(Topic ID: 17072)

Pop bumper fires with flipper activity

By Pinwiz1985

11 years ago


Topic Heartbeat

Topic Stats

  • 228 posts
  • 39 Pinsiders participating
  • Latest reply 6 years ago by CNKay
  • Topic is favorited by 54 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

thumb_IMG_2013_1024 (resized).jpg
IMG_2012 (resized).JPG
IMG_2011 (resized).JPG
IMG_1911 (resized).JPG
IMAG1985 (resized).jpg
IMG_3544.jpg
diswasher.jpg
IMG_35431.jpg
IMG_35421.jpg
DSC00211.JPG
IMAG2653.jpg
2.jpg
6.jpg
5.jpg
4.jpg
3.jpg

Topic index (key posts)

2 key posts have been marked in this topic

Display key post list sorted by: Post date | Keypost summary | User name

Post #84 Potential simple solution to random coils firing. Posted by MrBally (9 years ago)

Post #155 Checklist to address random coils firing in Classic Bally Machines. Posted by BJM-Maxx (8 years ago)


Topic indices are generated from key posts and maintained by Pinside Editors. For more information, or to become an editor yourself read this post!

There are 228 posts in this topic. You are on page 1 of 5.
#1 11 years ago

Hey Guys! First time here! Looking forward to gaining some knowledge and assistance on my issue.

I have a 1981 Bally Embryon I'm restoring. I've already done a great deal of component replacement and repair. All 6 circuit boards have had 99% of all the individual components replaced with exception to the MPU board. The CPU and memory are original, but the PIA's are both changed.

Here are some videos, attached some pics of some very bad connectors and rusted components on the MPU board i replaced.

The guy who had this machine before me never did a single thing to it except change BR1, everything else was original. it was also in a damp cold environment. Found tons of burnt connectors and melted resistors and rusted caps, you name it.

Basically, the left top bumper and the bottom left bumper occasionally fire AND score when hitting the flippers a lot. First off, this happens even when the switch is blocked with a piece of paper, so the switch on the coils are not too close. Switch test shows no bad switches, all test correctly. So its definitely EMI coming from the flippers I think.

I also just added some 2.2uF 250v non polar caps to the EOS switches as well to get rid of the arcing (the machine did not have them originally for some reason) My thought was too that adding these caps would also minimize the EMI from the flipper's magnetic field collapsing. I've also added some ferrite chokes on the wires at the flipper buttons.

After all this, i still have the problem

Here is a list of what I have already changed or fixed:

-All components on all circuit boards changed except the MPU board, just the U10 and U11 PIA's changed
-All the PF switch diodes and caps
-ALL solenoid coil diodes replaced
-Flipper button leaf switches
-EOS caps for arcing and EMI reduction
-EOS switches on all flippers
-All burnt out lights and cleaned/resoldered any bad sockets

That's what I have so far. I'm starting to pull my hair out and run out of places to check for problems. I still yet to replace all the molex connectors in the machine too. That is on the agenda to fix, however, they all looked still in decent condition. I have diligently checked all the boards for broken solder joints and have fixed them all. I lucked out and the machine did have its battery relocated in a safe position so no alkaline damage on any of the boards.

My first thought is its got to be something on the MPU board, maybe a bad capacitor or diode?

Thanks so much for the help guys, hope we can trace this issue down!

539863_940229979877_45602245_38103382_.jpg539863_940229979877_45602245_38103382_.jpg 577808_942844919517_45602245_38115348_.jpg577808_942844919517_45602245_38115348_.jpg

Post edited by Pinwiz1985 : added video links and photos

#2 11 years ago

This happens on some older Ballys for some reason. Did you try clipping a capacitor leg off the pop bumper switches?

#3 11 years ago

That i did on the caps. The switch test revealed nothing abnormal, i did notice that the bumper was not as quick responsive, instead of a quick bounce i got a slightly slower bounce, still just as powerful. Even after the cap was removed, the flipper activity still caused it to phantom fire.

#4 11 years ago

I'm gonna turn my attention to the MPU board tonight. Could someone tell me if a leaky capacitor or diode on the MPU board would be the culprit for my problem? Seems that the pulse thats supposed to be read by the switch returns is being misinterpreted with flipper EMI noise. The switches for the bumpers is on J2, when I remove that connector i can go crazy on the flippers all day and it will not fire randomly. I'll remove the PF cap and rest the PF diode before doing further tests

***I do notice that the displays will flicker garbled characters when hitting the flippers like crazy. This doesnt happen with no flipper activity Leads me to beleive the same phenomena is happening to the displays as the bumpers.

Also as a new question, what are the risks of high tapping the transformer? Its currently set for 120V, but my house wiring is always at 116-118. I even have the pin on a separate circuit without any high current draw devices. If I jumped it for 115 would that cause damage? Reason I bring this up too is maybe I can get some slightly stronger flippers.

