are some common
chips that can cause this sort of failure. I would suggest
getting a spare video PCB, if you can, just to rule it out.
Also, checkout both Brasington's and MikeArcade pages for
some common chip failures. It's worth a shot!

I'm able to work on other vendor PCBs, but Nintendo PCBs,
just are not my cup of tea. :(

Scott C.

[Larry]

> Larry- Hide quoted text -

>
> - Show quoted text -

Updated Information:

I have tried swapping out power supplies with a known good PS from a
donor machine, and I the machine boots up with the same problem.

I have removed the Sound board completely - it doesn't change a thing
(well, no sound, of course) - same non-boot problem.

I have tried lightly flexing the boards with power applied, and I see
no changes.

Any ideas?

Larry

[Scott Caldwell]

If you have a logic probe, you could check the video
RAM (on a 2 board Video half, it's 2 P/R and 6 P/R,
but I have no idea on a 4-board set). I think Scott's
site does list those location, though.

Scott C.

[Larry]

I don't have a logic probe (nor have I ever used one), but I do have
an O-scope. Can I use that to test, and if so, what should I be
looking for?

Larry

[Scott Caldwell]

Larry, I don't have those layouts in front of me, but the
they are 2114 and 2148 RAM on the 2-board set. You can
check the states of the outputs and see if they are stuck,
pulsing, etc. The few bad RAMs I've found this way had
either very weird outputs or none at all (the chip died).
You can also compare the two chips (the 2114 to each
other and the 2148 to each other) and see if you get any
differences.

I haven't ever used a scope, so I can't help there. You
may consider picking-up a probe, since they are usually
pretty cheap used, but working.

[Scott Caldwell]

Someone had a nice web page set-up that gives a quick (and good)
explanation of how to use a log probe. It's really pretty easy
once you get used to it and know what you're looking for.

The following is a schematic posting on the vector group today.
The red dots mean that chip leg is currently showing a "high"
state and the green dots mean it's showing "low". There can
also be pulsing (my probe has that as yellow), which usually
means the pin is showing some kind of activity (generally a
good sign). If you understand logic states (several good web
sites that teach this) and you know the layout of the chip
you're looking at, then it is fairly easy to tell if the chip
is working or not. IE. If it receives a signal on pin x and
another signal on pin x, then it should output a signal (or
some other action) on pin x. If it doesn't, then the chip may
be bad. That's a very, very brief explanation, but you get
the idea.

http://games.rossiters.com/Cine-Clock-Circuit.jpg

If you have another working set of that same game, you can
always swap in the working board and see what a good chip
measures. There are some chips on my DK 2-board set that I
don't have the chip details for, so I've been comparing two
video halves to see if the various chip states agree. If
they do, then chances are good that chip is not the problem.
It takes more time, but it is also an easy way to build-up a
list of states for various chips you don't already know or
have documentation on.

[Larry]

Thanks, Scott!

That's very helpful. I think I can use my scope for the same purpose
- though I admit it's a bit overkill, and probably harder than using a
probe. Now that I think about it, I'll probably hunt a probe down - I
have a Moon Cresta that needs some TLC, and I'm sure I'd get further
use out of a LP beyond that.

I guess this'll be a longer process than I first thought, but I'll
definitely learn a lot about my DK in the as I go. (I just want to
play it so bad!)

Wouldn't it be nice if Radio Snack carried a decent array of
components, so I could pick up a couple of spare 2148s on the way
home?

BTW, what would you recommend as a good reference for the pin
descriptions for these older chips?

Thanks again!

Larry

[atari...@yahoo.com]

Larry-

For what it's worth, I had a 4-PCB board set with a similar problem,
and it ended up being a broken trace between the program ROM chips.

-Steve

[Larry]

Thanks, Steve,

I'll check that out too. Hopefully it's something easy and I don't
spend the next month tracking this down :)

Larry