(Topic ID: 270879)

Firepower pop bumper coil blowing CPU


By robm

7 months ago



Topic Stats

  • 34 posts
  • 10 Pinsiders participating
  • Latest reply 7 months ago by Schwaggs
  • No one calls this topic a favorite

You

Linked Games

Topic Gallery

There have been 6 images uploaded to this topic. (View topic image gallery).

20200626_172810 (resized).jpg
20200621_101601 (resized).jpg
20200621_101611 (resized).jpg
20200613_093110 (resized).jpg
20200613_093113 (resized).jpg
20200613_093105 (resized).jpg

#1 7 months ago

I recently puchased a non working Firepower. The game didn't boot with original boards in, so i installed a spare Rottendog 327 MPU/driver combo.

The game worked perfectly, probably played 30 games on it. At the same time, i had purchased a Tri Zone project (non working), so put the Rottendog combo board in it, and it worked fine for a few games.

Then i put the Rottendog combo board back in Firepower - after resetting the board to Firepower, the game booted fine, started a game and after plunging the ball, i am 99% sure that exactly when it hit the top left pop bumper, the game died - all lights out and it took out the solenoid fuse on the power supply board. I also smelled a light electrical smell from the rottendog board, but couldn't visually see any damaged chips, or feel any that were hotter than others. The board then wouldn't boot.

NOTE: I did not do anything else to the Firepower while the board was in Tri Zone, so it appears to be a very unfortunate coincidence it happened after the board swap. I also triple checked the plugs and their positions and am quite certain i got that correct.

So i sent it off, along with the original MPU and driver boards for repair. All boards were repaired, for interests sake, the following was replaced on the rottendog board:
Replace 2 x 74hct08 chips , 1 x 14 pin socket , 1 x 74hct240 chip and socket , 1 x 74hct10 chip and socket , 1 x 74hct244 chip and socket , 1 x Uln2803 chip and socket , 1 x 6821 chip , 1 x crystal.

Received all boards back, and installed the orignal CPU and Driver in the machine. I did notice there was a coil firing every couple of seconds - the same coil and i am fairly sure it was one of the kickout holes, so assume a switch may have been a bit sticky. I power cycled the machine and it booted fine a 2nd time, no coils firing and added credits. Plunged a ball, and again, am 99% sure it happened exactly when the ball hit the top left pop bumper, machine went dead, no displays, no lights and a buring electrical smell from the CPU area. I have attached pictures of the wiring - as i really made sure i had the plugs installed correctly, referencing this page: http://firepowerpinball.com/downloads/BackboxWiring.pdf Now the machine has both LEDs locked on upon power up.

So now i am quite frustrated, obviously quite a few $ down the drain in repair costs as twice it appears there is another problem that is taking out parts of the CPU board. I am aware i might be focussing on the top left pop bumper a bit too much, but am fairly convinced its either related to when this coil fires, or maybe when any coil fires?

After inspecting the top left pop bumper, i saw nothing unusual visually. However, i did find 2 things;

1. The top right (not left which is appearing to be the problem one) pop bumper coil had the signal wire fallen off - it doesn't appear to have shorted on anything the way it was hanging - or had a previous owner cut this wire off? - I don't think this is the case as i don't recall it not working when playing the game previously, but i can't be certain

2. Using the capacitor test on my multimeter, the top left pop bumper capacitor on the switch tested quite differently (sometimes 1500, other times 1 - 10) to the other 3 caps on pop bumper switches that all consistently tested aroudn 130 nf. All 4 caps appear to have been replaced by a previous owner and are 22uf 63V electros

I tested the coil diodes (snipped a lead) and they were all fine, along with coil resistance being around 4.3-4.5 ohms on all coils.

So i am starting to think the cap on the switch may be the culprit, however am unsure if this failing could cause such problems on the CPU? I am also very cautious and unsure about how to test anything, as i really don't want to keep frying my boards, paying for repairs and 2 lots of postage! I do have the repaired rottendog board i can use for testing. I will also replace the wire off the top right bumper coil, not sure if this is related or not.

So long post, but do you think i am on the right track with replacing the switch capacitor of the pop bumper, and is there any way of testing without taking more boards out - or are there other checks i should do? Or, could the problem reoccur when any coil fires?

20200613_093105 (resized).jpg20200613_093110 (resized).jpg20200613_093113 (resized).jpg
#2 7 months ago

Even if the pop switch capacitor were shorted I'd think it would just be the same as if the switch were stuck closed, but you don't say that coil is locking on. Since you did find that wire hanging there I would make sure all the wires and diodes on the pop coil, pop activation switch and pop score switch on both pops were not screwed up by previous owner, maybe post a pic of those areas. Are you measuring the caps with one lead unsoldered also?

#3 7 months ago
Quoted from frenchmarky:

Even if the pop switch capacitor were shorted I'd think it would just be the same as if the switch were stuck closed, but you don't say that coil is locking on. Since you did find that wire hanging there I would make sure all the wires and diodes on the pop coil, pop activation switch and pop score switch on both pops were not screwed up by previous owner, maybe post a pic of those areas. Are you measuring the caps with one lead unsoldered also?