#5 11 years ago

You've about exhausted all the normal Pop Bumper fixes for Bally's....Here's a thread with additional links that might be of some use:

http://rgparchive-removed.com/rgpforum/showthread.php?t=280559

As for the transformer, I'm surprised it's on 120V anyway. Most times they are tapped at 115. Won't hurt a thing.

#6 11 years ago

Thanks Robertmee. I will def tap the xformer at 115. That should solve my weak flipper issues and any further voltage dip.

I spent some time tonight replacing the 390pF caps for the J2 connector. Problem still exists

Question, can diodes and caps test good on a DMM but be leaky/bad while under load?

#7 11 years ago
Quoted from Pinwiz1985:

, can diodes and caps test good on a DMM but be leaky/bad while under load?

Yes

#8 11 years ago
Quoted from pdman:

Pinwiz1985 said:

, can diodes and caps test good on a DMM but be leaky/bad while under load?
Yes

Then before I start ripping into the MPU board I'll go and change the diodes and caps again on those bumpers. Maybe I heated them up too much with the solder gun the first time I changed them ( I hope its that easy of a solution) Funny that it is only the two left bumpers and not the two on the right. Also all the bumpers are the end switch on separate columns in the switch matrix. I already changed them and diodes tested good, caps dont pass DC so hard to check those.

If that don't work, then the monumental task to replace all the caps and diodes on the MPU board is the next step. Hell might as well do the molex connectors and header pins while I'm at it! Still cheaper than buying a new Alltek board, and its still the original board too.

Question, I never replaced the CPU or memory chips on the MPU, could those be faulty though I get 7 flashes of the LED?

#9 11 years ago

Hey everyone, okay so I had a bit of a breakthrough. I removed all the caps from the 4 pop bumper boards and replaced both diodes on the pop bumpers that are firing randomly...the result

Now only ONE, the top left bumper fires randomly and DOES NOT SCORE! All the bumpers are still pretty responsive as before. So I guess this means that my my problem is not on the MPU board but on the solenoid driver board??

Also does anyone have a pin-out picture of the AS-3071-2 transformer jumper setting for 115 volt? I have seen the schematic, just want to make sure that I have the right pins, last thing I need is jumping this thing wrong and I have a $200 blown xformer or worse.

Also have the replacement header pins and molex connectors on order, maybe this will solve my last few problems I'm hoping!

#10 11 years ago

jumpers for 115

#11 11 years ago

Grr....doesn't support TIF and converting to BMP or JPEG is 5MB....

Go to www.ipdb.org, search for embryon, scroll down to schematics not in manual, and open up the power.tif file. It's a picture of the jumpers for 115V.

#12 11 years ago

Hey Robert! Hope your Memorial day weekend is treating you well! Found the schematic on the link you gave me, thank you! I got the connectors and pins on order. I'll post back with updates soon. Thank you everyone for all the help!

#13 11 years ago

Robert, check this out, further research showed that I'm not the only one that has experienced this exact issue. This thread below echo's my problem.

http://rgparchive-removed.com/rgpforum/showthread.php?t=4484

One of the commenters suggested that the "pull up" resistors are out of spec causing the bumper to phantom fire. In light of the fact that now the bumper does NOT score after removing the caps suggests the problem be on the solenoid driver board. Seeing that I'm jumpered for 120 currently and not 115 (and line voltage is always around 117-118V) I'm speculating that low voltage combined with weak out of spec resistors is confusing the transistor gates on the SDB & MPU when to open??

I'm electronics capable but I'm still learning the mechanics of logic circuits, but can anyone explain to me what a pull up resistor is and why they are needed? Wikipedia gives me the very tech answer but its over my head ATM.

#14 11 years ago
Quoted from Pinwiz1985:

I'm electronics capable but I'm still learning the mechanics of logic circuits, but can anyone explain to me what a pull up resistor is and why they are needed? Wikipedia gives me the very tech answer but its over my head ATM.

Pull up resistors tie logic states high (a digital "1" or 5v), rather than leaving the logic state "floating" (ie. neither high or low). Many switches on these older machines or even some old game system controllers have pull-up resistors so that the logic state of the input to other circuitry is known/guaranteed. It just means the logic state of an input (be it transistor, IC, etc) is being forced to a "1" (5v) or a "0" (0v) and not left undecided.

Hope that helps

#15 11 years ago

Thanks Acebathound, So if I'm understanding this correctly, then an out of spec resistor will not give a clear known logic state at any given time, so the IC or transistor will confuse the state, which in my case leads to phantom firing of solenoids???

Since removing the caps on the leaf switches the phantom pops do not score (with the caps they DID) and have been greatly reduced. However, the problem still exists though.

