Repair & Maintenance Log
12/18/08: Acquired game. Replaced power cord.
12/19/08: Problem: Stuck shooter lane switch. Solution: Cut out leaky switch cap. Problem: Background sound randomly stops during heavy ball action. Solution: This was another leaky/intermittent switch cap on the tube exit switch (Note: The exit switch stops the background sound before the kickout saucer resets the background sound). Problem: Game serves both balls to the shooter lane. The outhole kicker was unreliably kicking the second ball over the trough hump. If the ball rolled back onto the outhole switch, the game became confused. Solution: Cleaned kicker assembly and adjusted outhole switch. The marvin3m.com repair manuals also suggest that this problem could be caused by corrupt RAM data (see below). Problem: Inoperative center and right coin switch. Solution: Rewire miswired left coin switch.
12/27/08: Replaced missing knocker assembly. Why would someone scrap the knocker out of a perfectly good game?
Some previous person had removed the NiCad battery from MPU before it could leak and destroy the board. But no battery replacement circuit was ever implemented to maintain the volatile RAM. Not only are some game settings lost when the power is turned off and back on, but the game itself could act squirrelly from semi-corrupted RAM data (I guess). Instead of implementing a new battery backup circuit, I decided to replace the old RAM chip (which can be unreliable anyway) with a modern non-volatile memory chip from Tom Callahan at pin-logic.com.
Shown above is the new memory chip (red arrow). The chip comes with a small circuit board such that the wider new chip can piggyback on the old narrower socket. Assuming the old socket is healthy, this is a plug and play modification. I folded down the TP6 wire hoop so as to not interfere with the new larger chip (yellow arrow). Blue circles indicate the old battery connections.
04/26/09: Solenoid driver board repair. Connector J3 was broke in half and two of the J3 pins were bypassed with alligator jumpers. Time to try my hand a connector repair. None of the connectors or pins looked great so I pulled the board and replaced all the J1-J5 header pins and connectors.
I also replaced Q19, the coin lockout driver.
While I was at it, I also made the ground modifications as recommended at pinrepair.com. Note: pinrepair.com recommends NOT connecting the +43 volt ground (J3 pins 23-24) to the +5 volt ground (J3 pins 18-22). However, my board has a trace connecting all of J3 pins 18-24.
Old, busted up J3 (left) and new connectors (right).
New header pins and new Q19.
Recommended pinrepair.com board mods: TP1 jumped to TP3 (red arrow), negative lead of C23 jumped to ground trace (blue circle) and negative lead of C26 jumped to ground trace (yellow arrow).
05/10/09: Coin door repair. I don't set my games to free play. Coining the game is part of the fun and I keep a big dish of quarters nearby. But Xenon takes dollars as well. Unfortunately, none of this worked because the coin lockout driver Q19 was bad (see board repair above). Also the dollar mech was missing. The dollar slot had been disabled by wrapping the coin drop window with electric tape. After removing the decades-old tape I had a nice S.B.A drop window. I added a surplus dollar mech, adjusted the switch and I was good to go. The coin door is still a cruddy mess, but now everything works.
S.B.A. coin drop window (5 plays for a dollar).
04/03/10: New playfield glass.
03/24/12: Installed standard-keyed (751) lock on coin door.
01/11/16: Ball trough modification. I've had this game for more than seven years. The game has sat idle for many of those years thanks to my frustration over a persistent double ball serve problem. Scroll toward the bottom of this page for more details. I've written about this problem extensively. I still have no solution so it's time for a work around. My goal is to make the outhole kicker absolutely reliable.
I went to eBay and found a spare lower ball trough guide. These parts are plentiful. It's the same part you'd find in most any Bally game through the 1970s. I pulled the lower guide out of my game and cut them both as shown below. I reassembled the parts with screws and spacers such that the beginning of the guide now sits within the playfield slot about ¼" higher from its original position. This does two things. It reduces the leading slope of the guide. And it places the ball more directly in front of the outhole kicker arm. The outhole switch wireform required minor adjustment to account for the new ball position. Hopefully the kicker will now propel the ball over the trough hump such that the ball never rolls back onto the outhole switch. **So far so good**
Ball guide cuts.
Reassembled parts.
Modified guide back in the game.
The ball sits more directly in front of the outhole kicker arm.
01/16: Shopped playfield. Everything topside was disassembled, cleaned and polished. New rings and new balls.
Looks like there was some preliminary intention to have two more lamps near the top of the playfield.
Missing GI sockets?
The metal post by the spinner was bent. I looked through my scrap box and found a beefier post with a slightly wider base
New beefier post by the spinner.
The game flyer shows a metal post by the tube shot. My game had a plastic post. The original hole was blown out and there was some playfield wear. So there was no going back to a metal post. I filled the hole with a two-part epoxy. Then I re-drilled the hole for a tight fit and added a new blue post.
Blown out post hole.
Hole filled with two-part epoxy.
