I'm starting to hate Asteroids ...

[CheffoJeffo]

My travails with Asteroids continue ... 

After fixing the vertical collapse that happened within the first hour of my bringing this game home from auction, then replacing some bad RAM (and not doing a pretty job of it), the other day I found my Asteroids was suffering from two vertical strips where the images were a mirror image of what they should have been (kinda like two rectangular black holes each 2 inches wide).

So, I reviewed the Asteroids Repair Encyclopedia and the repair logs at GLS. Then I pulled the board and use Randy Fromm's quick and dirty method of checking the voltage drops across pins on the IC.

I finally concluded that the 74LS273 at K7 was bad. OK, I can fix that.

Took a quick drive to A1 Electronics and picked up a handful of 74LS273 chips and sockets. Fired up my shiny new soldering station (WOW! What a difference from using those crappy pencil irons!!!) and replaced K7. Nice clean job. I'm very proud of it.

Checked my work and made sure there were no broken traces or loose solder bits. Put the board in and fired the cab up.

I was faced with a totally mangled display. 

Flipped the test switch and got 'bom, bom, BOM, BOM'.

Frak! I just replaced those RAM chips. I swapped in two known good chips and no change.

Checked K7 and still see weird voltage drops across the IC (weird being different from the other 74LS273s). Hmm. Check the resistors near K7. R34 is only testing at 3.3K and should be 22K. Check the bands and it;s 22K. Pull up one leg and test and it's 22K. Damn. Maybe Randy's method doesn't always work and I need to learn something about what the chip should actually be doing.

OK, time to check the voltages. Start at the ARII board. 5V is about 5.3V. Check at the game board and it's 5V.

So, I decide to check the voltage at the RAM chips in question. 5V. It's 5V everywhere ... until I get to the 6502, where it is 4V.

Pull the board and gently pry up the 6502. It has a broken leg. Solder a jumper to the leg and fire up the board again. Now getting 2.5V. Doublecheck the jumper. It's fine,

Crap ... I have a box full of Z80s, but no 6502s. Check A1's website and they don't stock 6502s.

Have to order a replacement and wait while I rapidly lose hold of the belief that I finally knew what I was doing and was doing it right with the right tools.

 

I'm starting to hate Asteroids.

 

[RayB]

Pull one from a Commdore 64. 

[Level42]

Pull one from a Commdore 64. 

That uses a 6510 not a 6502 at it is NOT pin compatible.



Keep it up Cheffo, you'll tackle it sooner or later and Asteroids is one of the best of the best.

6502's available here:
http://www.arcadecomponents.com/cpusupport.html

or here:

http://www.coinopchips.com

[ChadTower]

Is that the same 6502 as in the NES?

[CheffoJeffo]

I ordered a couple of 6502As and a pair of 74LS491s (used in Tron spinner boards) from ArcadeComponents.

Then I realized that I do have another 6502 from my Atari Football project.

Swapped that in and get no change  but at least I'm not waiting anymore. 

EDIT: Was killing time waiting for my family to arrive for the weekend, so grabbed a magnifying glass and meter and started looking the board over in detail. Found a broken lead on a disc cap that needs replacing (not enough lead exposed to jump it). The question is was at which point was it broken ... or did I just break it with my fat fingers during my inspection.

[RayB]

The question is was at which point was it broken ... or did I just break it with my fat fingers during my inspection.[/i]

I broke one last week inspecting my cocktail. So I blame you!

[CheffoJeffo]

Finally got back to the board again (the visiting family members have all gone back home) and replaced that radial cap ... it was my fat fingers (so maybe I am to blame for Ray's as well!) ... no change in behaviour.

I sat down and took a more ordered approach to things and, after checking the traces around the RAM chips for the eleventieth time, I had a look at what exactly was happening. In test mode, I get 2 regular beeps, then 2 low tones, which indicates RAM failures at M4 and R4 (vector RAM). I swapped the 4 vector RAM chips around, swapped good chips in, and got the same test results every time.

I decided to try to look at things logically instead of just trying to shotgun them. At this point, I assume that the RAM is good and the voltages are OK.

I looked for traces that touch M4 and R4, but not N4 and P4 (which aren't flagged as bad in the RAM test). There is only one -- the chip select on pin 8. Tracing pin 8 on R4 leads to the C7 cap, which, coincidentally is glass (?) and has a crack in it. I don't have a cap tester and while taking resistance measurements indicates something is happening, I don't trust it.

Sadly, I have a very early morning and I don't want to fire up the machine tonight while everybody else is asleep, so replacing C7 will have to wait until tommorrow morning's coffee break.

If that doesn't work, I may have to learn how to use that nifty logic probe I picked up today.