Episode 130 - Magic Screen Timing Bug

For Amusement Only EM and Bingo Pinball Podcast·podcast_episode·13m 56s·analyzed·Jul 19, 2015

Analysis

claude-haiku-4-5-20251001 · $0.022

TL;DR

Magic Screen bingo machines have a timing bug that can burn out the search index coil during large payouts.

Summary

Nick Baldridge discusses a critical design flaw in Bally Magic Screen bingo pinball machines produced between Sea Island and Bounty, where a timing bug can cause the search index coil to remain powered during payouts, potentially burning it out. The episode details the root cause (a race condition involving three switches in the control unit and timer), explains mitigation strategies, and reviews several technical fixes proposed by Phil Hooper, Jeffrey Lawton, and others in the community.

Key Claims

Magic Screen games produced between Sea Island and Bounty (exclusive) are affected by a design flaw where the search index coil can remain powered during payouts, causing it to burn out.
high confidence · Nick Baldridge, citing write-ups by Phil Hooper and Jeffrey Lawton on bingo.cdyn.com
The bug is caused by a race condition where three switches (Control Unit 1A, Control Unit 14B, and a drag switch) must close simultaneously while a payout is occurring, which can cause the timer to step up and shut off motors while the search index coil remains powered.
high confidence · Nick Baldridge, explaining Phil Hooper's technical analysis
Starting with Sea Island, Bally began shutting down moving parts shortly after the fifth ball is shot to reduce wear and tear, accomplishing this via the timer unit stepping to its top position.
high confidence · Phil Hooper, as cited by Nick Baldridge
On Bounty, Bally added another 120-volt circuit path through the replay cam's index unit to prevent the timer from stepping up during payouts.
high confidence · Phil Hooper, as cited by Nick Baldridge
Bally eventually moved the extra 120-volt circuit fix (likely on Border Beauty) by relocating the Control Unit 14B switch to the replay cam's index unit instead of adding a new circuit.
medium confidence · Phil Hooper speculation, as cited by Nick Baldridge
The bug only manifests with large payouts (192+ wins, 300+ wins, 600+ wins), not with small wins like four replays.
high confidence · Nick Baldridge, explaining the technical conditions
Jeffrey Lawton recommends moving the Control Unit 1A switch wires to a new switch on the search wiper stack as the best technical solution because it avoids potential issues with fifth-ball detection that moving 14B would create.
high confidence · Jeffrey Lawton, as cited by Nick Baldridge

Notable Quotes

“the search index coil, it can typically be burned out. And the reason is that it's gotten too much power and it's stayed on for too long.”
Nick Baldridge @ early_discussion — Explains the core failure mechanism of the bug
“If the machine is paying out, and it times out, then that search index coil will stay engaged and continue to receive the full amount of power that's causing it to hold the search disk in place. This is not good. It causes it to catch fire.”
Nick Baldridge @ core_explanation — Dramatic articulation of the severity and consequence of the bug
“it becomes an intermittent race condition. If you happen to push the R button to start the payout while Control Unit 1A is closed, then Control Unit 1A and Control Unit 14B will open and close together. When all three switches close enough times, the timer unit steps up, and the motors shut off while the search index coil is still powered.”
Nick Baldridge (citing Phil Hooper) @ technical_explanation — Clear articulation of the race condition that triggers the bug
“I prefer having another 120 volt circuit in my game if it prevented a coil that has to be used frequently from getting locked on in some instances.”
Nick Baldridge @ discussion_of_fixes — Expresses preference for Bounty's solution over the original design
“If you move the leads off of switch 1A and put an additional switch on the stack on the outside of the search disk... This simple modification will prevent you from having a fire.”
Nick Baldridge (quoting Jeffrey Lawton) @ solution_section — Presents the recommended fix from Jeffrey Lawton
“Switch 1A always works, because it isn't activated until after the fifth ball has been shot.”
Nick Baldridge (quoting Jeffrey Lawton) @ solution_justification — Explains why Jeffrey's solution is technically superior to simply moving 14B
“I think Jeffrey's solution is probably the best not only technically because he mentions a particular problem with stepping up the timer for after 5th ball but also it's probably the easiest to perform”

Entities

Nick BaldridgepersonPhil HooperpersonJeffrey LawtonpersonAlanpersonMagic ScreengameSea Islandgame Bountygame Border Beautygame

