Asteroids & WG 19V2000 Repair Log

Picked up a (non-working) Asteroids a few weeks ago. Managed to fix it using information found on the 'net, including the archives of this mailing list; so I thought I'd join the list and share my repair notes.

The sideart is in nice shape; coin door is one of the older "round entry" types and needs repainting and polishing of the chrome; plexi in nice shape (and has the older yellow text); cardboard surround is usable with a little cleaning (monitor schmutz); so work has focused on the PCB & monitor. Starting point: powering up gave me repeated high pitched beeps and spot-killer LED on the montitor. (Note that this is my first vector game, so I lack a known-good vector PCB or vector monitor.)

Asteroids PCB: S/N 1847 (matching S/N on AR board).
-preliminary checks verified the power supply block was putting out all the proper voltages, as was the AR board. +5VDC was present and well-adjusted on the game PCB.
-logic probe on pin 40 of the 6502 confirmed is was being repeatedly reset, presumably by the watchdog circuit. Grounding the WD Reset line led to a hang.
-as I lack a way to verify the ROMs and I was planning on eventually using the Asteroids High Score Save kit, I decided to go ahead and get it. This kit has custom game code so it effectively replaces all three of the game ROMs (C1, D/E1 & F1). In addition, it comes with an -02 vector ROM (N/P3). As a result, this kit replaced all of my ROMs, and eliminated them as suspects. However, the game was still watchdogging...
-I tried the DMAGO trick (socketing the 74LS42 @ L6 and lifting pin 1). This stopped the resetting and allowed my to play the game blind. It also cleared the microprocessor side of the game, and directed my attention to the vector generator side.
-For no justifiable reason, I decided to replace the bipolar PROM @ C8 (the only wonky IC in the circuit). Ataricade.com programmed & verified it for me. However, this didn't fix it for me.
-Test mode (with DMAGO lifted) gave a single high beep, so I assumed my RAM was OK. I broke out the schematics, logic probe & datasheets to begin a logic probe marathon...
-Note that I had to reconnect DAMGO so it would try starting the vector generator and exercize the circuit. I eventually found that on the 74LS109 @ A9, pin 6 was stuck high. It is supposed to be a complimentary output to pin 7, which means it should always be the opposite state. Pin 7 was changing state, but not pin 6.
-Replacing the 74LS109 @ A9 fixed the board, and allowed it to run w/o resetting with DMAGO connected. I also verified signals on the X & Y video output lines.
-Playing blind isn't really any more fun with DMAGO connected than disconnected, but now I had a video signal to try fixing the monitor...

WG 19V2000: This had been hacked into the cabinet, presumably due to the different connector configuration from the OEM monitor (G05-801?). It has light Battlezone burn (interestingly, the burn-in is mirrored left-to-right...). Now that a signal was comming from the PCB, the spot killer was now off, but still no picture.
-There was no neck glow. Tracing the 6.3VAC found that it was making it onto the deflection board, but not off of it. I pulled the deflection board and took a good look at it; lots of cracks at the header pins. This solder was filthy and difficult to melt. I eventually got it off and got all of the header pins re-soldered. Hooked it back up; finally... video!
-There was a picture, but it bloomed quickly and badly. The 90VDC line (return from the HV board to the defl. board) started at about 60VDC when it was turned on and dropped to about 35V in a couple minutes. After reading this is a symptom of a bad HV diode, I began looking for one. Bob Roberts was out of them, but I ordered a cap kit & the 6800uf filter caps from him anyway. I found the HV diode in stock at arcadechips.com (Mark Capps).
-The cap kit arrived first, so I replaced all the electrolytics. Of course this didn't fix the blooming but I feel better about having all new caps in it.
-The HV diode (H1809) arrived. It was a pain to remove the old one. There was greasy goop all over the diode, boot, anode cap & wire, and the diode didn't want to come out. The heat-gun-on-the-rubber-boots trick eventually worked well. The springs inside the boots were in bad shape (rusty). I salvaged one by sanding off the rust. The other spring broke all apart during extraction. I cleaned off the brass cup, stripped back the HV wire, soldered it back on, and soldered in a piece of spring I found in my spring collection. The old diode had the leads cut off short (one diode radius long) and folded straight down. I didn't know the best way to prep the new ones, but I ended up coiling the full lead into a spiral on the end of the diode. I reassembled with dielectric grease (like I use on spark plug boots) to try prevent future rusting of the springs. The blooming was now fixed.
-Somewhere along the way, the picture starting flickering and partially collapsing on the X-axis. Eventually traced this to a bad X linearity pot. BTW, I can't figure out why this is called a "linearity" adjustment; from what I can tell it does nothing but adjust the size (which can be done on the game PCB also). I haven't found a replacement yet, but there are only a few places along the adjustment that are noisy, so I adjusted to find a steady point then adjusted the size on the PCB.
-The picture was now solid, but shifted up from center. Some fiddling with the yoke adjust rings/tabs finally centered it up. Perfect image now!

Not too bad for a $50 non-working game! I think I'm ready for a Tempest now...

df

      
---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Sun Sep 20 14:44:04 2009