4-player Williams EM owners... I may have found a disturbance in the Pinball Force.

Maybe. Let me explain. This one comes from a '72 Fan-Tas-Tic.

fanfail (resized).png

The section at the upper left is a switch in the tilt relay. The score reels (and free credit for high score) get their power through this switch. It is normally closed (i.e. when game is not tilted).

IMG_5971 (resized).JPG

Our culprit, pictured up close. Yellow common on one side, white-orange on the other.

The question to ponder: Why cut power to the score reels when the game is tilted? Tilting the game already cuts power to the score relays, so why worry about the reels?

The action to ponder: If this switch for whatever reason fails to stay closed (game not tilted), the reels won't move. However, the score relays can still be energized! And thus, the nightmare situation... cooked chimes, match coils, and relay coils. Possibly several all at once.

For safety's sake, why not eliminate this switch altogether? This will guarantee power to the reels at all times.

I don't know why Williams did this. Correct me if I'm missing something, but I don't see any benefit to cutting power to the reels during a tilt. The score relays get cut anyway, so what is the point? Seems better to eliminate this potential source of failure (and fire) and permanently jump it closed... just in case.

Right?

I looked at Space Mission/Odyssey and it has the same circuit design. Looks like many other Williams'ses share it as well.

Anyone see any reason why this switch should be preserved?

Nothin'. The top segment is the free credit after reaching a high score portion. Again, seems irrelevant to disable it if you can't score while in tilt.

It did cause a problem in my case. The switch wasn't making good contact, causing infinite relay holds. Luckily only the match coil smoked, but it could have been a lot worse!

I tested my theory with a jumper. Works fine when tilted. Or not... obviously.

I'm going to do this to Space Mission as well. It has the same layout. In fact, EVERY 4-player Williams I checked has the same configuration from Hot Tip (1977) all the way back to A-Go-Go (1966). No 2-player or 1-players use it. Interesting, eh?

I have an emerging theory: Laziness.

Big Chief (1965) was the last 4p released without a separate tilt relay portion. All games before it have BOTH the score reels and relays controlled by the same tilt switch. With A-Go-Go, a second tilt switch was inserted up the line... without going back to remove the first one.

Perhaps the engineers were cruising on autopilot and simply copied the 4p layouts year-to-year?

I say solder it shut. Doesn't seem to do anything except potentially cause a big problem.

Time to issue a sweeping technical service bulletin for 4p Williams owners?

Hi Nick. I see that, but the speed of a reel-click must be in the hundredths of a second. Furthermore, if a tilt occurs *after* a relay was struck, it makes more sense that the point should be scored versus not.

It is hard to believe Williams would have installed this vulnerability just to interrupt wildly improbable reel clicks (which should be awarded anyway). To me, the evidence looks like a design oversight that was copied year after year.

I could be wrong, but at this point I don't see how. If the relays control the reels and depend upon them to unlock, the power should only be switchable at the relays for safety's sake.

They can't. If the score relay power is also cut, the reels won't turn.

It is confusing, but essentially there are two locations in which the tilt relay cuts power. One to the reels, the other to the relays. If the reel-portion is faulty (i.e. switch is physically closed but isn't passing current effectively)... the relay portion will still remain active and burn up.

It's more of a safety/bulletproofing thing. Seems better to let one switch do the job versus risk the redundant one failing and taking out multiple coils.

Right. If that always-on tilt switch is faulty for any reason, it'll cause big problems.

Imagine this... switch becomes faulty during a game (bad gap? dirt? something falling in there?). P1 10 reel locks, then 100, then 1000. 10, 100, 1000 chimes also locked on. Match locked on. Game is still on and flips, but not scoring. Player walks away. Game cooks itself to death.

Already did, and it does. Now commencing with "the tilt safety fix" on all of my 1966-onward 4p Williams'es. Haven't checked every single 4p schematic since A-Go-Go, but it's a good educated guess considering the consistency through the end of EM-production.

Anyone who trolls with "Well, the game works fine if adjusted properly..." is getting a boot to the head, Ti Kwan Leep style!

Yes. Join yellow to white-orange.

fanfail (resized).png

This is one switch in the tilt relay. It feeds power to the reels.

fanfail2 (resized).png

This is another. It feeds power to the relays.