No, i was measureing the caps with both legs connected.

However i can't work out why the game worked fine for 30 games when i first got it

#4 7 months ago

Did you activate the flippers, like when the ball went through the top lanes to perform a lane change? Maybe you have a shorted diode on a flipper. Long shot but figure it was worth checking.

#5 7 months ago

Small update.

I removed the playfield from the machine and put on the workbench for easier work. Replaced coil diodes (they tested good), and remeasured caps. All 4 pop bumper caps now measure exactly the same 16-17 uf. Have looked at the coil wiring and can't see anywhere the power or signal wire to/from the coil could be shorting around there.

Also tested for continuity/shorts between all switch lugs in the pop bumper switch stack and the coil wires - all tested fine.

So now i have no idea what to test/check next!

#6 7 months ago
Quoted from Completist:

Did you activate the flippers, like when the ball went through the top lanes to perform a lane change? Maybe you have a shorted diode on a flipper. Long shot but figure it was worth checking.

Pretty sure i flipped the flippers prior to plunging and they were OK. Just replaced diodes on flipepr coils and lane change switch for good measure - they all tested good when i removed them.

#7 7 months ago

Consider adding a Solenoid Saver inline to the Special Solenoid connector.

Saver places an individual fuse between each coil trigger/ground lead, and your driver board transistors (to ground). Fuse much cheaper to replace than driver board repair.

#8 7 months ago

Are you using the original William displays?
I had a Flash game that did what your saying and it ended up being
high voltage from the display feeding back into the mpu.

#9 7 months ago

When the CPU jumps out of program and the blanking stays high the lamp column resistors may heat up and produce a stank.

For now I'd pull out the solenoid fuse and see if the game still boots up. Go back to basics and measure the +5vdc at the MPU and driver board. It could be as simple as the CPU locks up when the solenoid fired because the +5v dropped out. The CPU jumped out of a program and locked on the lamp column drivers making the resistors have a burn smell.

There is a 100uF capacitor on the MPU and the driver board which is a good place the measure +5v across of. See what it reads at the MPU and Driver.

You mention 74HCT08 was used. That flavor of 7408 should not be used at driver board IC1,2,3,4,8, or 9. They can't handle the current to turn on the 2n4401 in the solenoid driver circuits.

#10 7 months ago
Quoted from keith20mm:

Consider adding a Solenoid Saver inline to the Special Solenoid connector.
Saver places an individual fuse between each coil trigger/ground lead, and your driver board transistors (to ground). Fuse much cheaper to replace than driver board repair.

Yep, thats installed, on the left of the driver board

#11 7 months ago
Quoted from wdennie:

Are you using the original William displays?
I had a Flash game that did what your saying and it ended up being
high voltage from the display feeding back into the mpu.

Hmm, interesting. Yes they are original displays. So did your flash boot up OK, but then shut down when a particualr coil was activated? Note the machine shoots the ball into the shooter lane no worries and flips flippers, it only seems to occur when the pop bumper is triggered.

When i either find an issue or get the confidence, i will test all coils before plunging a ball

#12 7 months ago
Quoted from barakandl:

When the CPU jumps out of program and the blanking stays high the lamp column resistors may heat up and produce a stank.
For now I'd pull out the solenoid fuse and see if the game still boots up. Go back to basics and measure the +5vdc at the MPU and driver board. It could be as simple as the CPU locks up when the solenoid fired because the +5v dropped out. The CPU jumped out of a program and locked on the lamp column drivers making the resistors have a burn smell.
There is a 100uF capacitor on the MPU and the driver board which is a good place the measure +5v across of. See what it reads at the MPU and Driver.
You mention 74HCT08 was used. That flavor of 7408 should not be used at driver board IC1,2,3,4,8, or 9. They can't handle the current to turn on the 2n4401 in the solenoid driver circuits.

Thnks for that, will try as you've suggested when i get home. The 74HCT08 was on the rottendog board repair, not the original - wonder does it have different circuitry?

#13 7 months ago
Quoted from robm:

Thnks for that, will try as you've suggested when i get home. The 74HCT08 was on the rottendog board repair, not the original - wonder does it have different circuitry?

it probably does. i assume Rottendog uses mosfets which a 74hct08 could turn on pretty much directly.

#14 7 months ago
Quoted from robm:

Hmm, interesting. Yes they are original displays. So did your flash boot up OK, but then shut down when a particualr coil was activated? Note the machine shoots the ball into the shooter lane no worries and flips flippers, it only seems to occur when the pop bumper is triggered.
When i either find an issue or get the confidence, i will test all coils before plunging a ball

Once, blow mpu after about 4 minutes sling coil locked on.
replaced with rottendog, that lasted a couple minutes.sent it back to rottendog to have him fix it,
he said that every chip that had blown on it got hit with a high voltage spike.
I then got the rottendog led displays installed the repaired rottendog mpu, never had a problem since

#15 7 months ago

Another thing you might want to check, the rottendog mpu, if you install on the original mounting brackets,
there's a chance that the male pins on the back side of the board can short out against that bracket.
There is a posting on pinside about this problem.

