From: woodcock@crchh75b.bnr.ca (Gregg Woodcock) Newsgroups: rec.games.video.arcade.collecting Subject: Re: Typical Failures of Early Williams Games Date: 22 Feb 1994 23:29:15 GMT richard.l.schieve (rls@cbnewsc.cb.att.com) wrote: > Several people have asked me to repost some of my old articles. > I suspect many have seen the following which was one of my first > "Tech Tips" ;-) Sorry to do this to you again, Rick, but I have added a BUNCH of stuff to your file and would like to "share", too (I can never start these files but *BOY* can I update them :). Here is my revised text. ========= Well, I've been collecting, fixing, and restoring mostly early 80s video games for around 8 years now and have gotten pretty good at working on them. Some of these games have the same problems over and over and I thought some of you might like to hear about some of these problems. Much of this is information I've sent directly to people when they have posted requests for help. I'm not an electronics wizard or anything. I've just spent enough time trying to fix games that I've learned some typical failures. Todays Tech Tip is on the early Williams games, Defender, Joust, Robotron, Stargate, Sinistar, Bubbles, and I think Blaster but I've never worked on one. NOTE: Defender comes in 2 flavors, the standard Williams package (discussed herein) and a more compact 2 board version done by Taito (not discussed). Defender has several other important differences, too (ROM boards are different). Sinistar uses different ROM and interface PCBs than the other games but the rest of the boards are the same. Some versions of Stargate contain a CPU board which will only work in Stargate and none of the other games. One quick errata before we begin in regards to the Williams manuals: Some of them (Stargate and Joust for sure) drawings are incomplete. Your interboard wiring drawing probably shows video coming out of a 7 pin connector (IJ3 "Monitor Control") as follows: N/C 7 H Sync 6 V Sync 5 GND 4 Blue 3 Green 2 Red 1 If so, it's wrong. Composite sync is available at the unused pin (number 7 cabled "N/C"). It is indicated on the schematic pages but not on the wiring harness diagram. The problem with this sync is that it is an active high composite sync so you have to run it through a TTL inverter as MAMMA is looking for active low composite sync. These games all share the same basic logic board arrangement as follows: +-----------------+ +----------+ +----------+ +-----+ | | | | | | | S A | | | | | | | | P U | | | | ROM/PIA | | SOUND | | E X | | CPU/VIDEO/RAM | | | | | | E S | | | | | | | | C N | | | | * | | | | H D | | | +----------+ +----------+ +-----+ | | | X X X | +--------+ | X R X | |INTER- | | X A X | | FACE | <--INPUT | X M X | +--------+ | X X X | +-----------------+ Here is a blowup of the "X" area of the CPU/VIDEO/RAM board: BANK BANK BANK 3 2 1 : C 1 3K,31 2K,21 1K,11| | H 2 3L,32 2L,22 1L,12| | I 3 3M,33 2M,23 1M,13| | P 4 3N,34 2N,24 1N,14| | 5 3O,35 2O,25 1O,15| | N 6 3P,36 2P,26 1P,16| | U 7 3Q,37 2Q,27 1Q,17| | M 8 3R,38 2R,28 1R,18| ..-------------------+ If your game is not working, the first thing you should do is open up the cabinet and locate the 7 segment (alpha-numeric) "digital" LED on the ROM board (denoted by the "*"). Watch this when the game first turns on. Normally, it will show "0" and remain that way. If there is a detected failure, the display will repeatedly cycle 1 or 3 numbers. The first number is the failure type; 2=ROM, 1=RAM, 3=CMOS RAM. For ROM errors, the last 2 digits are the ROM number (ROMs should have stickers which are labeled to tell you which ones they are). For RAM errors, the last 2 digits are the Bank number and then the chip number (see previous picture). For CMOS RAM, there are no other digits. SPECIAL NOTE: Some (all?) Defender ROM boards do not have the digital LED but instead have a row of 4 regular LEDs. I do not have a Defender manual so I am not sure about how to read these lights other than the fact that they only indicate the bank numbers for RAM errors; there is no chip number information. If you have a ROM error, you can just burn a new