If the tilt relay closes... both switches get cut. This is fine and normal.

If, however, the "reel portion" switch is faulty, WITHOUT the tilt relay closing = problems.

What am I missing?

If the reel EOS doesn't release the relay because the reel cannot move, what usually occurs next?

A different switch in the tilt relay (red -> red/yel/wht) then a switch to reset relay (red/yel/wht -> blu/yel/wht) then a switch to game over relay (blu/yel/wht -> black).

I confess... the answer eludes me. Find it?

Whatever it does, I doubt it is a backdoor to the EOS switches... so the vulnerability is still likely to exist. Just an educated guess.

I'll see if I can find the other end of that wire today. Elusive little sucker on the schematic!

Aha! Bonus points to Rolf! Thanks everyone for your insights. That switch ensures the ball index relay closes (ball saver ends) before a 10/100/1000 point target was hit. It prevents someone from tilting their way out of a bad plunge and getting an extra ball.

At this point I'm pretty confident to call this an action item. All Williams 4p EM's starting with A-Go-Go should have the tilt relay yel/white-orange connection soldered shut for safety's sake... ASSUMING all of the schematics follow suit.

(which they certainly appear to do)

I recommend the safety fix for Grand Prix as well. Circuit topology is the same.

Tilt relay - yellow to white/orange. Solder it closed.

If you'd like to test it yourself... start a game and block that switch with a piece of paper or qtip or whatever is your non-conductive material of choice. Then activate some 10/100/1000 point targets and see what happens. But don't wait very long before unplugging it.

Engineers sometimes build upon what previously existed, right or wrong. Seems like the case here.

...except it actually began as a problem. Perhaps as the machines age, more will pop up.

Typical answer: "Grind away them contacts and hope it doesn't happen again".

Not good enough. The circuit design seemed fundamentally pointless and needlessly vulnerable.

And, as it turns out, it is. Unless someone can explain otherwise?

So...

burns (resized).jpg

Switch was clean and wiping well with good pressure. However, it was one of those edge cases. Some contacts can be a bit pointy on the surface area, which I suspect might have been the case. Then again, I've seen this happen with well-formed contacts too... so I don't know. Maybe 1 out of every 200-300 pairs just needs a little extra persuasion/adjustment. Unfortunately, it just happened to occur on THIS particular switch, which was obviously one of the worst places for that to happen.

In hindsight, a good thing. Now it definitely won't happen again.

True, but this is the case for all EM's, isn't it? At least the fix removes a large variable from the "things to go awry" list. At least seven coils could lock this way, possibly more. Might as well do it.

In global IT circles, we refer to these issues as "vulnerabilities" and deploy "hotfixes" to remedy them. This would qualify as such.

I need some. Badly. Getting a bit grouchy lately... uh oh. Don't want to contract PHOFitis.

Just to pry, is it a "keeping things original" motivation? In the grand scheme of things, a tiny switch in a tilt relay in an antiquated amusement machine is a pretty tiny thing indeed. But it does seem that soldering it closed makes things just a wee bit more certain than not... so why not? *poke* *poke*

...and that multi-bingo is one HELL of a machine. And I want one! So I won't pick on your insistence upon originality. But I'm not certain the engineers had a rock-solid reason for doing this. Some of the stories I've heard at Lockheed... oh... nevermind.

Obviously you and I can adjust relays properly. I'm pretty sure I've rebuilt well over a thousand by now. But the underlying circuit just bugs me. The possibility of seven (or more) coils locking at once isn't something I ever want to happen.

As Mr. EMKC has stated, the likelihood is low. But since it happened to me, I had to figure out what the tradeoff is... and there doesn't seem to be any upside to leaving it as-is.

So, away it goes. *shrug*

Also... I still want to replace my Bally jones plugs with strips of banana terminals.

Original? No. Better? Likely!

But that is a discussion for another day...

In the car world, Zac and I would be resto-modders.

Blasphemy to some, delicious air conditioning in Texas summers to others.

“Your mileage may vary”

I’ve imagined a number of scenarios but fortunately the circuit is a simple one.

The tilt relay is almost always feeding power, so the variable to ponder is when it opens the switch. Is anything lost by *not* cutting the connection while in tilt?