Attached is a pic from Big Daddy Enterprises. All the parts labeled I have replaced with new components, however, still MANY resistors above that section that are original. I probably should change all those and see if the issue goes away. Same applies to the MPU board, all the components need to be changed there for good measure.

SDB_Driver_section.jpgSDB_Driver_section.jpg

#16 11 years ago
Quoted from Pinwiz1985:

So if I'm understanding this correctly, then an out of spec resistor will not give a clear known logic state at any given time, so the IC or transistor will confuse the state, which in my case leads to phantom firing of solenoids???

There's a lot of things that could be the issue, that's what makes diagnosing fun (or a *real* PITA depending on the day). Voltage feeding back into a circuit, bad capacitor somewhere in the circuit.. broken/cold solder joint on the board(s) or something with the wire harnesses. I had a machine where you'd hit the flipper button & it would go into self test mode. Turned out to be a bad capacitor on the MPU board (I think for the diagnostic switch circuit). Took me *forever* to figure out that one.

It looks like you've done quite a bit of work replacing components on the boards. A lot of people have different philosophies toward that and mine is generally if it ain't broke, don't fix it. Most of the old electronics in these machines are still working very well and will continue to work well for years and years -- exceptions would be some of the electrolytic capacitors used on any pinball machine that's 20-30 years old since they dry out over time. If there's burnt connectors, broken or melted parts, alkaline damaged parts, definitely replace those -- otherwise just replacing everything cause it's old will cost a lot of $$$ when it's really not necessary.

Anyway, with most issues the easiest way to diagnose is if you have a few spare boards around so you can knock the issue down to a particular board. Sometimes you don't have that luxury and have to look at the schematic and logic through things.. then investigate with a multimeter and logic probe. Sometimes this stuff defies logic =) I had 1n4004 diodes on the solenoid board of a Stern Nugent machine that were bad and causing the pop bumpers to work intermittently -- they tested fine with the diode test on the meter. It took me a while to decide to replace what amounts to a few cents in components just to see if it would work.. and it cleared up their random behavior.

If I were you I'd pull the MPU board & driver board and look for broken solder joints on the backside of each connector. Sometimes wiggling the connectors or tapping lightly on the board can expose issues with broken solder connections. Then.. go from there

#17 11 years ago

Thanks Ace! Normally I am of the same viewpoint of if it aint broken don't fix it, but seeing a 31 year old machine, there is bound to be stuff not working right, even if they test good with a DMM, and rather than trying to nitpick every single board, I just rebuilt them all, with the exception to the few components I mentioned above. A bad capacitor on the MPU could very well be my problem too like you, but hunting for which one I have no patience for, just gonna replace them all, the caps are cheap.

I agree with you on pulling the MPU and SDB to see whats up. At this stage of the game, I've already gone this far with major component overall, might as well do the entire machine and not stop short of what may be broken or could break later. While this job has truely been fun, i don't want to make it a habit where 'm working on the machine more than I'm playing it!! HAHA!

I got new molex connectors, pins, and caps/diodes/transistors for the MPU board. I'll post back with some results soon. Thanks again Ace!

1 week later
#18 11 years ago

Robert/Ace

I HATE, I'll say it again, I HATE MOLEX CONNECTORS and worn header pins!! Molex pins are tedious to repin and crimp! Takes so much time! I just got all the parts in yesterday and started from the rectifier board. I replaced all the connectors on the board and repinned the xformer for 115 volts. Now I have a good 45.5 volts for the coils, and all the other TP's are showing in spec, maybe a bit higher on the safe range. AWESOME!

My SDB was in rough shape before I replaced all the diodes/resistors/transistors, broken trace that connected J3 and J2 on the SCB was replaced with a jumper wire and soldered in before I got the machine. This wire when I was testing the rectifier and SCB voltages, it sparked and shorted out the display fuse! The pins that arced were the 190V supply and solenoid coil/rectifier ground returns judging from the schematic.

LUCKILY THE MPU BOARD WAS DISCONNECTED! It blew a few TIP102's, burnt out my left flipper coil and locked some sling and bumper coils on, I replaced them and fixed the stuck coils but my flipper coil is toast, dead shorted on the low power winding, high power winding still good but it buzzes and bounces wildly. After hooking up everything, now half my PF switchs and bumpers do not work, BUT all the solenoids work in self test. MPU boots fine.

My PF switch issue may be completley un-related because I could take a jumper lead to J2 on the MPU and the non working switches off the PF respond when manually jumped, even the non working coils too. So this may be the cause of all my problems with the bumpers. (I FRICKING HOPE) The J2 connector is black is def oxidized (which I discovered is silver dioxide according to PinWiki) and the pins are rough, not smooth like someone sanded them down. So I can add poor maintenance on top of poor storage conditions.

