Tempest weirdness

[erdna77]

I have a tempest machine that boots up with 40 credits and unresponsive controls. First happened after a dip switch adjustment.

Tried swapping pcb with known good one. No change.

Tried AR2 rebuild and Big Blue swap. No change.

Ideas?

 

[brzezicki]

you swapped the CPU boards with a known working one and you still get those issues?

 

[erdna77]

Yes. I've never seen anything like it before.

 

[scottmaggie]

tempest

1. If you tried a completely different board set and got the same problem then the dip switches had nothing to do with it. If that is the case you most likely have a wiring issue, probably grounding something somewhere.

2. Disconnect the coin door harness (black molex) and power up. Does the issue still happen? If so you can rule out the coin door. I am guessing that when you disconnect the coin door the problem goes away though...If it does clear up the problem you need to check all the wiring, switches, etc. on the coin door.

3. If disconnecting the coin door doesn't resolve the issue, check the wiring by the volume control and test switch.

Please post updates so we can help you further. Good luck.

 

[Cricket117]

Hey guys, so I am doing work for erdna77 on his tempest machine and i took a video of the occurring problem.

https://www.youtube.com/watch?v=YbHgtP2qg4Y

as you can see when the test switch is turned off the credits quickly climb to 40 and are not simply stuck. And when turned back on 2 of the IO reading are jumping up and down ( i believe 2 and 3 for the difficulty dip switches) and we are not able to change the difficulty from medium no matter the configuration but all other dip switches work fine.

Also, no controls respond but the IO reads the coins and slam switches fine.

I removed the connector from the secondary board as well as pulled the pins for coin R, coin L, Coin C, Coin ctr R, coin ctr L, coin ctr C, and self test as per this pinout diagram.

http://www.mikesarcade.com/cgi-bin/spies.cgi?action=url&type=pinout&page=Tempest.txt

even running it with out all of these connections the problem still exists.

I also tried unplugging the molex connectors for the controls and coin doors at the front as suggested by scottmaggie.


were really stuck here

 

[scottmaggie]

Super helpful video

Ok so first thing I see is that none or the buttons or the spinner are registering on the self test. The next thing is toward the end of the video you see the red instructions as the top of the screen bouncing around.

Again. I'd lean toward something shorted but if you have eliminated the wiring as a possibility you have to look up what chip handles the controls. I don't have it in front of me but pretty sure it's on the aux (smaller) board.

 

[Dokert]

Put the game into test, and see what Math Box error comes up. Your bad component is on the Aux board not the CPU.
Dok

 

[erdna77]

Put the game into test, and see what Math Box error comes up. Your bad component is on the Aux board not the CPU.
Dok

Weird that the problem persisted even when swapped with a known good PCB set that I purchased from komodo.

 

[scottmaggie]

forgot that

I forgot that when I replied before. Well there is only so much wiring. It's got to be a bad ground or crossed wire somewhere. Got to just go piece by piece.

You removed the coin door and controls as culprits by disconnecting the molex
What about the wiring around the transformer (and under it)?
All that's left is the edge connectors on the board
Heck even disconnect the light & speaker

Just go slow, piece by piece.

 

[barristr]

Swap the pokeys around on the auxiliary board to see if the problem changes. The pokeys handle the controls, and can produce some interesting problems when bad.

 

[ieure]

A lot of folks in here seem to be confused. Cricket117 unplugged the aux PCB to eliminate that as the source of the issues, which is why self-test is showing the EQP M codes. It's not a mathbox problem. The coin inputs do not come from the POKEY anyhow.

erdna77 / Cricket117, when the credits rack up, does the coin counter click up? What happens if you trip one of the coin switches? If you stick a logic probe on the COIN L / C / R test loops, does it pulse?

What are you getting on the +5v test point on the AVG PCB?

When you swapped the PCB, did you swap in a full set (AVG+AUX), or only one of the two boards?

Have you reset the DIPs per the troubleshooting guide? "Set all of the DIP switches on the main board on except for N13 #7 and all of the switches on K10/11 (turn them off). Turn all of the DIP switches on the aux board off."

 

[Cricket117]

To answer ieure's questions:

-the counter does not click when the credits rack up but it does click when the coin switches are pressed. And the IO test recognizes the coin switches and slam but nothing else.

-I believe I was getting 5V but I will need to double check. I know I have 5V at the watch dog.

-first we swapped both boards with no luck and i swapped the aux board and just the pokeys from a third board with still no luck

-yes I have tried setting the switches as per the trouble shooting guide with again no luck.

Thanks for your time guys.

 

[ieure]

To answer ieure's questions:

-the counter does not click when the credits rack up but it does click when the coin switches are pressed. And the IO test recognizes the coin switches and slam but nothing else.

What I think this means is that whatever is causing credits to rack up is either corrupting RAM or hosing the address/data bus. I have no idea how the PCB would pass self-test in this condition.

The fact that the coin counter can click up but doesn't tells you that the problem has nothing to do with the IO side of things.

-I believe I was getting 5V but I will need to double check. I know I have 5V at the watch dog.

Please check and report back.

-first we swapped both boards with no luck and i swapped the aux board and just the pokeys from a third board with still no luck

Just to confirm, when you say "with no luck," you mean that it exhibited the same symptoms each time?

-yes I have tried setting the switches as per the trouble shooting guide with again no luck.

What if you set it back to 1 coin / 1 credit?

 

[Cricket117]

yes, by no luck I mean they both had the EXACT same symptoms.

And how would the ram on BOTH boards be corrupt in the exact same way if that were the case ?

 

[scottmaggie]

question

You tried a second known good set in the cab and it caused the same issue. Did you try the original boards in a known good cab to see if they work 100%? And also try the known good boards back in a known good cab to make sure you didn't cause an issue with them now?

Once you can confirm that at the very least the known good boards are still good (and hopefully both boardsets) then you will 100% know you have a cab issue.

 

[Cricket117]

We don't have another cab the boards will fit in at the arcade but I will check with erdna77 to see if he was one around his place.

 

[Cricket117]

Haven't been able to get down and test anther cab and I am not even sure if we have one. But since we tried 2 different main boards and 3 aux boards, I would assume we could rule out the boards because they all gave the exact same issues.

so in that case, where would you guys start to troubleshoot ?

Remember, the AR was already recapped and tried a new AR board, and big blue was replaced.

thanks