Asteroids Cocktail reset fun

Blades

Member
Joined
Aug 4, 2008
Messages
960
Reaction score
5
Location
Nebraska
Just a question.

My cab has been working fine, but recently when I turn it on, it keeps reseting the board. As-in I turn on the power and at first, my board will reset over and over(with beep), then be fine for a while, then it will reset again and again. At that time it would just reset the PCB but Now recently it's been reseting the WHOLE cab, as-in the coin mechs making there clicking sound over and over(like it's lose power to it and regaining it), the monitor display spot killer turns off and on(like when it's not getting a good signal from the PCB).

I've done a cap kit a couple months ago, replaced big blue about the same time. I've looked at the power supply and don't see any obvious problem. It's getting the voltages out that it's suppose to. The AR board is also putting out it's required voltage(just did a recap kit on it, not too long ago).

Do you think that the AR board is causing the problem, does the coin mechs work off of that board, which would explain why it's clicking off and on(losing it's 12v) when it resets.

Not sure where to go from here, PCB(cap kit done) seems to be ok, I just wonder if that AR is causing havoc or if the ISO is shitting out on me. What do you think?
I have gone over it and can't see any loose connections

Now keep in mind, I can't MAKE it reset, as in, if I wiggle wires or jiggle the cabinet, push buttons etc.., it will not reset. It just does it on its own. Sometimes constantly and sometimes once or twice every half hour.


;
 
Mine was doing the exact same thing and turned out to be a board problem. I tried cleaning all the chips which didn't help. I never figured it out as I replaced it with a spare working board.
 
Mine was doing the exact same thing and turned out to be a board problem. I tried cleaning all the chips which didn't help. I never figured it out as I replaced it with a spare working board.


Cool at least one response. Now did you replace AR board and the PCB board or just one or the other?
 
Cool at least one response. Now did you replace AR board and the PCB board or just one or the other?

I just replaced the main pcb. AR was fine as the main pcb was getting all good voltages.
 
CPU logic controls the coin counter operation. Check to be sure you are getting VERY close to +5.00 volts DC at chip level. This is not at the 5V test points but on the actual chips.

1) Try measuring voltage on pin 8 of the 6502 CPU at location C3.

2) Try measuring voltage on pin 40 of the 6502 CPU at location C3.

Pin 40 of the 6502 is the /RESET line and should be HIGH (+5v) during normal operation. If you see this line dropping to zero or something significantly less than 5v then the pcb is resetting itself for some reason.

Debugging starts to get a bit more complex from this point forward.
 
Last edited:
Ok, I've been able to play with this a little more.

BTW the clicking I was hearing was the coin counter counting up every time it reset itself. I have unplugged it to get it out of the picture and the board still resets.

I have checked pin 40(reset pin) on the 6502 and it reads 4.17 volts. On pin 8 i get 5.28

On the PCB test plugs I get 5.32

On the AR board I get 5.45.

(And obviously the lower I dial down the 5v the more often it resets itself. So right now the AR board 5v is on as high as the adjuster will allow me to go, which is the 5.45 volts.......is it supposed to go higher?) Obviously I don't want to go too high....anyway..

Now one thing I did notice in the Asteroids Circuitry manual. was that it stated that the MPU has to be a 6502A and will not function properly on just a 6502 MPU. Here is what it says via manual:

Note: The MPU in this game operates at a frequency of 1.5mhz. Therefore the MPU chip must be 6502A. The 6502 Maxium frequency is 1 Mhz and is not compatible with this game.


Can anyone confirm this? Maybe this might be part of my reset problems.



When the board resets (which is at random), the voltage on the AR nor PCB (test points) drop at all, ONLY the voltage on Pin 40 drop.

Still, as XRAY2 says, pin 40 should be at 5+ volts(normal operation) and it's only 4.17 volts.

Anything else that anyone can think of?

Also can anyone confirm the 6502A chip spec in there own asteroids machine to see if this is required. Maybe I just have a krappie PCB.

Thanks!!
 