The repair slowly continues with some bumps and scraps but I think I'm on the right track, time to order some extra TIP102's just in case. I also got every component to replace on the MPU board. Glad I'm replacing all the connectors, If J2 is bad I'm sure others are just as bad.

I'll get some pictures up soon today or tomorrow. Right now I'm around ~$300 for a complete electronics rebuild, plus the $20 or so bucks to buy a new flipper coil now. Thankfully my fiance loves this game or I'd be sleeping on the couch for spending so much flip and time on it. This is officially my "hot rod" she tells me.

#19 11 years ago

Perseverance will be rewarded

#20 11 years ago

Rob/Ace

Replaced flipper coil, all repaired!

Fixed non working PF switchs, after re pinning j2 i still had the problem. Wound up making a jumper wire to the return wire which shares all the bumpers and slingshots between the right slingshot and the left top bumper. There must have been a break in the wire somewhere in the harness.

Repinned connectors and header pins on SDB, MPU, and rectifier boards...this was a PITA!!!!

replaced all caps and diodes on MPU except for 820pf caps near j4. I'll do that later this week. Gonna repin the lamp boards next.

After all this guys, that top left bumper still phantom pops but dont score. I'm at my wits end with this issue. Im gonna do more troubleshooting later this week. Ill post my findings, maybe a possible solution will come up..

#21 11 years ago

This may sound really stupid, but are you sure the pop bumper switches are adjusted correctly? Because if vibration is what's causing them to pop, it's sounds like something under the PF. Time to take the 'fix it stick' and start tapping things. Just an observation, you sound like you know what you're doing.

#22 11 years ago

Okay gonna take a more systematic appreach to this issue after reading the schematic more...

When the MPU senses a switch closure on either J2 or J3, it register/score for that switch. If a solenoid is meant to fire from that switch then that logic signal is sent out of J4 of the MPU to J4 of the SDB. From there the signal is routed much like the switch matrix to the appropriate solenoid circuit. Its received by the 1 to 16 binary decoder on the SDB and then sent to one of three IC's which drive the transistor/diode/resistor associated with that solenoid.

So if I'm getting no score with a phantom pop then to isolate the MPU from sending a bad or weak logic signal, removing J4 off the SDB will isolate the MPU from being the cuplrit. If the bumpers phantom pop with that disconnected, then the issue is soley on the SDB. Thats when I will suspect a pull up resistor, all the IC's were changed. If it doesn't phantom pop with J4 removed, then the issue is on the MPU board...

Still have to replace those 820pF caps near J4 on the MPU...wondering now if they are meant for the logic signals to the SDB??

I remember reading somewhere that ALL Bally MPU board sets were prone to leakage and noisy logic lines. Was there any specific reason why? lack of filter caps on the logic lines maybe?

The wheels are turning here....Keep ya posted!

#23 11 years ago

Pinball Shawn, yes, bumper switches adjusted properly. I have even put copy papaer inbetween them, even un soldered the leads and it still phantom fired. Thank you for the suggestion, it was the first place I thought to look. I think I'm battling an inherent design flaw in the MPU logic circuits I'm thinking.

#24 11 years ago

Cool, just making sure A few weeks ago I had a friend who's game wasn't registering when the ball went down the hole. We had 3 guys tearing the boards out and going through everything, I kept asking him, "did you check the switch in the ball return?", 'It's not that' he kept saying while they kept working. Until I reached up and pushed the switch in manually and it registered. But to be honest they still messed with the boards for 5 more minutes until I just adjusted the damn thing!
But if you disconnected it and it's still doing it, you got further issues. Carry on...

#25 11 years ago

Thanks Shawn!

UPDATE: Removing J4 from the SDB also removes the ability to fire the flippers from the flipper buttons! GAHH!! Removing J4 from the MPU while J4 connected of the SDB, the MPU is unpowered unable to boot! DAMN IT! How the hell am I supposed to isolate both of these boards if I can't fire the flippers to test the bumper chatter?

Next course of action is to replace all the 820pF caps (orange/green components) for J4 of the MPU, which by the schematic ARE for momentary solenoid firing logic messaging. The picture below shows my work area. It was taken before I replaced the header pins and the J1, J2, & J3 caps. Seems there was at some point minor alkaline damage but was cleaned up. I already replaced VR1 zener diode no change either. I got replacement transistors (two of them towards the right near the big resistor) I will change out too. I'll check the resistors too while I'm at it. Keep you all posted.

2012-05-11_08-50-22_1.jpg2012-05-11_08-50-22_1.jpg

#26 11 years ago

Shawn/Ace/Robert

Made some interesting progress tonight. I changed out all the 820 pF caps tonight. The phantom popping is still there!!!! GAHHH!!

I have isolated the problem to exclusively the MPU board. I was able to manually jump Q15 of the SDB to activate the flipper enable relay. J4 to the SDB is disconnected, so in theory all the switches should register but the MPU would be unable to message the SDB to fire the corresponding solenoid.