New post.
The tube is supported by a pair of giant white nylon cable clamps. My left clamp was by the far left side of the tube whereas the game flyer shows the left clamp more toward the middle. My game has an empty T-nutted hole corresponding to the configuration shown by the flyer. I reassembled my tube with the clamp back on the far left before I noticed this discrepancy. My "misplaced" clamp could have been from a previous tech guy who wasn't paying attention when he reassembled the tube. On the other hand I have found pictures of other Xenons that are set up like mine. IPDB, for example, shows pictures of several different games where the mount is positioned both ways. Looking through various pictures on the Internet seems to show about a 50/50 split. So maybe the clamp position is a legitimate factory variation.
My tube clamp at the far left side of the tube.
Game flyer showing the left tube clamp more toward the middle.
Empty T-nutted hole for alternate (correct?) tube clamp mounting.
There was some black electric tape along the path to the Exit Chamber. Maybe someone was trying to prevent further playfield wear. I cleaned and waxed the area and applied a Mylar patch. If someone wants to restore this playfield someday the wax should make the Mylar easier to get up.
Mylar patch covering the playfield's only significant spot of wear.
The manual shows a 2" ring below the exit chamber in a rectangular configuration. I found a playfield dimple as if there could have been a fourth post. Either way a 2" ring was too big. I used a 1 ¼" ring.
1 ¼" ring below the exit chamber in a triangular configuration.
The thumper bumpers were in decent shape except the old white skirts were beat. I disassembled and cleaned the mechanisms. I decided to try new black skirts. Bumper sockets are my nemesis. I can clean them, adjust them or replace them and I still end up with one or two that give me unreliable contact. Time for an experiment... I dispensed with the sockets all together. I decided to take advantage of the LED's long life expectancy and hardwire them right into the game. I picked frosted blue LEDs from CoinTaker.
Leads soldered directly to the LEDs.
No more bumper sockets.
I like how the black bumper skirts disappear into the playfield.
I had a set of new drop targets I installed. They were not quite identical to the originals.
Old and new drop target.
There are three white nylon posts behind the drop targets. Each post had two additional homemade mounting holes. Based on playfield fade it would appear that the posts spent most of their lives in the wrong place. The original holes were not stripped out so I don't know why someone would want to move the posts around. I moved all the posts back to their original positions. I did notice that it's possible for the ball to deflect the ring slightly into the lamp sleeves. But it doesn't seem like a severe problem that would warrant moving the posts forward. We'll see...
White posts back where they belong.
Drop Target Update: After a few test games it became obvious what was going on with the three white posts. With the posts back in their original positions a ball could get stuck along the side of a lower drop target after an upper target has been knocked down. It took a significant amount of nudging to free the ball. First I replaced the white posts with short blue finned posts. This pushed the ring out a little further. And the blue posts mostly concealed the ugly homemade holes which were all uneven and haphazardly drilled. One ring still wasn't enough so I doubled it up. Ball hang-ups are still possible, but rare and are easily cleared with a gentle nudge.
Ball hangup.
Three short blue posts and double rings.
I've had a bunch of late 70s Bally games, but Xenon was my first encounter with the linear-style slingshot kickers and flippers. The linear kickers have a bigger footprint and I had to de-solder the slingshot switch leads to get the mechanism out for cleaning. I didn't really like the bare switch leads so when I re-soldered the leads I added scraps of heat shrink tube.
Linear slingshot kicker.
The linear flipper mechs were well worn so I rebuilt them entirely. I replaced the right coil since the old sleeve was melted in place. Half way through my rebuild I realized I didn't have new flipper bats. I reused the old bats, but the shafts were too beat up to get them adjusted the way I wanted. A lot of people don't like the "feel" of these linear flippers and convert back to the more traditional style. I already had the linear rebuild kit so that's what I used. I'm sure I'm not a good enough player to tell the difference.
Rebuilt right flipper mech.
The game came with red flipper rings. The flyer shows yellow. The ring kit I bought came with blue. So I tried blue.
Blue flipper rings.
My game had the infamous faded to pink/clear playfield posts. It's hard to believe that these things started off as blue. I bought new blue posts from Pinball Resource. There's some debate on the as-new shade of the original posts. Some say pale blue. Some say deep blue. In any event, deep blue looks better than what I had so I'm happy. In total I used 33 tall posts plus three short posts for the drop target modification noted above. I replaced all the posts except the Kmiec red post which belongs by the upper left ball gate.
Old and new posts.
The Kmiec red post.
I also bought an array of colored bulb sleeves. I wanted to replicate the original arrangement, but I haven't found any definitive information about what the original arrangement may have been. I guessed based on what I found on my game plus what I could make out from the game flyer. I had four red sleeves behind the drop targets. I couldn't confirm that from the game flyer. However, the four sleeves could be confirmed from that old Omni documentary about Xenon (go to YouTube and search "Suzanne Ciani Xenon"). There were four red sleeves across the top arch which looked correct based on the game flyer. There was a blue sleeve by the tube exit which looked correct based on the game flyer. There were no more sleeves on my game, but the flyer makes it look like there could have been two more reds by the exit chamber. So that's one blue and ten reds.
