BK2000 Pop Bumper problem

[Ron Kral]

All of a sudden, one of the pop bumpers on my System 11 BK2000 stopped working. Here's what I found:
1) The coil fires if I short the coil tab - not a coil failure
2) The transistors at Q75 and Q74 test normal using a DMM
3) There are two pop bumper switches, both normally open and both are aligned correctly.
4) In diagnostic mode, the coil does NOT fire (all the others do)
5) When the coil fires, it operates the 2nd switch which IS in the switch matrix to cause the scoring. This 2nd switch DOES register in diagnostic mode when operated manually.
6) The board headers and plugs are clean and reseated
7) All the other switches in it's column and row register normally in diagnostic mode.

This is a strange setup compared to more recent machines.
The leaf switch that the cup/point operates is NOT in the switch matrix.
When the top switch closes, the coil fires and operates the lower switch which IS in the switch matrix and causes the scoring to occur.

When this first started, I fooled around operating the ring manually and the coil fired intermittently a couple more times before giving up completely.

Mystery...
1) Why did it just stop out of the blue?
2) Why did it work intermittently for a bit longer? Is this indicative of the issue?
3) There appears to be a resistor? and a capacitor? in the pop bumper switch assembly. What are these for? Could they be the failure? What are these part's ratings?
4) Since the top switch isn't in the switch matrix, how does the CPU recognize it and fire the coil?

Any insight on what could be causing the failure and how to locate it or where to go from here is most appreciated.

Ron Kral

[John Robertson]

Post by Ron Kral
1) The coil fires if I short the coil tab - not a coil failure
2) The transistors at Q75 and Q74 test normal using a DMM
3) There are two pop bumper switches, both normally open and both are aligned correctly.
4) In diagnostic mode, the coil does NOT fire (all the others do)
5) When the coil fires, it operates the 2nd switch which IS in the switch matrix to cause the scoring. This 2nd switch DOES register in diagnostic mode when operated manually.
6) The board headers and plugs are clean and reseated
7) All the other switches in it's column and row register normally in diagnostic mode.
This is a strange setup compared to more recent machines.
The leaf switch that the cup/point operates is NOT in the switch matrix.
When the top switch closes, the coil fires and operates the lower switch which IS in the switch matrix and causes the scoring to occur.
When this first started, I fooled around operating the ring manually and the coil fired intermittently a couple more times before giving up completely.
Mystery...
1) Why did it just stop out of the blue?
2) Why did it work intermittently for a bit longer? Is this indicative of the issue?
3) There appears to be a resistor? and a capacitor? in the pop bumper switch assembly. What are these for? Could they be the failure? What are these part's ratings?
4) Since the top switch isn't in the switch matrix, how does the CPU recognize it and fire the coil?
Any insight on what could be causing the failure and how to locate it or where to go from here is most appreciated.
Ron Kral

Hi Ron,

The computer only recognizes that the pop bumper has fired when the pop
bumper yoke activates the scoring switch. This switch must be clean and
properly adjusted for the pop bumper to operate. The most common problem
is a skirt (and spoon) driven switch not registering closed long enough
for the driver transistor to fire. the resistor and capacitor are there
to stretch the closure so the coil will fire even if the skirt is only
pushed enough to close the skirt switch momentarily.

It sounds to me like the skirt switch is dirty, so clean it before
looking elsewhere. If you have an ohmmeter then read across the switch
with the power off and make sure the resistance drops close to zero ohms
when the switch is activated by pressing on the skirt. If it does, then
turn the game power on and make sure the switch shows around 5.0VDC
across the contacts.

One of the above is the most likely problem...

John :-#)#

[seymour.shabow]

Post by Ron Kral
This is a strange setup compared to more recent machines.
The leaf switch that the cup/point operates is NOT in the switch matrix.
Mystery...
4) Since the top switch isn't in the switch matrix, how does the CPU recognize it and fire the coil?

The cpu doesn't recognize it, it's connected to completely different
circuitry that fires it directly via logic chips. Williams did this as
it was felt that the cpu couldn't react fast enough to a regular switch
matrix hit and fire the solenoid that way.

You can read up on Special Solenoids at pinwiki.com and various other
online resources if you are interested.