In the instances below I do see a slight flickering of all the displays too, possible stray EMI in the lines.

1. Tried this setup in game mode without the switch caps on the Pop bumpers. I go fricking nuts on the flipper buttons...no score, no solenoids firing at all.

2. Next I RE-ADDED the switch caps to all the pop bumpers, again J4 to SDB disconnected. Continued to go bat shit crazy on the flipper buttons...NO SCORE (VERY INTERESTING), no solenoids firing!

3. I reconnect J4 to the SDB, go crazy on the flippers, now I replaced the switch caps mind you...Phantom pops AND phantom scores.

4. Removed the switch caps, reconnected J4 on SDB, and removed J2 from the MPU board so it couldnt register ANY switches...NO SCORE & NO PHANTOM FIRING!

So my hunch is two things, and it may be one or both causing it. The fact that no phantom fire or score happens with the switches disconnected from the MPU tells me that the pull-up or pull-down resistors are not doing their job properly failing under load, or the simple fact that the board was simply not shielded well enough by design and the wire harness acts like a big antenna picking up stray EMI from the coils.

So I will eventually check into the resistors on the MPU and SDB for the hell of it, BUT I GIVE UP...FOR NOW.

I put some good time into this problem. least it doesn't phantom score so i can live with the popping for now. I'll revisit it later when I'm ready to rip apart the playfield for artwork restore, then I'll see if I can replace the wire with some shielded type. Till then I'm enjoying my game.

Thanks for all the insight and help everyone!!

#27 11 years ago

Thanks for the update....Interesting reading, but you've concluded at the end of the same path historically that others have faced with old Bally's. Stray EMF from the magnetic fields collapsing on the flipper coils.

You could always isolate your test further between EMF and vibration by removing the flipper coil from the playfield and just let it fire in the air. If you still get stray pops, you know it's purely electrical noise and not some mechanical vibration.

#28 11 years ago

Its def EMF, I dont see how it could be mechanical vibration if the bumpers still popped with an index card between the leaf switches.

Check this out Rob... My buddy has a Gottlieb Haunted House. He had issues with a mod of adding a pop bumper board to the up kicker from the cellar. Its a well documented manditory mod. The original relay would always burn out, So adding a PBDB to control it, but it required a polar 4.7mF electroytic cap added between the 5V logic line and ground unlike the stock configuration which originally didn't call for it. This prevented the kicker from firing erroneously when flippers were activated.

I think I might try this approach to the logic lines off J2 of the MPU and see if it makes any difference. Just that connector disconnected prevents phantom popping suggests to me the noise is being backfed from there. The wire harness itself is a big antenna, no getting around that. Sure adding shielded cable MAY help but thats alot of work for a solution that only MAY work. To make the MPU ignore the noise with a filter cap may be just what the doctor ordered.

The stock 390pf of J2 may not be enough capacitance to ignore the noise, or the pull down resistors need to be a lower value (strengthen its "pull" to ground) so when the strobes scan it will ignore the eroneous noise. The software of the machine unfortunately I don't have access to modify which is expecting a certain value range to detect a switch closure, so all bets are off.

When I eventually get around to it I'll do some experimenting. Till then, I must say I'm pretty burnt out troubleshooting this issue. I will continue the fight at a later date however, for future reseachers of the thread here.

#29 11 years ago

One easy test is just lift one leg of the resistors R66 to R69....These feed the base of the pre-driver transistor array from the logic demultiplexer. If the pops don't phantom pop then, you'll know the signal is logic oriented coming from the demux. If you still get phantom pops (you won't get regular pops), then something is turning the pre-driver transistor on besides the logic circuit. There are several paths around the pre-driver collector and emitter to the driver transistor that fires the pop.

#30 11 years ago

R66-69 is for J3. R47 to R57 is for J2. So lift those and check for phantom pops? The value is 56k ohms, was gonna try a 47 k ohms in place see if it phantom pops then. Theory goes less resistance will pull the logic gate to 0 volts stronger forcing there to be a clearer logic signal and raise the threshold for a pop

#31 11 years ago
Quoted from Pinwiz1985:

R66-69 is for J3. R47 to R57 is for J2. So lift those and check for phantom pops? The value is 56k ohms, was gonna try a 47 k ohms in place see if it phantom pops then. Theory goes less resistance will pull the logic gate to 0 volts stronger forcing there to be a clearer logic signal and raise the threshold for a pop

Unless the schematics are wrong J5 IS the thumper bumpers. J2 is the saucer, outhole, drop target reset, knocker, etc. And J2 is R57 to R61, not R47 to R57?

I'm talking about the resistors from the demux to the pre-drivers on the solenoid board, not the MPU.

I'm looking at the embryon solenoid schematic on IPDB.