What's left of my original sleeves. One blue and eight red.
Four new red sleeves behind the drop targets.
Four new red sleeves across the top arch.
Two more reds and a blue along the exit chamber.
I had a new set of CPR reproduction plastics to go on the game. I've had this set for years and it's obviously one of CPR's earlier efforts. The edges were rough and there was no protective film on the art side. They just weren't as nice as what I've become used to. The CPR products I've bought more recently have been excellent. Hats off to CPR; they've come a long way. I like to apply a protective undercoating of Mylar to my new plastics to help protect the art from screw heads and other hardware. In this case I found a partial set of transparent plastics protectors on eBay. My old plastics had zero ball damage so I'm not sure the protectors are necessary in that regard. Nevertheless, I used the protectors in place of my usual Mylar treatment. I applied Mylar to only three plastics which were the two top arch plastics and the drop targets plastic.
Partial set of transparent plastics protectors (before removing the protective films).
New instruction and pricing cards.
Done!
02/16: LEDs, sockets and lamp drivers. The blue playfield inserts looked dim. The dingy old Mylar wasn't helping. This game has full Mylar. It's not factory original and it looks like it would pull up pretty easy. But I won't risk it. I'd rather have dingy Mylar than missing playfield art. So I thought I'd add blue LEDs under the blue inserts. Then I decided to go ahead and just do all the playfield inserts using color-matched super bright LEDs from CoinTaker. I stuck with incandescent bulbs for all other game lighting except for the thumper bumper lighting as noted above and the tube lighting.
The LEDs caused two problems. With LEDs some of the inserts exhibit a rapid flashing effect that wasn't there with incandescent bulbs. I wouldn't call it "ghosting" because the LEDs don't flicker when they should be OFF. When they're OFF, they're OFF. When they're ON some light solidly while others rapidly flash. I don't really like the effect. But I'm trying to convince myself that the flashing could have been a plausible design intent. Maybe the game programmer was trying to make the game look "flashier". Maybe not.
I thought the whole point of these newer pinball-specific LEDs was that they were designed to properly load the driver circuits. Guess not. They do work if properly loaded. If, for example, I jump a 1k ohm resistor across a flickering LED, it stops flickering. I'm not interested in replacing the lamp driver boards with the newer Alltek boards. I'm not interested in soldering a resistor across every lamp socket. Perhaps I'll try a set of those flicker eliminator adapter boards that piggyback on the lamp driver board connectors. That looks like a relatively simple and inexpensive solution that is easily reversible. Reversibility is a must. Conclusion: LEDs are a pain in the ass. Don't do it.
The second problem was that the bright LEDs made it glaringly obviously that all my lamp sockets weren't up to snuff. The first amber "N" between the flippers was flat out dead. This traced back to a bad SCR at Q51 on the lamp driver board. The 2N5060 appears to be obsolete so I used the 2N5064 which is the replacement that all the suppliers now sell.
New SCR (2N5064) at Q51.
Next I went through every insert socket and added a solder bridge between the lamp bracket and the socket body. I started by using a cutoff disk in a Dremal to grind away some metal at each point where I intended to solder. I tried a small wire wheel, but that didn't seem aggressive enough to get down to clean metal. But using a cutoff disk around wiring harnesses isn't a great idea. It takes a lot of care and patience. I worked on just a few sockets at a time and took a lot of breaks. Next I added a dab of paste flux at each solder point. This is critical. Just a tiny dab makes all the difference. After soldering all the sockets, but before turning ON the game again, I tested all the sockets for shorts. Actually I did this twice. The last thing I wanted to do is blow up more lamp drivers. I soldered every socket except for the drop target awards. This was for a lack of work space. Sometimes it's also helpful to remove the control wire from the solder tab and solder the wire directly to the back of the spring contact in the middle of the socket. Again, I use the cutoff disk and paste flux to prepare the spring contact for soldering.
Here are some solder bridges (red arrows) and a relocated control wire (green arrow).
I still had about four uncooperative sockets so I just replaced them. In the interest of future lamp access I added solder tabs between the mounting screws and lamp brackets. All the new sockets can be removed without de-soldering the power line. I also replaced the upper GI socket behind the left slingshot which was perpetually dim.
New lamp socket and solder tab.
Power line soldered to the solder tab, not the lamp socket bracket.
When I bought this game it had already been retrofitted with a new blue LED light board for the tube.
A close up of the tube's preexisting blue LED circuit board.
Done!