Signals

?
product_concern: Critical design flaw in Magic Screen bingo machines (between Sea Island and Bounty) where a timing race condition can cause the search index coil to remain powered during large payouts, potentially burning it out or catching fire.
high · Detailed technical analysis by Phil Hooper and Jeffrey Lawton; specific game range identified; multiple mitigation strategies documented
?
restoration_signal: Multiple repair strategies documented by community experts (Phil Hooper, Jeffrey Lawton, Alan) for addressing the Magic Screen timing bug, ranging from operational workarounds to permanent electrical modifications.
high · Jeffrey Lawton's recommended fix (moving Control Unit 1A to search wiper stack); Alan's alternative (extra switch approach); Bally's factory solutions (Bounty and Border Beauty variants)
?
historical_signal: Bally's iterative design improvements across Magic Screen product line: initial bug introduced at Sea Island, first fix at Bounty (additional 120V circuit), refined fix on Border Beauty (relocated control unit switch).
high · Phil Hooper's historical analysis of design changes between Sea Island, Bounty, and Border Beauty
?
community_signal: Strong community collaboration on technical problem-solving: Phil Hooper's documentation on bingo.cdyn.com serves as central knowledge hub; multiple independent experts (Jeffrey Lawton, Alan) contribute repair solutions.
high · Multiple experts cited documenting solutions; central repository at bingo.cdyn.com; peer review of technical approaches
?
gameplay_signal: Bug manifests only with large payout scenarios (192+ wins, 300+ wins, 600+ wins), not with small wins, indicating the race condition requires extended payout sequences to trigger.

Topics

Magic Screen timing bug design flawprimarySearch index coil burnout mechanismprimaryTimer unit and control unit operation in bingo machinesprimaryRace condition in switch timing logicprimaryTechnical repair and mitigation strategiesprimaryBally design evolution across Magic Screen product linesecondaryBingo pinball machine electrical systemssecondaryCommunity technical documentation and knowledge sharingsecondary

Sentiment

neutral(0.5)— Nick Baldridge maintains a technical, educational tone throughout, presenting the bug as a serious but solvable problem. He expresses appreciation for community technical experts and advocates strongly for implementing fixes to prevent damage. The tone is informative rather than alarmist or celebratory.