#32 11 years ago

Ahh my fault, I thought you meant the MPU board.

While I do have plans to replace those resistors on the SDB, with J2 disconnected from the MPU I do not get any phantom popping, and all that time I am going nuts with the flippers the MPU and SDB connected via J4. So the pull down resistors on the MPU are suspect to me.

1 week later
#33 11 years ago

Rob/Ace

Quick update. I have confirmed the problem! The problem is only made worse if your boards are not all tied to a common ground which makes problems like this worse. YES it is true Bally had a poor design of the MPU board where it was susceptable NOT to EMI "noise" from the flippers, but something similar called "ground bounce" and "Vcc Sag." I won't get too much into it here as its fairly technical but I invite everyone to read up on the following pages.

http://en.wikipedia.org/wiki/Ground_bounce
http://www.altera.com/literature/wp/wp_grndbnce.pdf

http://www.allaboutcircuits.com/vol_4/chpt_3/10.html This site is very informative how the switch matrix of a pinball works.

In order to compensate for these issues (which exist to some degree in any electronic circuit) they used Pull UP & Pull DOWN resistors to prevent the logic gate from "floating" causing an unclear logic state. This makes sense to me that the bumper will phantom pop but not score with the switch caps absent, and when they are installed on the switch it will score. The phantom pop is simply too quick for the MPU to detect without them.

The 2nd aspect of the problem is a simple one...HEAT STRESS of the 1/4 watt 5% carbon resistors on the SDB and MPU were more than 5% out of spec. Resistors as they heat up they actually get LOWER in resistance. The lower the resistance of a pull up or pull down resistor means it has a stronger "pull" to either 0 volts ground or 5 volts.

As a test I put a 100kohm resistor in parallel with the 56k pull down resistor (total of 35k ohms, so stronger pull to logic low) for the return line of the top left pop bumper thats been giving me the trouble. Surprisingly the phantom popping STOPPED, but now the bumper next to it phantom pops, which it never did before. So its def the resistors in question, making the resistor values too strong and you will def have other issues, but the 5% tolarance of these components is rather narrow, so its very tough to control.

Thats my phantom pop problem in a nutshell, so the next step is to replace all the resistors that deal with the switch matrix. Hope my findings will help others here, luckily there is something that can be done to fix this issue, thought I admit (and I'm sure Bally would too) is that the fix is not always 100% due to Bally not incorporating enough compensating factors for logic errors in this circuit board generation. Guess they accepted a "good enough" conclusion for this problem back then, and tech limitations of the time also had some play in this too.

Post edited by Pinwiz1985 : too wordy

10 months later
#34 10 years ago

I have the exact same problem with a Bally Space Invaders and was wondering if you finally fixed the problem? If so, could you please give an update. Thanks!

#35 10 years ago

Hey Millertnt,

I was able to fix the problem to the point it only pops once in a blue moon. A few resistors on my MPU were out of spec, and I changed all the caps on the MPU. Its a limitation of the original circuit design. Too much EMI noise on the logic lines, and while adding stronger (lower ohm) resistors it wont completely fix the problem...but it will help alot.

8 months later
#36 10 years ago

I have the same problem, also on an Embryon. My two left bumpers are the only two that ever false fire and the second from the left does it about 10x more than the left one. I have done a few things, most of which did nothing. As part of this I put a scope on the switch matrix on the worst bumper after the diode but before the switch. When firing the bumper there should not be any sign of anything but on my machine I get the following:

Here is the normal strobe pattern which on my machine is pretty messy.
IMG_2972.JPGIMG_2972.JPG

Here is what happens when I fire a flipper. This spike is huge and shows up most of the time.
IMG_2980-638.JPGIMG_2980-638.JPG

Here is another example, this spike is really wide.
IMG_2981-657.JPGIMG_2981-657.JPG

After a while I figured out that when the spike happened at the end of the strobe pulse like this it would cause the bumper to fire. This completely explains the random nature of the misfire. The matrix is strobed at a relatively slow ~120 Hz. When the timing of the flipper fire lined up properly it misfired every time.
IMG_2982.JPGIMG_2982.JPG

I added 3.3 microfarad caps to the EOS switches on the three flippers and while the rate of misfires went way down it still happens occasionally. On the scope there is no obvious difference.

I have a Bally Black Jack which is the from the same generation of Bally and there is no induced spike at all. So something is quite different in Embryon.

#37 10 years ago

BJM,

Great investigative work. It only confirms my theory that the construction and layout of the wire harness for Embryon and early Bally 35 board sets have an inherent shielding problem. Others have discovered that editing the fast scanning of the switches actually virtually eliminates the phantom pops. I see your scope only goes up to 10mhz. The RF spike is probably spiking in the amateur radio and lower broadband range too. I have circumvented this problem best I could with ferrites near the switch matrix connectors near the boards. It has helped a lot. The RF generated by the collapsing magnetic field affects the logic lines, but it also travels on the power lines to the coils too.