02/18/16: Replaced missing playfield switch capacitors. These capacitors tend to short which in turn shorts out the switch. Repair techs typically clip bad capacitors and leave it at that. I clipped a few myself years ago just to get the game working. But the capacitors are necessary to achieve proper switch sensitivity. Not all switches have a capacitor. I used the switch matrix schematic as a guide. But there are inconsistencies. For example, the four top rollover buttons have what appear to be original capacitors installed, but they're not shown on the schematic. All in all I replaced about four capacitors that were clipped or missing.
General Note: Like many Bally games from this era, my game is missing the transparent protective shield that goes over the rectifier board in the bottom of the cabinet. After doing any work on the bottom of the playfield (especially after soldering), I run a shop-vac over the rectifier board before turning the game back ON.
02/21/16: The cabinet tilt switches and cabinet slam tilt switches were all none functional. Neither of these two switch circuits are part of the switch matrix. Both problems were due to faulty connector pins which happened to be adjacent pins 15 and 16 at MPU connector J3. So I rebuilt the connector J3. I probably should have pulled the MPU board and replaced the header pins as well. I probably should have replaced all the header pins and all the connectors. I'll save all that for another day. Many people would not bother fixing tilts in a home use environment. I like aggravating aggressive guests.
New MPU J3 connector.
New SBA coin door decal.
04/22/16: New flipper bats.
New flipper bats.
03/16/17: New Alltek Ultimate MPU. One day I turned on the game and was suddenly getting an automatic extra ball on ball 1. The number of players was irrelevant. Everyone got an extra ball on ball 1. The extra ball was awarded as soon as any playfield switch was closed (even a 30 point switch). Balls 2 and 3 played normal. I was stumped. Someone suggested that the U8 data had been altered. That did not entirely make sense to me, but I checked it out. Self test positions 16 and 17 should be 03 and 03 for replay or 02 and 02 for extra ball. Mine were 00 and 02 - somewhere between extra ball and garbage. Note that I was still getting replays for playfield specials. So I changed the values back to 03 and 03. That fixed the game. But I still didn't know why the data had become corrupt in the first place.
A closer look showed that I now have some battery corrosion along the bottom edge of the MPU board. I suspect that the U8 socket has gone bad. When I bought this game back in 2008 it had no battery. It also appeared to have no battery corrosion. Instead of replacing the battery I replaced the U8 RAM with a new nonvolatile memory chip. But I didn't replace the U8 socket. Compare the picture at the top of the page with the one below. The board has noticeably deteriorated over the last eight years even with no battery.
Battery corrosion.
I didn't feel like messing with a battery corrosion repair project so I just ordered a new Alltek Ultimate MPU. In preparation for the new board I replaced all the MPU connectors J1, J2 and J4. Note that I had already replaced connector J3 last year.
The new Alltek board is half the size of the old (-133) MPU.
New MPU and connectors.
It wasn't until after I sold the old MPU that it occurred to me that the old MPU was the wrong board. The -133 board is a modified version of the standard -35. The -133 is for the pinball/video games Baby Pac-Man and Granny and the Gators (this particular board was from a Baby Pac-Man). It should not have mattered. In fact, except for improved reliability, there was no change in the game's behavior with the new Alltek board.
09/02/17: Scrapped linear flipper mechs. My rebuilt linear flippers were not completely reliable. It seemed to be a plunger alignment issue. Whereas conventional plungers have two points of alignment, linear plungers have three. If the three points don't line up, the plunger sticks. The flippers worked fine when cold. But after everything warmed up for a few hours, the right plunger would occasionally stick. I replaced the linear plunger and pawl with an old-style conventional plunger, link and pawl assembly. It's a straight forward swap. Some folks rig the pawl with a DIY bracket to retain the new-style extension spring. I just used an old fashioned compression spring.
Linear parts (top) swapped for conventional parts (bottom).
Half way done.
The linear plunger guide becomes the link stop.
09/07/17: Fixed Exit Chamber side saucer glitch. In a nutshell, the problem was not with the side saucer, but with the ball release rollover switch (switch 02) on the right side of the ball trough. If the ball release rollover switch (switch 02) is open, the game seems to assume that there are two (non-multiball) balls in play. Accordingly, the game holds a ball at the side saucer. I eventually solved the problem by cleaning and adjusting the ball release rollover switch (switch 02). This problem was so maddeningly intermittent and so long standing that I wrote a more elaborately detailed story which may be found at the bottom of this page.
Ball stuck in the Exit Chamber side saucer.
Ball release rollover switch (switch 02) from the top.
Ball release rollover switch (switch 02) from the bottom.
09/21/17: Success! After two weeks of testing, I'm calling this game bug-free. Only took nine years of on-again/off-again tinkering.
02/15/18: Replaced ball release rollover switch (switch 02). The switch still wasn't perfect so it was time to replace the damn thing. This is Bally part ASW-A1-211. I could find no such part at any of the usual suppliers so I bought an ASW-A1-155 and made it into an ASW-A1-211. I disassembled the switch and made two modifications.
