Pop Bumper Misfiring

[Steve2010]

I have a Bally Skateball that is misbehaving.  The left pop bumper fires randomly, but not completely randomly (I will explain).

At first I suspected it to be a faulty switch capacitor.  I cut the capacitor, but it still misfired randomly.  I began to notice that the coil misfired usually always (if not always) coincidentally when the flippers were activated.  The coil does not misfire every time the flippers are used, in fact, it takes a lot of flipper action to eventually misfire the pop bumper.  But when it does misfire, it always seems to misfire when the flippers are being used.

The other thing that is noticed is that scoring occurs most of the time the coil misfires.  So, this would mean that the MPU is “seeing” a switch closure, even though the switch is not being closed (especially with the switch capacitor cut).  I dismantled the bumper switch assembly just to make sure, and still got the same result.

I then suspected something may not be right with the SD board, and tested the pop bumper coils by grounding the main driver transistor.  The central and right bumpers fired when grounded (as expected) and did not score points (as expected).  Grounding the left bumper coil transistor not only fired the coil (which was expected), it also scored points (not expected).  I then removed the J2 connector on the MPU board, and retested.  This time the left bumper only fired when its transistor was grounded, and did not score points.  This is beginning to look like EMF backlash/inducted voltage from somewhere.

Back to the flippers ….. I also began to notice that sometimes the left bumper would misfire after releasing one of the flippers from a ball trap position.  Web searches reveal EOS switch electrical spikes causing interesting problems, and some with bumper misfirings.  The conclusion here seems to point to the benefit of adding a capacitor across the EOS leaf switches to suppress the electrical arcs.  This also leads to another area – in the Skateball switch matrix schematic (which is available on the ipdb), it DOES show high voltage (500V) caps across the EOS switches, but this is referenced in note 5 at the bottom of the schematic to be used for Germany.  My skateball does not have caps across the EOS switches, and I am curious why only for Germany was this recommended?  As Skateball has 4 flippers, there are 4 EOS switch capacitors as this option.  I am also wondering if a cap may be needed for the leaf switches (2 of these, co-located with the bottom 2 flipper EOS switches) which eventually power the upper flippers.  This is not suggested to do so in the schematic.

Back to the pop bumpers ….. I still wondered why the left bumper coil scored points when the SD board transistor was grounded.  This seems to be yet another source of problem (and may also be independent of what is going on with the flippers?).  I had a closer look at the bumper coils, and then I noticed it.  The diodes on all three bumper coils are not the expected 1N4004 diodes, but very much look like the 1N4148 diodes used for the switch diodes (ie. the coil diodes look exactly the same as the switch matrix diodes).  Maybe the diode across the left bumper is not 100% anymore, and allowing a backlash of EMF when the coil releases causing further EMF to be induced enough to score points in the switch matrix?

Don’t know guys …. maybe I’m dealing with two issues here rather than one.  It would appear to be one or both possible actions.  First could be to replace the bumper diodes with 1N4004.  Second could be caps across the EOS switches.

Suggestions or comments most welcome.

[Strangeways]

I had a similar issue with my Space Invaders. I ended up replacing EVERY capacitor AND diode across every switch in the switch matrix (according to the schematic). The left Pop Bumper would randomly fire when pressing the left or right flipper button. This would happen once in every ten games. Replacing the components in the switch matrix did not solve the problem.

I had an Altek board, and I understand that due to the architecture of the board ( and the clock speed is faster ), "phantom" pop bumper firing would be totally eliminated. I installed the Altek, and there was an improvement. The "phantom" pop bumper firing was not totally eliminated - it still happened once every 20 - 30 games.

These switch matrix issues can be a real pain to figure out.

I would swap the PIAs over and test. This will eliminate the MPU.

You could try swapping out the SD board - CA3081 chip might be firing the pop, but the pop happens when the flippers are pressed.

1N4148 are faster switching diodes. I try to use 1N4004 / 1N4007 in the switch matrix.

Your Skateball was most likely from Germany.

[ddstoys]

I've had a similar issue on a gottlieb may not berelated but check for dry joints on the solenoid board one flipper wouldn affect it but doing both flippers would cause the issue.   But for the few minutes and a few cents I'd be replacing all the diodes for good measure.  

     Report your findings will be good to know the cause

[FirePower]

– in the Skateball switch matrix schematic (which is available on the ipdb), it DOES show high voltage (500V) caps across the EOS switches, but this is referenced in note 5 at the bottom of the schematic to be used for Germany.  My skateball does not have caps across the EOS switches, and I am curious why only for Germany was this recommended? 

I'd guess tighter EMS noise specification / requirements in Germany is the reason for the cap across the EOS.

I'd replace the switch matrix diodes and use 1N4004

Are there time delay filter caps on the switches? If so have a look at them, just remove to test - you'll miss some scoring on that switch (they just make it look as though the switch is closed longer than it actually is) but it will remove the false triggering if that is  the issue.

I'd also have a look around the area where the solenoids are fired. I'm not familiar with your game to know which boards used, but I have mid 80's Bally that uses a BCD to 10 way chip to decode and select which solenoid to fire. I suspect mine might occasionally "monkeys around" in the way it fires.

[Homepin]

1N4148 (1N914 same same) are faster and will do the job across the coils perfectly BUT, being of glass construction are not really suited for the application across coils as they are not as mechanically robust as the 1N400X family and will very likely fracture/break/become intermittant.

[Steve2010]

Hey guys, many thanks for the comments and suggestions.  I will have another go at it hopefully soon-ish - it's challenging to get "investigate & repair time" with the pins, with 3 young kids telling me "enough pinball Dad, c'mon"!  It may be a while till I have time for another update.