Ideally though, to Band-Aid a bad design to the point you get almost no phantom triggering is to add ferrites on every single connector going to the boards. Every wire whether power or signal wire acts like an antenna and would need suppression. You can try ferrites just on those logic lines to those bumpers and it should suppress noise on that one wire, but you got literally hundreds of wires where that RF pulse can gum up the works.

The theory is sound, and parts are pretty inexpensive, its just time consuming to construct a plausible near foolproof fix.

#38 10 years ago

There is another test I tried that I forgot to mention. The coil power feed and returns are all intertwined with the lower voltage lamp strobes and returns and the switch matrix strobes and returns making a giant coupling nightmare. The flipper power makes it's way from the top of the playfield past the right flipper and then down to the lower flippers. Then as in most Bally's of this era, it goes through a 1 A slow blow fuse that then meanders its way back up the playfield past all the other 10 or so coils on the playfield. In Embryon, that is a lot of wiring going back and forth.

I pulled the 1 A slow blow which kills power to all the coils but leaves the flippers working. I re-scoped the same address lines at the bumper. Although it could no longer fire, I wanted to see what noise might still be there due to the flippers. What I saw was a much reduced interference. I had effectively removed a lot of power wiring that can couple into the switch matrix and saw a very reasonable improvement in the switch matrix response.

That being said, I don't understand how these machines could get worse over time, the wiring harness was built poorly from the get-go. My machine is using new Alltek boards in most locations so many aging components have been changed. Playfield caps have all been replaced with cheap ceramic caps (although perhaps they are EMI sensitive too). Does anyone know what these machines were like when new? Maybe they always had this problem.

I am wondering if there are diode issues in the matrix where some diodes are barely working. I actually found my right side single drop target was wired wrong with the diode not being in the circuit at all and it looked like factory work. I fixed that and all sorts of weird things (related to that target) went away.

My last fix will be to run a custom dedicated flipper power circuit away from the existing snarl of the main harness. I would use twisted wire to minimize magnetic fields. My Bally Black Jack has absolutely no interference from the flippers. However someone bypassed the flipper relay at some point, this alone removes a lot of parallel wiring in the harness. I have yet examine what that person did but that might be a clue before creating a whole custom harness.

#39 10 years ago

And I thought I was going above and beyond thinking outside the box to fix this issue ha!

Yes the conducted & radiated RF emissions will always be a problem for these older machines. I don't even think FCC 15 ratings/rules even apply to them. (not that those regulations mean much these days)

As you have read in this thread I have gone through every single board, every single diode, cap, IC, connector and component there is in this machine. There is no foolproof fix for a bad electronic design. Modern pins put all the logic lines on one side of the PF, one harness, and the power wires are opposite side, one harness.

The best way to deal with RF in the older Ballys is to fix any bad component, and use ferrite chokes at key points of the machine to suppress that spike, such as before the circuit boards, at the bumper itself, and along the line of the wire harness. Its a band aid procedure but has proven effective as you have discovered.

#40 10 years ago

I agree with your points but what is the actual degradation mechanism. Why were these machines better when new? Or were they?

#41 10 years ago

This has been a fascinating thread read. Good information in here that goes deeper than the normal "Just replace the chip" response I normally give people here, hahaha. I can't even remember the last time I pulled out my Velleman Scope to probe logic, but now I may be more inclined to do so.

#42 10 years ago
Quoted from BJM-Maxx:

I agree with your points but what is the actual degradation mechanism. Why were these machines better when new? Or were they?

Like with any electronic circuit the tolerance of components changes over time, a capacitance value, resistance value, trigger voltage, current, transistor junction areas heat and become weak etc. I honestly think these machines brand new had the same exact problems. Another difference is these days we have so many more devices creating interference that the minimal shielding in these machines was at time adequate back in the 80's. Today, that shielding is insufficient. There is an RF filter on the incoming AC line on all these machines. They were very basic filtering. Corcom now makes power entry filters with greater common mode and differential mode noise suppression with greater than 60db attenuation.

Even if you had them in perfect working order, bad circuit design will still rear its ugly head. Plenty of work arounds in the world of electronics with ferrite, pull down/up resistors, capacitive decoupling, regulation, and logic signal triggering techniques to "smarten up" the MPU to know what noise is, and what is an actual switch closure. These factors are all part of circuit design.

#43 10 years ago
Quoted from thedefog:

This has been a fascinating thread read. Good information in here that goes deeper than the normal "Just replace the chip" response I normally give people here, hahaha. I can't even remember the last time I pulled out my Velleman Scope to probe logic, but now I may be more inclined to do so.