1) I added an insulator immediately beneath the long blade of the switch (see part labels below). This isolates the switch from the rollover wireform and the rest of the metal ball trough components.
2) I flipped the column terminal so as to not interfere with the ball release mech (solenoid 3).
Switch parts.
New ball release rollover switch (switch 02).
A common problem with this title is that a fast moving ball might skip right over the Exit Chamber side saucer. There are posts leading to the Exit Chamber that are intended to rattle and thereby slow down a ball from the tube. Sometimes it's not enough. I added an extra 7/16 ring below the 5/16 ring on the post by the one-way gate. This creates a gap only a bit wider then the diameter of a ball. This arrangement rattles and slows the ball such that it lands in the side saucer every time.
Extra ring (red arrow).
02/22/19: LED flicker fix. This is a follow on to my 2016 LED project (see above). To solve my LED flicker problem I bought a Siegecraft Electronics adapter boards kit. There are different kits for different lamp driver boards. I didn't buy a kit for the aux lamp driver board because I never added LEDs to any of those circuits. I liked this solution because it was simple, relatively inexpensive and easily reversible. It all went together exactly as the instructions said it would and my LED flicker was eliminated.
Each of the three boards requires power from the controlled lamp power bus. The kit includes a gray power wire which is daisy chained between each board. The end of the gray wire came with an alligator clip for clipping to the controlled lamp power bus. This didn't seem like a very durable approach. So I cut off the alligator clip and soldered the gray wire directly to the controlled lamp power bus at the High Score to Date lamps on the insert panel.
I don't like the idea of having oddball modifications that might confuse a future owner. So I tagged the gray wire where it's soldered to the controlled lamp power bus. I also left a copy of the installation instructions in the bottom cabinet and another copy in the game manual. Adding LEDs to a game is like going down a rabbit hole. I could have bought about 700 incandescent bulbs for what I have into this project.
These boards work by placing an extra load resistance between the controlled lamp power bus and each lamp driver output. This is electrically the same as adding a resistor across each lamp socket. The resistors are nominally 715 ohms. I measured between 711 and 713 ohms.
Siegecraft Electronics adapter boards kit.
Boards installed (note gray power wire).
Power connection on the insert panel. Note the existing blue wire which originates from BR1 on the rectifier board.
03/08/19: Replaced rectifiers. While futzing around with the Siegecraft adapter boards (see above) I noticed my controlled lamp power bus was down to about 3.6 volts. Pathetic. I had 10 volts ac going to the rectifier BR1 and everything else looked good. So I figured it was time for a new BR1. This AS-2518-54 rectifier board appears to have been a substantial improvement over the previous AS-2518-18 board (see my Star Trek repair page). There's lots of Internet repair information for the -18 board and its too-small 8 amp rectifiers, but not so much for the -54 board and its larger 25 amp rectifiers.
I bought some KBPC3504W 35 amp rectifiers which were just the right size to replace the old 25 amp rectifiers. There appeared to be nothing wrong with BR2, but while I was at it I replaced both BR1 and BR2. I tried some Tarn-X on the two high current fuse clips F1 and F5, but there wasn't much effect. In any event, the clips were still shiny on the inside where they actually contact the fuse. I suck at checking fuses when I buy a game. All the old fuses were at least in the ballpark. But a few were off which I replaced accordingly.
After separating the rectifier board from the transformer chassis I cleaned away the old heat sink compound with some naphtha and vacuumed up the area. The old rectifier leads are folded over which is a pain. But after sucking away enough solder, the leads can be straightened and the components removed. Note that each rectifier has a spacer which is reused. It's important that both rectifiers be coplanar with the transformer chassis. This is another reason to replace BR1 and BR2 so they're both dimensionally identical. Before soldering the new rectifiers I temporarily reassembled the board to the chassis to make sure the rectifiers and chassis were perfectly coplanar. After soldering, I cleaned up my work, double checked everything, added some fresh heat sink compound and put it all back together. I initially checked all voltages without plugging in the output connectors J1-J4. All looked well!
Transformer chassis before cleaning up the old heat sink compound.
Old BR2 with spacer.
Old rectifier compared to new.
Temporarily reassembled and ready for soldering.
Done!
With a new BR1 my controlled lamp power bus jumped to 7.4 volts under load which brightened up the backbox nicely. I should have done this years ago and maybe avoided the whole LED fiasco.
07/23-8/23: Yet another MPU.
***Project in progress; this may not work out.***
Back in 2017 I replaced the game's existing MPU board with an Alltek. Then I was annoyed with myself for selling the old board instead of trying to fix it. In April 2023 I found a Xenon MPU on eBay that looked like it might be worth while. Unlike the old Baby Pac-Man board, this was an actual Xenon board with correct part and serial number decals. And the board appeared relatively unmolested. There was a remote battery pack which I cut away. There may have been some battery corrosion along the bottom edge. So I did some remedial sanding and vinegar scrubbing followed by rinses of water and alcohol.