The fun ones for the uninitiated is when you have it firing correctly in
solenoid test, but not with the activation switch. Then you realize
that these are 2 completely different circuit paths, but it helps as
well then that you know it can't be the opposite path causing the issue.

BTW, the activation switch just grounds an input on the main board, so
you can test the path by grounding that leaf that goes to the board.
(Usually a single wire instead of a daisy-chained wire.) Do not hold it
on the tab though as it's just like ground the transistor tab and will
burn out the transistor if held on too long. (The fuse should pop in
this situation but often doesn't on WMS games)

A common problem is the header pins are cracked on the main board
causing intermittent failures.

[Ron Kral]

Thank you both for responding to my post...I used your suggestions to test some more.
UPDATE:
Double checked the switch. It's clean and gapped properly.
Continuity check when the switch is operated manually (or via the playfield ring) = GOOD
Reflowed solder on 1J18 pins (They looked fine - no cracked joints - this was just a precaution)
Tested continuity from the switch tabs to 1J18 Molex plug - GOOD
There are 4 wires into 1J18 Molex - a common and a home run for each pop bumper.
Used the DMD to test switch operation on all three wires against common
by connecting and operating the corresponding pop bumper ring - all switches test OK
Just for fun, (re)punched down the 4 wires into the 1J18 Molex plug
All 3 resistors test good at about 100 OHMs
Continuity on the board from the Molex plug top of each 1J18 home run wire to the first component on the board is GOOD
Shorting the tab on Q73, Q74 and Q75 operates each pop bumper coil in succession.
I was unable to get any of the 3 coils to fire by grounding the home run switch tab to the side rail - I must be doing something wrong.

After all that, the left pop bumper still doesn't operate either in game or diagnostic mode.

It's beginning to look like it's the logic chip that controls that pop bumper(s).
Anything else I should try??

BTW, it's clear the board has been repaired, perhaps a couple of times.
It's obvious that it had battery damage that's been cleaned up and repaired.

I have the manual, but I'm not great at reading schematics...
From what I can gather, this circuit goes through the 7402 chip at U45 which has been socketed and replaced.
Does each pop bumper have it's own control chip, or do they all run through the same one?
Could some other chip unique to this pop bumper be failing? Which one(s)?

Thanks again for you excellent advice and help.

Post by seymour.shabow

Post by Ron Kral
This is a strange setup compared to more recent machines.
The leaf switch that the cup/point operates is NOT in the switch matrix.
Mystery...
4) Since the top switch isn't in the switch matrix, how does the CPU recognize it and fire the coil?

The cpu doesn't recognize it, it's connected to completely different
circuitry that fires it directly via logic chips. Williams did this as
it was felt that the cpu couldn't react fast enough to a regular switch
matrix hit and fire the solenoid that way.
You can read up on Special Solenoids at pinwiki.com and various other
online resources if you are interested.
The fun ones for the uninitiated is when you have it firing correctly in
solenoid test, but not with the activation switch. Then you realize
that these are 2 completely different circuit paths, but it helps as
well then that you know it can't be the opposite path causing the issue.
BTW, the activation switch just grounds an input on the main board, so
you can test the path by grounding that leaf that goes to the board.
(Usually a single wire instead of a daisy-chained wire.) Do not hold it
on the tab though as it's just like ground the transistor tab and will
burn out the transistor if held on too long. (The fuse should pop in
this situation but often doesn't on WMS games)
A common problem is the header pins are cracked on the main board
causing intermittent failures.

[Bill Ung]

Left Jet Bumper is Solenoid #17, controlled by Q75; which is upstream controlled by Q74, a 2n4401 (TO-92) smaller transistor. This component lives between U45 (the 7402) and Q74.

If you have a 7402, I'd try swapping that first. If not, go through the Wiki and see how to test Q74. I haven't done much of it, myself.

Good luck!