ROM for it (try reseating it first). Most (if not all) Williams games use the same ROM board with different ROM data for different games. These ROM boards can be configured for either 2532s or 2732s. There are 2 pairs of jumper locations which determine which type of ROMs go in the board; they are W1/W2 and W3/W4. These jumpers are just to the left of ROM 5 (W1 to the far left, W2 to the right of W1) and to the left of ROM 6 (W3 to the far left, W4 to the right of W3). Jumpers W1 and W3 configure the board for 2532s while jumpers W2 and W4 configure the board for 2732s. It goes without saying that these are mutually exclusive pairings. SPECIAL NOTE: Some (all?) Defender ROM boards differ slightly from the other Williams games (they even have slightly different connectors) but since I don't have one, I am not sure about jumpers, etc. By far the most common failure of these machines is the DRAM (type 1 failure). This usually will produce a very organized "rug pattern" on the screen and the game will not "boot". Sometimes you get really weird video or else none at all. The solution is simple, get new DRAMs! These are type 4116 (450 ns access) 16K x 1 Dynamic RAMs and the cheapest places I have found are B.G. Micro at 214.271.5546 for 40 cents each and Altex in San Antonio, TX at 800.531.5369 for $1 each. This is an obsolete part so prices are bound to vary *wildly* due to local supply/demand. I am sure some of you can find them for pennies. If you want to pay through the nose, you can buy NTE2117 parts ($15+ near me). The three boards on the left are often mounted on a metal plate. The sound board is the same one used in pins and the AUX SND (speech) board is only used in Sinistar (I'm not sure on Blaster). The DIP switch at DS1 is supposed to enable/disable any auxiliary boards (like the AUX SND) and should not effect any game that does not have one. Here are some quick notes (from page 14 of the Sinistar Drawing Set 16-3004-103) about the jumpers on the sound board and how they configure different games: JUMPERS USED ROM TYPE FORMAT GAMES USED IN W2,5,7,9,10,15 2K x 8 Sound & Gorgar, Blackout Firepower, Black 2516/2716 Speech Knight, Jungle Lord, Pharaoh W1,2,5,7,9,10,15 2K x 8 Sound Defender (video & pin), Solar Fire 2516/2716 only Barracora, Hyperball, Stargate, Cosmic Gunfight, Varkon, Time Fantasy W3,4,5,7,10,15 4K x 8 Sound & Sinistar (upright & cockpit front) 2532 Speech W1,3,4,5,7,10,15 4K x 8 Sound Robotron, Joust (video & pin), 2532 only Bubbles, Sinistar (cockpit rear), W1,2,4,5,7,10,15 2K x 8 Sound Warlock 2516/2716 only W1,3,6,7,9,11,12 512 x 8 Sound Big Strike 15 7641 only The next most common problem I've seen with these games are loose connections that can become intermittent and sometimes overheat. How the loose connections affect on the game depends on where it is. All these boards including the power supply boards not shown are interconnected with female Molex connectors mating to male pins that are mounted perpendicularly to the boards. These male pins are fairly heavy gauge and are a bit hard to solder to. The solder connection between the board and pin often cracks when taking the connectors on and off. Also, at the female end, the connection can overheat if things get a bit corroded actually burning the connector shell and the female pin looses it's spring tension. People often complain of resets during game play. There are some software bugs (at least in Robotron) but more often than not the resets are due to the +5 VDC supply dropping below what TTL devices can operate with at the logic board itself. The Molex connections to and from the power supply board will often show signs of overheating. Try measuring the +5 at the power supply and again at the logic boards. If there is much of a difference and especially if the logic board is seeing less than +4.7 or so you have problems. If the female end has overheated considerably the female pin itself has lost spring tension and should be replaced. Often you can clean the male pins and work the female off and on a few times to get by for a while but chances are the female pin should be removed and