Tie the Watchdog Disable test point near D5 to ground. This will eliminate the reset if it was triggered by the CPU failing to write to WDCLEAR to address 0x3400 before the timer expired. Let's see how this test turns out and we'll take it from there.

You are probably seeing 4.1 volts on the /RESET line because it is running at less than 100% duty cycle. The line is being reset (grounded) multiple times during your voltage sampling with the meter. In essence, you are seeing the average of 5v around 80% and being 0v around 20% over the given sample period. This voltage will vary somewhat depending on the quality (speed) of your meter but it will not show a steady 5v. If your meter has logic probe functionality, this is the perfect time and place to utilize it to watch the /RESET line go from HIGH to LOW and back.

As for the CPU the MOS series 6502A were rated at 2MHz while the 6502 was just 1MHz. Many manufactures built licensed 6502 series processors including Rockwell and Synertek to name a few. Most had their own nomenclature for the part number but you do typically see the "A" on the end.

Here's a link to a Synertek datasheet.
http://www.gamearchive.com/General/Data_Sheets/6502.pdf

After reviewing the results of the WD Reset test, feel free to swap in a known good 6502A CPU for testing.
 
Last edited:
Ok I'll start with that tonight, when I get home from work. Thank you for the help so far!!!
 
Depending on your results, I may have a custom Asteroids diagnostic rom image I scan send you. PM me your email address when you get home.

-- Steve
 
Ok, I did as you suggested by tieing the watchdog leg to gnd. And when it would reset, it would just emit a long tone(beep), screen blank and would continue that tone until I pulled off the jumper. At that point it would reset itself and go about its business until it would reset at random times....

Whats also interesting(the more I watch this) is that when it resets, it resets violently, as in multiple resets at a time, very fast, then it would be ok for a minute or 5, and then reset once or twice, would go for a while more and then reset a sh*tload of times in a row. (No voltage to the board is dropped at anytime.) And continue this pattern.

The constant resetting doesn't happen instantly when you turn the unit on, after about 5 min. or so does it start.

Not sure where to go from here, Thank you for the offer of the Rom image, however I do not have a burner in my possession. (Should probably look into that though eh?)

I'm thinking this board is to blame. Not sure where to go, maybe a new 6502. Put an "A" version in it, maybe that would help?

What do you think? Thank you for you help so far!



;
 
Here is the MPU specs that are on the chip as Xray2 requested.....

MPS 6502----Top row
1278----Bottom row

Now you mentioned in your PM that the 1.5mhz freq being a bad thing, but my question is, in the clock spec portion of this atari PDF, it states that a normal 6502 will not work and must be a 6502a with the clock spec of 1.5mhz. It's a little paragraph on the Clock Circuit page. Is this wrong?

http://www.joesarcade.com/schematics/asteroids/Asteroids-sp.pdf

Thank you again for your continued help!!!




;
 
Yes, if you have a 1.0 mhz rated CPU and are attempting to run it at 1.5 mhz in the Asteroids circuit then unexpected behavior will become the norm. Grounding the watchdog disable test point should allow the game to run if /RESET was the issue. The fact you get erratic results points back to the CPU. (for starters)

There is no point to continue trouble shooting with an incorrect component in the circuit. As soon as you have a 6502A installed we will no longer be chasing ghosts.
 
Sounds good to me. I will start with that and replace the MPU.

Thanks again for your help and the links!!!


;
 
By the way...

MPS 6502----Top row
1278----Bottom row


MPS is probably MOS and the 6502 is just a 6502 (1 MHz) not a 6502A (2 MHz) so you've been lucky until now running on your over clocked CPU!

1278 is the date code

Week # 12
Year = 1978 -- An EARLY manufacture date considering the long run this processor had.

_
 
Closure post for future searches...

I received a PM from Blades over the weekend. He ordered a 6502A replacement CPU and installed it in his Asteroids. This eliminated all of the issues he was experiencing. Unless we hear back, installing the proper speed rated CPU did the trick!


Just wanted to let you know that I got the replacement chip with the proper faster MPU and so far so good. No random resets, no beeps, it's just working like it should. I've left it on for about 6 hours today and I will not reset like it used to. Sweet!
 
Back
Top Bottom