I put the board in the game and of course it didn't work. I got the infamous locked-ON LED, which is about the most ambiguous thing the LED can do. The reset circuit appeared to work. I could simulate a reset pulse at the microprocessor by shorting the junction of R1 and R3 to ground. No LED flicker. I shorted the microprocessor pins 39 and 40. No LED flicker. I was reading 2 to 3 volts on all the pins receiving a clock signal, so I thought the clock circuit was working.
I removed all chips but U6, U9 and U11. No LED flicker. I swapped U11 with U10. No LED flicker. I went to reseat U6 and noticed it was the only chip with visible tarnish on the legs. So I carefully sanded the tarnish. No LED flicker. I reseated the microprocessor U9. No LED flicker. At this point I decided to buy an ATX power supply and move myself to the workbench.
Show below is my cheap ATX power supply. I immediately disassembled the unit, internally grounded the power ON circuit and hacked away most of the output wires and connectors. All that's left is one wire each for +5 volts, +12 volts and GND.
ATX power supply.
On the bench.
Having moved to the bench, things became even more ambiguous. On one or two occasions I may have gotten a flicker followed by a few, but not all blinks. More typically I'd get a flicker and no blinks. Any attempt at a manual reset would lock ON the LED. I tried a new microprocessor; no difference. I tried a new PIA at U11 which caused the LED to always lock ON. I did continuity checks between all chip legs and adjacent board traces. So I'm reasonable confident that the sockets are healthy.
Next I went to Matt's Basement Arcade and purchased a Leon Borre test EPROM for the Bally -35. Leon Borre's original instruction could still be found on pinwiki and other places around the Internet. This comes on a 2716-type EPROM (I assume) and is intended for a U6 socket jumpered for the 2716/32. Since my board is jumpered for the 9332/2532, I built the Leon adapter, which is just an intermedary socket with ground going to pin 18 and 5 volts jumpered to pin 21.
Leon Borre test EPROM.
EPROM installed atop the adapter.
I saw activity at A13, which Leon says should be low. So I replaced the 6800. Activity around the new 6800 looked promising so I installed a new 6821 at U11. The PIA randomly starts and stops. I'm still not willing to accept that this could be a socket issue. I retested for continuity between all chip legs and adjacent board traces. I spent a lot of time pressing and poking and tapping on the chips and around the sockets. None of my physical interactions corrolated to the PIA outputs randomly starting and stopping.
I ordered a Pinitech 5105 NVRAM adapter.
Pintech 5105 NVRAM adapter.
Notes: Double Ball Serve Problem
Like many other Xenons, my game suffered from a ball serve problem where both balls were served to the shooter lane. I've read a lot of forum threads on this topic. But I never found circumstances quite like mine and I never found a definitive solution. I did take the time to carefully document all my observations. Here they are...
In summary, my game malfunctions when the outhole switch (switch 08) is closed twice. This happens when the outhole kicker fails to propel the ball over the ball trough hump and the ball rolls back on the outhole switch closing it for a second time. My only previous solution has been to keep the outhole kicker working as flawlessly as possible. If I cleaned the plunger and coil sleeve the game worked pretty well, but only for a while. I don't consider that a fix. The game ought to be able to cycle through all the trough mechanisms and still keep track of where the balls are supposed to be. Once both balls are in the shooter lane the game becomes hopelessly confused and a new game must be started.
Under normal circumstances there are not two balls on the right side of the ball trough. When the game is at rest, one ball is held at the ball release kicker on the right. The other ball is held at the outhole kicker. At the beginning of a game, the ball release kicker serves a ball to the shooter lane as ball one. A brief moment later the outhole kicker passes the other ball to the ball release kicker where it is held until the game requires another ball. When ball one drains the ball is held in the outhole while the ball release kicker serves the other ball to the shooter lane as ball two. A brief moment later the outhole kicker passes the first ball to the ball release kicker where it is held until the game requires another ball. When the last ball drains that ball is held in the outhole until a new game is started and the process begins again. This appears to be how the game is supposed to work. But if the outhole kicker fails to get the ball over the trough hump, the ball rolls back and closes the outhole switch for a second time. Then both balls end up in the shooter lane.
This problem is not intermittent. It is completely consistent. Close the outhole switch twice and both balls go to the shooter lane. When the ball rolls back and closes the outhole switch for the second time, the outhole kicker fires a second time and then a brief moment later the ball release kicker fires. That's how the second ball gets into the shooter lane. In other words, the ball release kicker always fires a brief moment after the outhole kicker fires for the second time. This sequence of events occurs regardless of the status of the two trough switches. If I force one or both switches to be closed, the sequence is unchanged. If I force one or both switches to be open, the sequence is unchanged.