replaced. Many hack artist coin-op repair types will simply cut off the female connector and solder wires right to the male pins. You see the same problems with Williams pinball games. If the physical connections are OK, try varying voltages from 4 up to 6 volts. I have seen machines that refused to work properly except in the range of 4.5-4.7 and others in the range of 5.5-5.7 so you will just need to experiment to see what voltage will keep your boards running without resetting. BTW, the software Robotron bug is supposed to go away if you turn off the "Fancy Attract Mode" which does away with the explosion when you zap a bad robot (thanks George). Loose connections at the logic boards seem more often to be cracked solder joints where the pins go through the boards. Any time unexpected things start happening while playing, it is worth the time to pull the board and check out the solder joints. I checked out about 6 board sets the other day and half didn't work but were fine after I resoldered the pins to the boards. Also, make sure to clean excess flux after resoldering the pins. I've seen too much left over flux scramble the inputs from the control panel. To clean excess flux I use 1,1,1-Trichloroethane called Carbo-Chor at Ace. It's amazing stuff as as strongly as it smells and as well as it works it does not burn so it relatively safe to use. I use one of the cheapie stiff bristled "glue" brushes and dip it in some trichlor and scrub away. One last thing, some versions of the CPU board which were only released inside Stargate games are incompatible with other games. If an unknown CPU board won't work in your Williams game, try it in a Stargate to be sure it is defective. Does anybody know how to tell just by looking if the board is of this type? I have seen 5 different labels on these boards; , Rev , Rev B, Rev C, and Rev D. If it works in Stargate, you can modify it to work in the other games. Here is a document from the October 1982 Star*Tech Journal that describes the procedure: ====================================================================== ========= Williams "Stargate" and "Robotron" PCB compatibility CPU/VIDEO BOARD CPU/Video Boards from "Stargate" may be used in "Robotron" by making two cuts and adding two jumpers. Modified boards and "Robotron" CPU/Video Boards may be used in "Stargate". Use a 27 watt or smaller (preferably grounded) soldering pencil and 60/40 solder for all soldering operations. Foils may be conveniently cut with a few strokes of a pocket or exacto knife. Use #22 or small gauge wire for the jumpers. See illustrations below [illustrations omitted for obvious reasons; the text steps are my descriptions of the pictures]: 1. On the chip at location 5A (under C63), cut the foil (trace) that connects pin 3 to ground. 2. On the chip at location 6F (under C75), cut the foil (trace) that connects pin 3 to ground. 3. Jumper the newly isolated pin 3 of both chips together. 4. Jumper pin 4 of the chip at location 5A to the feed thru hole (unused via) leading to pin 32 of connector IJ4 (above R26). NEW POWER SUPPLY BOARD Early "Stargate" games had a one piece D-8359 Power Supply Board Assembly. In "Robotron" and later "Stargate" games, this assembly has been superseded by the new D-8474 Power Board and outboard heat sink assembly C-8807. The new board features crowbar overvoltage protection on the +12vdc and +5vdc regulated outputs. NEW TRANSFORMER On upright and cabaret model games, a new power transformer with 8 primary and 12 secondary terminals (part no. 5610-09658) has also been substituted for the old transformer with 8 primary and 9 secondary terminals (part no. 5610-09353). The new transformer includes a 12.6v center-tapped secondary. Secondary terminal 20 and center-tap terminal 19 provide 6.3vac general illumination power to pins 14 and 15 of plug 4P1. Cocktail table model games continue to use a third power transformer with 6 primary and 9 secondary terminals (part no. 5610-09629) UPRIGHT GAMES New Transformer - new Power Supply: W1 in, W2 & W3 out. Old Transformer - new Power Supply: W2 & W3 in, W1 out and 2 ohm 5W resistor added