Each reel still has to go through a relay switch. Since those are also switched off in tilt... seems to be a moot point whether they have power or not.

I tend to make these types of decisions relative to downside risk. The switch isn’t likely to fail. But if it does, it’ll fail big... and doesn’t really do anything otherwise.

I don’t think Williams thought this one through. Looks like an absent-minded copy and paste job.

(really hoping I’m not missing anything by now!)

Bean-counting and second-guessing has nothing to do with it. This is simply a technical question being solicited for peer review. If Williams had a good reason for installing this switch, then what is it?

I don't see one.

I -do- see an extremely good reason -not- to do it.

With this one switch failure, at least seven coils can fry. Or more.

Figuring out what Norm Clark had in mind 52 years ago won't be possible, as he is sadly no longer with us. Then again, we don't know if he personally drew up that portion of the schematic.

Older games show the evolution of the tilt relay function. One switch used to be the "catch all"... then another was installed further up the line for the relays. But the old location remained in place.

Why?

At this juncture, two philosophies have emerged:

A) I don't know what Williams had in mind, but they must have known what they were doing despite the potential risk.

B) Williams accidentally overlooked a potential safety issue which typically isn't a problem, but when it is, it's a really big one.

I'm in camp B. Evidence to the contrary will move me to A, if it exists.

Plug failure? As in the jones plugs connecting the motherboard to the head?

Yes, but plugs and solder joints can always be problematic in and of themselves. Those are different issues. This fix only concerns the removal of the tilt relay switch... bridging it permanently closed for an extra hint of safety.

What do you mean by rewire the game?

Yes, this is the intent. The fix only eliminates the switch itself as a potential source of failure. However, it is the most significant piece of the puzzle, as it is the only portion of the circuit expressly designed to cut power to the reel coils.

Agreed, it would be most optimal to bridge at the head and bypass the redundant wire and connection points. But as an easy remedial fix, this should suffice.

Not necessarily. Color TV's were still new back then. Different era. Accounting/inventory practices hardly similar to today's JIT/Lean/6-Sigma methodologies. Also, accountants are not schematic designers... and this looks like a copy-and-paste operation carried over for convenience's sake. Aka "if it ain't broke..."

You are assuming because it is there, it should be. A little skepticism is healthy. Research and testing better still.

I like it. We should optimize things when it makes sense and improves reliability.

A new thread should be created for each issue. Otherwise it would be too easy to confuse different issues. I mean, look at where we are now with closing just one switch.

Do you want to open a new thread for the "Blue A" outhole switch question? Maybe start with "EM Puzzle" in the thread title too.

One more question to ponder: If switching the score coils on tilt was so essential, why only on 4-players?

Haven't seen this configuration on any Williams 2-players or 1-players.

Looking more and more like...

The only difference is a player stepper unit (selects 1-4) vs. a relay (selects either/or). Regardless, both are situated after the open relay switch on their respective circuits. Except for the bonus credit portion. Stepper comes first in that case, relay switch afterward.

Shouldn't make a difference. Stepper or relay or point-to-point wire... same job to do.

Wouldn't matter. It's only there to make a connection and the circuit is bridged by a score relay switch. If the relay isn't closed, nothing happens either way.

Probably not. It might cause two players to get points simultaneously, but I can't imagine two adjacent rivets would have been wired to be dangerous from a partial advance. We'd probably see extensive carbon tracking during normal operation if so... the rivets and wipers are pretty big.

No worries. I'm choosing to call it a mistake. It may not be one, but empirically the apparent uselessness of its default condition versus the rather large potential downside doesn't leave much room for generosity.

Only 80%? LOL I KID, I KID! Thanks JR... just sharing what I see. I've already decided to solder the fix in place. I've sincerely tried to find a downside... just don't see how it could possibly be better off without it.

Not at all & me too. EM puzzles are fun. Except bingos. Those I will leave for you... for now.

I believe the crossover point is Big Chief (1965) to A-Go-Go (1966).

From A-Go-Go onward, the layout remained the same until the end of the EM era for Williams.

Whatever the reason, it would have to answer this question: “Why do only 4-player games need to have the power yanked specifically from the score coils while tilted, despite having another switch which does a similar job at the relays?”

The thing is, the replay/extra ball award circuit still has to travel through a score relay switch, so how is cheating even possible?