Post by Ron Kral
Thank you both for responding to my post...I used your suggestions to test some more.
Double checked the switch. It's clean and gapped properly.
Continuity check when the switch is operated manually (or via the playfield ring) = GOOD
Reflowed solder on 1J18 pins (They looked fine - no cracked joints - this was just a precaution)
Tested continuity from the switch tabs to 1J18 Molex plug - GOOD
There are 4 wires into 1J18 Molex - a common and a home run for each pop bumper.
Used the DMD to test switch operation on all three wires against common
by connecting and operating the corresponding pop bumper ring - all switches test OK
Just for fun, (re)punched down the 4 wires into the 1J18 Molex plug
All 3 resistors test good at about 100 OHMs
Continuity on the board from the Molex plug top of each 1J18 home run wire to the first component on the board is GOOD
Shorting the tab on Q73, Q74 and Q75 operates each pop bumper coil in succession.
I was unable to get any of the 3 coils to fire by grounding the home run switch tab to the side rail - I must be doing something wrong.
After all that, the left pop bumper still doesn't operate either in game or diagnostic mode.
It's beginning to look like it's the logic chip that controls that pop bumper(s).
Anything else I should try??
BTW, it's clear the board has been repaired, perhaps a couple of times.
It's obvious that it had battery damage that's been cleaned up and repaired.
I have the manual, but I'm not great at reading schematics...
From what I can gather, this circuit goes through the 7402 chip at U45 which has been socketed and replaced.
Does each pop bumper have it's own control chip, or do they all run through the same one?
Could some other chip unique to this pop bumper be failing? Which one(s)?
Thanks again for you excellent advice and help.

Post by seymour.shabow

Post by Ron Kral
This is a strange setup compared to more recent machines.
The leaf switch that the cup/point operates is NOT in the switch matrix.
Mystery...
4) Since the top switch isn't in the switch matrix, how does the CPU recognize it and fire the coil?

The cpu doesn't recognize it, it's connected to completely different
circuitry that fires it directly via logic chips. Williams did this as
it was felt that the cpu couldn't react fast enough to a regular switch
matrix hit and fire the solenoid that way.
You can read up on Special Solenoids at pinwiki.com and various other
online resources if you are interested.
The fun ones for the uninitiated is when you have it firing correctly in
solenoid test, but not with the activation switch. Then you realize
that these are 2 completely different circuit paths, but it helps as
well then that you know it can't be the opposite path causing the issue.
BTW, the activation switch just grounds an input on the main board, so
you can test the path by grounding that leaf that goes to the board.
(Usually a single wire instead of a daisy-chained wire.) Do not hold it
on the tab though as it's just like ground the transistor tab and will
burn out the transistor if held on too long. (The fuse should pop in
this situation but often doesn't on WMS games)
A common problem is the header pins are cracked on the main board
causing intermittent failures.

[seymour.shabow]

Post by Ron Kral
It's beginning to look like it's the logic chip that controls that pop bumper(s).
Anything else I should try??

If you have a logic probe you can test the input and output from the
chip. The DMM isn't fast enough to react here.

Did you ground the inputs to that chip? That will fire the pop if the
transistors are good and the chip is good.

It's probably the chip though.

[Ron Kral]

Post by seymour.shabow

Post by Ron Kral
It's beginning to look like it's the logic chip that controls that pop bumper(s).
Anything else I should try??

If you have a logic probe you can test the input and output from the
chip. The DMM isn't fast enough to react here.
Did you ground the inputs to that chip? That will fire the pop if the
transistors are good and the chip is good.
It's probably the chip though.

I do have a logic probe...
What are the pins supposed to read?
What's a good 5V source for the probe in the backbox?
There does not appear to be a 5V test point on the CPU board.

[seymour.shabow]

Post by Ron Kral
I do have a logic probe...
What are the pins supposed to read?
What's a good 5V source for the probe in the backbox?
There does not appear to be a 5V test point on the CPU board.

https://www.pinwiki.com/wiki/index.php?title=Williams_System_9_-_11#Connecting_a_logic_probe_to_the_MPU_Board

U45 and U50 are the 7402 inputs. The input connector from the PF is
1j18. the pins that get grounded are 12, 5, 2, 9 with outputs to 13, 4,
1, 10.

So stick the probe on 12 of u45, should read high normally. Ground it
via 1j18-5 - probe shows low.

Stick probe on pin 13 of u45, should be high (there's a pullup resistor
doing so if the chip pin is floating to ensure this). Ground 1j18-5
again, probe should go from high to low.

Repeat for other pins. I just 'rake' the 1j18 input connector to find
the correct one for the pin I'm on.

Note you have to do all this in test mode or game mode, not attract mode.

The outputs of the 7402 are what enable the predriver and driver
transistors, which of course can be tested many other ways.

Don't discount the support passive components around the transistors
while rare they do go bad.