Haha yeah, see most pin guys I meet are only focused on the machine and playing them. That's great, but me I've always been a gearhead with electronics, wanting to know how they work, and taught myself how to find the problems in electronics. I'm not one to throw parts at a problem unless I'm beyond stumped. Root cause analysis for problems is part of my day job, so the training has permeated through all the other aspects of my life.

#44 10 years ago

One thing i have noticed is if you end of stroke switches spark a lot on a bally you get more phantom coils going off. Replacing EOS switches reduces this. Anyone ever try and put caps on the flipper coils like WMS did in the system 11 era?

#45 10 years ago

BTW, I had a similar issue on the Mr. & Mrs. Pac Man I fixed up recently, and it turned out to be an issue with a faulty switch on the coin door, and it is a very common issue. I cut a wire going to the coin switch (coin 3), and all the matrix issues and random firing stopped. I'd try starting a game up, then removing the connector for the coin door from the MPU, see if you still have the issues. At the very least, you can eliminate that from the equation.

#46 10 years ago
Quoted from barakandl:

One thing i have noticed is if you end of stroke switches spark a lot on a bally you get more phantom coils going off. Replacing EOS switches reduces this. Anyone ever try and put caps on the flipper coils like WMS did in the system 11 era?

The caps are not across the coils, but across the EOS - and only on parallel wound coils:

"System 11 EOS switches use a 2.2 mfd 250 volt capacitor to be used only (part number 5045-12095-00) with the parallel wound FL11630 flipper coils (used on F-14 Tomcat and later). This minimizes the high voltage electrical arc between the contacts of the EOS switch. Games High Speed to Millionaire that use the series wound FL23/600-30/2600 flipper coils do not need this capacitor (it won't help save the EOS switch because the coil is wound in series, not parallel). But on system 11 games with F11630 coils (F-14 Tomcat and later), this capacitor should be installed."
http://techniek.flipperwinkel.nl/wms11/index3.htm

#47 10 years ago

I put 2.2uf poly caps on the EOS switches. As mentioned it only cuts down on the switch arc and prevents the contacts from burning out. They are not required according Bally as the EOS switch's work regardless. They do help suppressing that spark and RF somewhat. Id rather have them than not have them.

#48 10 years ago
Quoted from Pinwiz1985:

I put 2.2uf poly caps on the EOS switches. As mentioned it only cuts down on the switch arc and prevents the contacts from burning out. They are not required according Bally as the EOS switch's work regardless. They do help suppressing that spark and RF somewhat. Id rather have them than not have them.

I'm not certain as to what effects this really has (are there any negative effects? loss of power?) when using a cap on the series wound coil EOS, since it is a different animal than the parallel wound coil. Williams suggests to upgrade the series wound coils to parallel coil to enable adding the EOS capacitor. Anyone know of a reason to avoid adding the cap to the series wound coils?

#49 10 years ago
Quoted from wayout440:

I'm not certain as to what effects this really has (are there any negative effects? loss of power?) when using a cap on the series wound coil EOS, since it is a different animal than the parallel wound coil. Williams suggests to upgrade the series wound coils to parallel coil to enable adding the EOS capacitor. Anyone know of a reason to avoid adding the cap to the series wound coils?

I did it, no downside at all. No significant change to the sparking but the phantom popping was greatly reduced.

#50 10 years ago
Quoted from wayout440:

I'm not certain as to what effects this really has (are there any negative effects? loss of power?) when using a cap on the series wound coil EOS, since it is a different animal than the parallel wound coil. Williams suggests to upgrade the series wound coils to parallel coil to enable adding the EOS capacitor. Anyone know of a reason to avoid adding the cap to the series wound coils?

The only effect is arc suppression. Eventually the extreme heat from the arc begins to corrode the EOS contacts, and that causes loss of power. The capacitor absorbs the extra energy and dissipates it away. Less arc means less RF due to the arc being shorted to the cap.

Promoted items from the Pinside Marketplace
From: $ 14.00
Electronics
Third Coast Pinball
Electronics
$ 30.00
Lighting - Led
Pinballrom
Led
$ 45.00
$ 44.99
Cabinet - Shooter Rods
Pinball Shark
Shooter rods
$ 1,099.00
Flipper Parts
Mircoplayfields
Flipper parts
$ 10.00
$ 18.95
Eproms
Pinballrom
Eproms
From: $ 1.25
Playfield - Other
Rocket City Pinball
Other
There are 228 posts in this topic. You are on page 1 of 5.

Reply

Wanna join the discussion? Please sign in to reply to this topic.

Hey there! Welcome to Pinside!

Donate to Pinside

Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


This page was printed from https://pinside.com/pinball/forum/topic/pop-bumper-fires-with-flipper-activity and we tried optimising it for printing. Some page elements may have been deliberately hidden.

Scan the QR code on the left to jump to the URL this document was printed from.