As a failsafe, it seems like a bad tradeoff. If the second tilt switch failed (stuck closed) during tilt, it would be apparent pretty quickly what the problem is and at least wouldn't melt down the machine. Whereas the consequences of the always-closed switch failing are much more serious. Also, the likelihood of a switch failing to open within a working relay is less likely than one which should stay closed at all times carrying live current to multiple coils.

However, I'm with you. Let's keep digging. I noticed the bonus wheel on A-Go-Go and Fan-Tas-Tic (and associated award relays) don't travel through the second tilt. Seems like an AHA moment! But as always... all points must ultimately travel through the score relays, so they would still be blocked.

And thus, I'm back to the original conclusion again.

Moving the smokestack "further back" is a win in my book. A smidge more reliable? I'll take it. True, there are further degrees of safety which could be followed (jumpering at the head/plugs/fuse installation)... but so it is with everything EM, eh?

That's an astronomical nitpick, Mr. Baldridge. A maladjusted relay would show issues at the start of every game, even if left original. The fact that it would "also happen in tilt" is moot... because if this is truly some kind of "fallback circuit", it STILL wouldn't help because the tilt relay snaps closed when the ball drains. Thus, the problem would resume immediately afterward.

Do you honestly believe the engineering team thought they should protect against a relay maladjusting itself... only during tilt... and only then requiring the operator to catch and fix the error before the ball drained?

Defending originality in some cases is hard work.

Yes, but again, if "secondary tilt protection" is necessary... why only 4-players and how would it stop malfunctioning during the non-tilted part of the game?

Maladjusted relays are always a risk. All EM platforms can fail when throwing sparks around with relays. Can't be helped. I don't see enough architecture in the tilt circuit alone to prevent it.

I'll admit, I'm getting a kick out of that possibility ("yay me!"). Of course, I may still be proven wrong and am open to that. All in good fun and learning. But you have to admit this is one heckuva puzzler! It isn't purely theoretical either. A real risk is present here. Unlikely, but with potentially big consequences otherwise.

No... what would happen is that a burning coil wouldn't stop burning for those few seconds on the way to the drain. It would never cause anything to burn on its own.

Nope. All enabled.

Now there's a new possibility. However, it wouldn't work. Rotating the player unit rotates all of the reel connections. Score thresholds wouldn't transfer to other players by turning it. Would take some sort of elaborate "just so" type of short to accomplish. Even then, the score relay is key, and it wouldn't be closed.

Yes, it indeed cuts power to the score coils and extra ball/replay coils. I just don't see why it ever should. Especially considering the potential downside.

Even as designed, the only way you wouldn't get the credit is if the tilt happened the first millisecond after hitting the target with the ball. Really splitting hairs on this one. 999 times out of 1000 you'd still get the credit... and this is only an issue when tilting *precisely* on the replay! How often does that happen?

If modified, that thousandth-of-a-second miracle replay tilt would be awarded. Just not an issue.

Replay and EB circuits go through the score relay switch. Can't lock on, modified or not. Why is EB safer?

All good. All of this is for learning and bulletproofing anyway. I do feel slightly more secure with the mod in place, considering it failed on-site with no one watching. Could have been worse. At least I'll sleep well knowing it can't happen again. At least, not at the switch... hahaa

Right, but how would the credit be any riskier? I mean, EB couldn't hold on either after a tilt.

Au contraire, my friend. At least on Fan-Tas-Tic, the extra ball relay is wired through the ball index relay... and that is wired through the other switch in the tilt relay. Thus, if the tilt trips, the index trips, and the extra ball hold is broken.

Tilt = buh-bye extra ball

True, but I'm good with the score relays doing the gatekeeping here. Won't fire or lock without it.

Tilt relay is also on that circuit path. It's the "other wire" Rolf found earlier... the grey-blu.

Appreciate all the thought you guys have put into this. If separating the current draw was the primary goal, I assume it must have been to protect the 4-player stepper unit rivets and wipers... which didn't exist on 2p and 1p games. That's the only architectural difference.

Still, why switch it? The score relay switches divide the circuit between the player unit and tilt switch. Who is to say which portion of the tilt relay would activate/throw sparks first?

(and now I'm going to watch some games from the backbox with people trying to tilt it... field test!)