at R27. Pins 13, 14, & 15 of 4J1 are left disconnected when transformer secondary plug is connected. New Transformer - old Power Supply: Pins 13, 14, and 15 of 4P1 (transformer secondary connector) are unterminated at Power Supply 4J1 12-pin connector. COMPACT (COCKTAIL) GAMES New Transformer - new Power Supply: W1 in, W2 & W3 out. Old Transformer - new Power Supply: W1, W2, & W3 in. Pins 13, 14, & 15 of 4J1 are left disconnected when transformer secondary plug is connected. New Transformer - old Power Supply: Pins 13, 14, and 15 of 4P1 (transformer secondary connector) are unterminated at Power Supply 4J1 12-pin connector. ====================================================================== ========= To people that have worked on these (and older Williams pinballs) this is all old news. I think that many that read this group have not had their hands in a machine until they are forced to and often have software backgrounds and are a bit afraid of the hardware. Rick Schieve rls@ihlpb.att.com additions by Gregg Woodcock woodcock@bnr.ca From dholcomb@microsoft.com Tue Dec 13 19:26:26 1994 Received: from netmail2.microsoft.com by halcyon.com with SMTP id AA00691 (5.65c/IDA-1.4.4 for ); Tue, 13 Dec 1994 19:26:19 -0800 Received: by netmail2.microsoft.com (5.65/25-eef) id AA09553; Tue, 13 Dec 94 19:26:31 -0800 Message-Id: <9412140326.AA09553@netmail2.microsoft.com> Received: by netmail2 using fxenixd 1.0 Tue, 13 Dec 94 19:26:31 PST X-Msmail-Message-Id: F299A4B7 X-Msmail-Conversation-Id: F299A4B7 From: David Holcomb To: holcomb@halcyon.com Date: Tue, 13 Dec 94 19:22:21 -0400 Subject: FW: (fwd) Re: Typical Failures of Early Williams Games Status: OR Newsgroups: rec.games.video.arcade.collecting From: rls@cbnewsc.cb.att.com (richard.l.schieve) Subject: Re: Typical Failures of Early Williams Games Date: Wed, 23 Feb 1994 15:15:06 GMT Keywords: Another old posting of mine. In article , astrachan@austlcm.sps.mot.com ('paul astrachan/xvt3') writes: > > I have been thinking about converting a Stargate to a Sinistar. > The following are two posts with a little conflicting info. > Can we sort this out? > > From RLS > >Todays Tech Tip ;-) is on the early William's games, Defender, > >Joust, Robotron, Stargate, Sinistar, and Bubbles. These game all > >share the same basic logic board arrangement as follows: > > > >------------------- ------------ ------------ ------- > >| | | | | | | | > >| | | | | | | AUX | > >| CPU/VIDEO | | ROM/PIA | | SOUND | | SND | > >| | | | | | | | > >| | | | | | | | > >| | ------------ ------------ ------- > >| | > >| | ---------- > >| | | | > >| | | | <--INPUT > >| | ---------- > >| | > >------------------- > > > >The three boards on the left are often mounted on a metal plate. > >The sound board is the same one used in pins and the AUX SND board > >is only used in Sinistar. > > > From Gregg W. > >Sinistar uses a different ROM board and interface board than the rest > >of the games do and requires an extra board that none of the others > >had for speech (and the sit-down game has a second sound board). It > >also uses a special zillion (48?) directional joystick for input. > > Both posts indicate the CPU/Video is the same. Is the ROM/PIA board > different from the Joust/Robotron ROM board? I said: "These game all share the same basic logic board arrangement" ^^^^^^^^^^^ not the same logic boards. Sinistar does not use the same CPU/Video or interface boards as Robotron/Joust. > Is the Interface board different? > > Is the Speech board the same as the aux board that was used on Williams > Pinball games like Alien Poker (I have an AP speech board) Yes. > What was the point of the second sound board on the sit down version. > Stereo effects? Yes, for Stereo. Kind of a reverberation effect. Rick Schieve rls@intgp1.att.com From dholcomb@microsoft.com Tue Dec 13 19:29:07 1994 Received: from netmail2.microsoft.com by halcyon.com with SMTP id AA01136 (5.65c/IDA-1.4.4 for ); Tue, 13 Dec 1994 19:28:25 -0800 Received: by netmail2.microsoft.com (5.65/25-eef) id AA09596; Tue, 13 Dec 94 19:27:53 -0800 Message-Id: <9412140327.AA09596@netmail2.microsoft.com> Received: by netmail2 using fxenixd 1.0 Tue, 13 Dec 94 19:27:53 PST X-Msmail-Message-Id: 66B48A1E X-Msmail-Conversation-Id: 66B48A1E From: David Holcomb To: holcomb@halcyon.com Date: Tue, 13 Dec 94 19:23:07 -0400 Subject: FW: (fwd) Re: Typical Failures of Early Williams Games Status: OR From: woodcock@crchh75b.bnr.ca (Gregg Woodcock) Newsgroups: rec.games.video.arcade.collecting Subject: Re: Typical Failures of Early Williams Games Date: 23 Feb 1994 04:16:35 GMT 'paul astrachan/xvt3' (astrachan@austlcm.sps.mot.com) wrote: > I have been thinking about converting a Stargate to a Sinistar. > The following are two posts with a little conflicting info. > Can we sort this out? > From RLS > >Todays Tech Tip ;-) is on the early William's games, Defender, > >Joust, Robotron, Stargate, Sinistar, and Bubbles. These game all > >share the same basic logic board arrangement as follows: > > > >------------------- ------------ ------------ ------- > >| | | | | | | | > >| | | | | | | AUX | > >| CPU/VIDEO | | ROM/PIA | | SOUND | | SND | > >| | | | | | | | > >| | | | | | | | > >| | ------------ ------------ ------- > >| | > >| | ---------- > >| | | | > >| | | | <--INPUT > >| | ---------- > >| | > >------------------- > > > >The three boards on the left are often mounted on a metal plate. > >The sound board is the same one used in pins and the AUX SND board > >is only used in Sinistar. > From Gregg W. > >Sinistar uses a different ROM board and interface board than the rest > >of the games do and requires an extra board that none of the others > >had for speech (and the sit-down game has a second sound board). It > >also uses a special zillion (48?) directional joystick for input. I am speaking from the perspective of the sit-down version which I have right next to me and I can assure you that all sit-downs I have ever seen have had the different ROM and interface boards. I *suppose* that the stand-up version be different (the same as the other games) but I very much doubt it. I am pretty sure that the reason the interface board is different is because it had to accomodate *48* different input signals which is highly unusual. This is why I said it would be better to start with a Sinistar and go backwards because you MIGHT be able to put the other games on the Sinistar ROM/interface boards (but I haven't looked into it). I KNOW that both versions of Sinistar use the same joystick but as I never paid attention to the stand-ups that passed through my hands, I cannot say FOR SURE that the ROM and interface boards are the same as the sit-down. > Both posts indicate the CPU/Video is the same. Is the ROM/PIA board > different from the Joust/Robotron ROM board? Yes. > Is the Interface board different? Yes. > Is the Speech board the same as the aux board that was used on Williams > Pinball games like Alien Poker (I have an AP speech board) Probably but can't say for sure (I bet Rick can). > What was the point of the second sound board on the sit down version. > Stereo effects? YOU GOT IT BABY!!!!!YBAB TI TOG UOY From dholcomb@microsoft.com Tue Dec 13 19:29:16 1994 Received: from netmail2.microsoft.com by halcyon.com with SMTP id AA01210 (5.65c/IDA-1.4.4 for ); Tue, 13 Dec 1994 19:28:44 -0800 Received: by netmail2.microsoft.com (5.65/25-eef) id AA09624; Tue, 13 Dec 94 19:28:02 -0800 Message-Id: <9412140328.AA09624@netmail2.microsoft.com> Received: by netmail2 using fxenixd 1.0 Tue, 13 Dec 94 19:28:02 PST X-Msmail-Message-Id: 411161E5 X-Msmail-Conversation-Id: 411161E5 From: David Holcomb To: holcomb@halcyon.com Date: Tue, 13 Dec 94 19:23:50 -0400 Subject: FW: (fwd) Re: Typical Failures of Early Williams Games Status: OR From: woodcock@crchh75b.bnr.ca (Gregg Woodcock) Newsgroups: rec.games.video.arcade.collecting Subject: Re: Typical Failures of Early Williams Games Date: 24 Feb 1994 02:19:13 GMT richard.l.schieve (rls@cbnewsc.cb.att.com) wrote: > Sinistar does not use the same CPU/Video or interface boards as > Robotron/Joust. In my experience (3 upright, 2 sit-down), this is false. CPU/video boards have always been compatible for me but your mileage may vary. I concur about the interface boards, though.