I'm reasonably certain that the trough switches are working. I always use a ball, not my fingers to test switches. All the switches register in switch test mode. They appear to work when tested with a meter. The ball release rollover switch diode (switch 02) tests okay. The micro switch (switch 28) has no diode. The switch matrix schematic shows a diode associated with the micro switch (switch 28), but it appears that no diode was actually installed as the games were originally build. I've read where others have tried adding a diode to no effect.
I don't get the purpose of the ball trough micro switch (switch 28). It's positioned to detect a second ball held by the ball release kicker. But that doesn't normally happen. As described above, there appears to be no normal scenario where both balls are held by the ball release kicker. I've tried physically removing this switch from the ball trough and it had no effect on game play. I tried physically shorting out the switch and it had no effect on ball though operation.
The ball release rollover switch (switch 02) under the ball release kicker has several purposes. The switch functions to "unlatch" the ball release kicker. That is, if I hold down the ball release rollover switch (switch 02), the ball release kicker will energize, but will not de-energize until the switch is released. So the switch serves to ensure that the ball has cleared to the shooter lane before the ball release kicker is de-energized. The ball release rollover switch (switch 02) also governs operation of the Exit Chamber side saucer as well as game start-up. In other words, it would be pretty obvious if the ball release rollover switch (switch 02) was improperly stuck open or stuck closed.
There's also a shooter lane switch (switch 25). But the switch is further up the shooter lane and would not detect a ball at rest in front of the shooter. The shooter lane switch appears to have no purpose except for the triggering of a sound effect.
Here's another way to think about all this... If a ball is plunged and drained without scoring any points, the player gets their ball back for another try. This is typical of how many pinball machines function. It's also another scenario where the outhole switch would be closed twice and in this case the game does what it should. This is the only scenario where both balls briefly end up behind the ball release kicker immediately before the ball release kicker serves up the player's replacement ball. Here's the scenario... Ball one is in the shooter lane ready to be plunged. The outhole kicker was the last coil to fire. The ball is plunged into play and manages to drain without scoring any points (not easy to do). The next switch to be closed is the outhole switch. Or in other words the outhole switch is closed for the second consecutive time. The outhole kicker fires for the second consecutive time. Then the ball release kicker fires a brief moment later thereby serving up the replacement ball.
It's as if the programmers did not account for the case where the ball fails to make it over the trough hump. Could this really be a programming oversight? All the hardware components appear to be working as they should. One could speculate that the whole multiball system was not well thought out. Xenon was Bally's first multiball game since switching to solid state. Pinball lore has it that Xenon was intended to be a single ball game. Multiball was a last minute addition. The Xenon ball trough doesn't look like the old multiball ball trough from the EM days. Nor does it look like the trough from Bally's next multiball game. In other words, Xenon appears to be a multiball experiment including a one-off ball trough arrangement. Maybe the design was rushed to production.
A lot of people attribute the Xenon double ball serve problem to some sort of unspecified switch matrix issue. I've tended to dismiss this because the switch matrix hasn't exhibited any other intermittent or flaky behavior. My double ball serve problem is well defined and has always been consistently reproducible. I've been through all the switches and any associated diodes, capacitors and wiring. All the hardware seems solid.
It would be great if someone out there in Internet-land could confirm (or dispute) my observations. If you're reading this and own a Xenon, do me a favor... Take your glass off, push the start button and wait for the first ball to serve. Before plunging the ball into play, reach under the apron and push the outhole switch (don't let the outhole kicker smash your finger). Let me know what happens.
I have no explanation for why the outhole kicker itself hasn't been more reliable. If the outhole kicker had always worked as it should then I would have never noticed this problem in the first place. I have meticulously inspected and cleaned all the kicker components on multiple occasions. I've replaced the coil sleeve. I even experimented with a slightly more powerful coil. Nothing kept it working consistently for more than a few weeks. I've had five other late 70s Bally games with this exact outhole kicker mechanism and none gave me trouble. Maybe something about the kicker or trough wasn't lined up quite right when the game was originally assembled. Who knows?
As noted further up the page I decided to shift my focus to the outhole kicker and modify the lower trough guide. Maybe this is more of a hack than a fix. So be it. I went to eBay and found a spare lower ball trough guide. These parts are plentiful. It's the same part you'd find in most any Bally game through the 1970s. I pulled the lower guide out of my game and cut them both as shown above. I reassembled the parts with screws and spacers such that the beginning of the guide now sits within the playfield slot about ¼" higher from its original position. This does two things. It reduces the leading slope of the guide. And it places the ball more directly in front of the outhole kicker arm. The outhole switch wireform required minor adjustment to account for the new ball position. Hopefully the kicker will now propel the ball over the trough hump such that the ball never rolls back onto the outhole switch. **So far so good**
Update: 14 months later... For unrelated reasons I replaced the original MPU with a new Alltek Ultimate MPU. The new MPU has had no effect on my underlying double ball serve problem. Close the outhole switch twice and both balls go to the shooter lane. Nevertheless, my modified ball trough has worked flawlessly. Accordingly, the problem continues to be moot.