Transcript

groq_whisper · $0.042

What's that sound? It's For Amusement Only, the EM and Bingo Pinball Podcast. Welcome back to For Amusement Only. This is Nick Baldrige. Tonight I wanted to talk about a fairly big design flaw in some of the bingos. and I'm going to discuss using some write-ups that both Phil Hooper and Jeffrey Lawton have written up and are posted over on Phil's site, bingo.cdyn.com. The games that we're talking about are Magic Screen games, and specifically the games between Sea Island and Bounty. Not inclusive. So Sea Island and Bounty are not affected, but every other game that was produced between those two are. So the idea is that when you win, say you get three, four, or five in a row, you push the R button on the foot rail, and the game will start a search sequence. In this sequence, several things are going to step and move depending on the positions of the magic screen, and based on the position of the odds steppers, the machine will start to rack up credits. There's a few other switches which are critically important in this sequence, but we're not worried about those at the moment. Those affect how quickly the payout stops. so the problem that we're going to talk about today has to do with the search index coil now this coil when it's powered will stop the search wipers from turning and if you've heard a lot of my complaining podcasts about the search disc in my ticker tape it can be a temperamental beast And I certainly complain about the adjustment on the search index coil switches because it is terribly sensitive. But all that aside, the search index coil, it can typically be burned out. And the reason is that it's gotten too much power and it's stayed on for too long. so when this happens what's going on is that the machine will time out the machine contains something called a timer unit which is going to keep track of both the number of balls that have been shot and also a certain preset amount of time that the machine should stay on in concert with that unit and the rotation of the control unit Now, the control unit is like the brains of the game. I've mentioned before, when I get to score motors, I'll be talking more about the control unit. It functions in a similar way, but much more complex and much cooler. So the timer will step up all the way to say on a Magic Screen game the 8th position or so and it will shut off all the motors Now, if the machine is paying out, and it times out, then that search index coil will stay engaged and continue to receive the full amount of power that's causing it to hold the search disk in place. This is not good. It causes it to catch fire. So, as Phil writes, the sequence of operations is supposed to be a winner's detected, the search index coil is activated, holding the search wipers on the winner, the replay cams are released, and the control unit motor turns them. The replay cams provide pulses to the replay counters and replay registers to award credits. Next, the replay counter reaches the payout level and cuts off power to the search index coil and replay cam's index coil. Now, the replay counter is a separate unit, and there's one for each color of odds. So you can have winners in all the different colors, and the game keeps track of how many replays you've won for each color to verify your payout. It's a very clever system, and it's really cool when you get in the back of one of these. So, starting on C Island, Phil says, Bally decided to reduce wear and tear on the mechanisms by shutting down the moving parts shortly after the fifth ball is shot. The timer unit accomplishes this when it reaches its top step, the eighth in this particular instance. it's a bad thing when the motor stops in the middle of a payout because it keeps that coil engaged as I said so Phil says the obvious way to stop this from happening is to make sure the timer unit can't step up while payout's occurring Phil gives a breakdown of the schematic and he explains how the timer can step up normally in the game. In order to understand the bug, you have to know a little bit about how the timer actually steps. So I'm going to do my best to explain it in a logical way that you can digest in audio form. So, Control Unit Cam 1A. So this is the first switch on the bottom of the switch stack that's closest to the control unit motor itself. And the drag arm switch closing at the same time provide the pulse to the step-up coil. Control unit cam 14B breaks the circuit to the step-up coil while payouts occurring. Control unit 14B is on the same set of cams that control replay winnings. winnings. So they're in rotation constantly while the machine is counting up credits. And so if 1A and the drag switch close at the same time, it's going to step up the timer. The addition of that extra switch at 14B, and that's pretty far away from the motor, it 14 stacks away in fact it makes it less likely for the timer to step up is you have to have three things happen at the same time But it still possible and I certainly seen evidence of that if you listened to my last podcast. So as Phil says, it becomes an intermittent race condition. If you happen to push the R button to start the payout while Control Unit 1A is closed, then Control Unit 1A and Control Unit 14B will open and close together. When all three switches close enough times, the timer unit steps up, and the motors shut off while the search index coil is still powered. You'll need a larger payout for this to happen. So it's not going to do it for your four replay win, but it will do it if you have 192 wins, or if you have 600 wins, or if you have 300 wins, something like that. So let's talk about mitigation. One way to mitigate the problem is just by pushing the R button again. Every time you push the R button, it will reset the timer and will continue paying out. If you're not standing at the machine and you don't push the R button, then you're in trouble. The other thing you can do is turn the machine off and back on. This will disengage the anti-cheat relay. and of course when you cut power to the machine it cuts power to the coil as well. So Phil remarks that on Bounty, Bally added another path on the motor power circuit which went through a switch on the replay cam's index unit. This circuit kept power on the motors while the replay cam's index coil was powered but had the disadvantage of adding another 120 volt circuit to the game. Now, I don't know about you, but I prefer having another 120 volt circuit in my game if it prevented a coil that has to be used frequently from getting locked on in some instances. So, eventually, Bally wised up. Phil speculates probably on border beauty. and Bally moved that extra 120 volt circuit out of the machine and they just moved that control unit 14B switch to the replay cam's index unit, which does the same thing that Bounty does, but in a nicer way. There's a few things, as I mentioned, that you can do to remediate it. There's do nothing, which means that you've got to pay attention any time it's paying out a big winner and make sure that you get paid the right amount or you're on hand to push the R button or turn it off and back on. Another thing you can do is add another switch on the replay cam's index unit and move the two wires from control unit 14B to it. This switch should be adjusted to close only when the replay cams are locked So that's essentially the bally fix Another fix would be to disable control unit 1A By bending up the switch blade so it can't ever close Doing this means the timer unit never steps up far enough to turn off the motors So if you leave the game on when not playing This is not a good option Another fix From a man named Alan Add an extra switch to the stack at the search wiper where the feed ring and R button index coil are located and use the 110 volt circuit for the control motor thus not allowing the control motor to stop running unless the search wiper is back to the zero position. Another fix from Jeffrey Lawton is to move the switch 1A wires over to a new switch on the search wiper stack. Jeffrey's write-up is also on Phil's site, and he says, If you move the leads off of switch 1A and put an additional switch on the stack on the outside of the search disk, the stack contains switches that prevent the R button and the magic screen buttons from activating during register cycle. This simple modification will prevent you from having a fire. I initially thought you could simply move the wires off switch 14B, but if you do that, it's possible for the game to miss the count pulse needed to advance the timer for fifth ball detection, so that won't work. Switch 1A always works, because it isn't activated until after the fifth ball has been shot. He also notes, using this method, the motors will shut down very quickly after the fifth ball is shot, which is okay, in home use especially. If you want the original one- to two-minute delay, wire the extra switch in the search wiper stack in series with switch 1A. this is Nick editorializing but I think Jeffrey's solution is probably the best not only technically because he mentions a particular problem with stepping up the timer for after 5th ball but also it's probably the easiest to perform one of the solutions Phil mentions is very similar to the Bally solution and it's pretty good as well but I've never tried that and I don't know what that would do now I should also say that I've never done any of these because my games are not affected thankfully but if I did have one of these games I would highly suggest it because it will prevent tragedy from occurring when you're not looking. If you want to read up on the problem, you can look at bingo.cdyn.com slash techno slash commonproblems slash sicoil for search index coil. It's a fascinating reading, and Phil highlights the entire timer circuit in the schematic. And I tip my hat to all the guys who have figured out solutions for this particular problem. Phil and Jeffrey and this man named Alan and anybody else. And thank you for joining me. My name again is Nicholas Baldridge. You can reach me at 4amusementonlypodcast at gmail.com. And you can also call us on the bingo line, which is 724-BINGOS1. That's 724-246-4671. you can listen to us on iTunes, Stitcher, Pocketcast via RSS, on Facebook, on Twitter at Bingo Podcast you can follow me on Instagram at nbaldridge and you can listen to us on our website which is foramusementonly.libsyn.com thanks very much for listening and I'll talk to you next time