Notes: Exit Chamber Side Saucer Glitch
Here's yet another Xenon odyssey that was driving me insane. I thought I'd elaborate and maybe save someone else some pain. Sometimes when the ball would land in the Exit Chamber side saucer it would just stay there. There's no ball search technology here. Once the ball was stuck, it was stuck. I'd have to pull the glass or restart the machine. The problem was maddeningly intermittent (but getting worse over time). It was frequent enough that the game was no fun to play, but not frequent enough to reliably reproduce under test. This is another problem that dragged on for years before I finally figured it out.
There were two scenarios where I might see the malfunction...
1) Lock is not lit. There are two distinct sound effects associated with the Exit Chamber side saucer. First there's a sound effect associated with the Exit Chamber lamps as they score and advance. The ball is held during this process. Next there's a sound effect associated with the ball actually being ejected. When the game malfunctioned, I'd get the first sound effect, but not the second sound effect and the ball would not eject. That led me to believe that it wasn't a simple issue with the side saucer switch (switch 33). The game knew the ball was there because I'd get the first sound effect and the lamp advance. But I also didn't think it was a simple issue with the side saucer solenoid (solenoid 04) because I didn't get the associated eject sound effect that goes with the firing of the side saucer solenoid (solenoid 04).
2) Lock is lit ("Try tube shot"). There are two distinct sound effects associated with the Exit Chamber side saucer. First there's a sound effect associated with the Exit Chamber lamps as they score and advance. Next there's a sound effect associated with the second ball being served to the shooter lane ("Ahhh"). When the game malfunctioned, I'd get the first sound effect, but not the second sound effect and the second ball would not be served to the shooter lane. That led me to believe that it wasn't a simple issue with the side saucer switch (switch 33). The game knew the ball was there because I'd get the first sound effect and lamp advance. But I also didn't think it was a simple issue with the ball release solenoid (solenoid 03) because I didn't get the associated sound effect that goes with the second ball being served.
I had not observed that two balls ever jam up at the side saucer after one ball is already locked. The game always ejects the first ball before the second ball takes its place. And I didn't recall that the game ever malfunctioned during multiball. The problem only seemed to occur during single ball play before lock was lit or as ball one was locked.
There is one normal in-game scenario that sort of mimicked the malfunction. That is when one ball is locked, but before multiball is started. When the second ball is shot through the tube, the first ball is ejected as soon as the second ball hits the Exit Chamber rollover switch (switch 01). When the second ball hits the saucer switch (switch 33), the Exit Chamber award is scored and advanced, but the second ball is not ejected. The second ball merely replaces the first ball as the locked ball. This is the only normal scenario where a ball would be held at the side saucer without a secondary sound effect, without the side saucer solenoid (solenoid 04) firing or without the ball release solenoid (solenoid 03) firing.
This turned out to be the critical clue. The problem had nothing to do with the side saucer and everything to do with the faulty ball release rollover switch (switch 02). This is the rollover switch on the right side of the ball trough adjacent to the ball release mechanism (solenoid 03). Once I figured this out, the problem was easily simulated by removing the second ball from the ball trough. If the ball release rollover switch (switch 02) is open, the game seems to assume that there are two (non-multiball) balls in play. Accordingly, the game holds one ball at the Exit Chamber side saucer. As noted further up this page, I cleaned and adjusted the ball release rollover switch (switch 02) and I was back in business.
But cleaning the ball release rollover switch (switch 02) was easier said than done. Usually I swipe the contacts a few times with a piece of heavy paper. That wasn't working in this case which had me continuing to question my diagnosis. I re-soldered the switch terminals just for fun. I replaced the switch diode even though the old one tested good. Finally I removed the switch and scrubbed the contacts from several different angles. That did the trick. So I had a tenacious speck of dirt or tarnish in exactly the wrong place. Or the contacts are worn out. Time will tell. But at least I now know where to look.
Note that closing the ball release rollover switch (switch 02) after a ball is stuck in the side saucer does not clear the problem. In other words, it only takes a momentary glitch of the ball release rollover switch (switch 02) and you could be screwed. So, for example, pounding on the game to vibrate the switch contacts won't help. The ball release rollover switch (switch 02) must function cleanly and consistently.
Also note that the ball release rollover switch (switch 02) must be closed in order for the game to complete the start-up sequence. For reasons I can't explain, this was never a problem. The game always started reliably.
Here's one more footnote about how I was able to trigger this problem even with all switches working. Xenon is unusual in that the game can be played without the apron. So the trough action is easily observable. During multiball I managed to simultaneously drain one ball while shooting the tube with the other ball. The first ball landed on the outhole switch and registered as a lost ball. The second ball landed in the side saucer just after the outhole kicker fired, but before the first ball fully traversed the trough and closed the ball release rollover switch (switch 02). Welcome to stuck. It only takes a split second for a ball to cross the trough so this is a highly rare scenario.