#16 7 months ago
Quoted from wdennie:

Another thing you might want to check, the rottendog mpu, if you install on the original mounting brackets,
there's a chance that the male pins on the back side of the board can short out against that bracket.
There is a posting on pinside about this problem.

Yep, was aware of that one thanks. My board looked pretty good, in that it didn't look as though it could short, but i ground the mounting lug back a bit anyway

#17 7 months ago

OK, another (probably expensive) update.

I have been over all the switches and coils on the playfield, and could not see any phyiscal shorts, solder splashes or anything dodgy.

Put the repaired rottendog board back in (remember the original fried after it was repaired), and left the solenoid fuse (2.5A SB, 2nd from left on power supply board) out, as well as the playfield disconnected, and it booted fine.

Then put the fuse in (with playfield still out), and it booted fine, could add credits and the knocker would fire no problems.

So then i put the playfield back in, fired it up with the solenoid fuse out and started a game. It worked fine when target and rollover switches were triggered (ie: scored and made sounds), so then i tried a slingshot, and it killed the MPU again!!!!!!

I did notice IC11 (on the rottendog board) was very hot this time.

So to me, there must be something up with the special solenoids, or switches that drive it. I have checked plugs 2J12 and 2J13 time and time again, and the wire colours and orientation is correct. See pics for wiring.

So, what could cause an activation of a switch that drives the special solenoids to take out chips on on the MPU, with the solenoid fuse out on the power supply board????

20200621_101601 (resized).jpg20200621_101611 (resized).jpg
#18 7 months ago

Probably time to post some pics of the switch in question. The problem at least appears to be playfield-related so far...

#19 7 months ago
Quoted from robm:

OK, another (probably expensive) update.
I have been over all the switches and coils on the playfield, and could not see any phyiscal shorts, solder splashes or anything dodgy.
Put the repaired rottendog board back in (remember the original fried after it was repaired), and left the solenoid fuse (2.5A SB, 2nd from left on power supply board) out, as well as the playfield disconnected, and it booted fine.
Then put the fuse in (with playfield still out), and it booted fine, could add credits and the knocker would fire no problems.
So then i put the playfield back in, fired it up with the solenoid fuse out and started a game. It worked fine when target and rollover switches were triggered (ie: scored and made sounds), so then i tried a slingshot, and it killed the MPU again!!!!!!
I did notice IC11 (on the rottendog board) was very hot this time.
So to me, there must be something up with the special solenoids, or switches that drive it. I have checked plugs 2J12 and 2J13 time and time again, and the wire colours and orientation is correct. See pics for wiring.
So, what could cause an activation of a switch that drives the special solenoids to take out chips on on the MPU, with the solenoid fuse out on the power supply board????
[quoted image][quoted image]

edit nevermind. SSS looks fine.

Slingshot is a special solenoid coil. There is a switch matrix switch and a coil fire switch. Your game is probably letting the 25v enter the switch matrix when that coil slingshot goes full stroke. Really inspect it for problems. that would break the insulation between the 5v and 25v circuits.

#20 7 months ago
Quoted from barakandl:

edit nevermind. SSS looks fine.
Slingshot is a special solenoid coil. There is a switch matrix switch and a coil fire switch. Your game is probably letting the 25v enter the switch matrix when that coil slingshot goes full stroke. Really inspect it for problems. that would break the insulation between the 5v and 25v circuits.

Thats what i was originally thinking, however that wound't happen with the solenoid fuse removed would it? Or does the solenoid fuse only remove power for the standard solenoids? Given its fried the CPU with both a pop bumper and a slingshot, i'm thinking its likely something to do with the common power wire daisy chained? And was going along the thought process that because it happened with the fuse removed, it might be a short to a lamp somehwere?

I'm really just guessing, but haven't worked out a way to logically isolate the problem by testing somehow...

#21 7 months ago
Quoted from frunch:

Probably time to post some pics of the switch in question. The problem at least appears to be playfield-related so far...

I'm starting to think the issue is not one switch, its now fried 2 CPU boards with special solenoid switches (pop bumper and slingshot), once when the solenoid fuse was removed.

#22 7 months ago

So the solenoid fuse was out of the circuit and all you did was trigger the switch? No Solenoid fired?

Measure the voltage at both sides of the slingshot switch. One side should be ground, the other leads to the driver board special inputs. The ground side of the special switch circuit is provided by J3 pin 3 (black wire) on the power board. Double check J3 and make sure it is pinned correctly.

This wire also traverses pin 2 on one of the connectors between the playfield and cabinet. Make sure your connectors are not swapped and that this black wire from the power board is pinned correctly on that connector.

Thinking being that the ground wire on the special switches circuit is not connected to ground.

#23 7 months ago
Quoted from Schwaggs:

Thinking being that the ground wire on the special switches circuit is not connected to ground.

Stands to reason that the ground runs in a chain as well, maybe one set it attached to something it shouldn't be. Williams loved to use those yellow jumpers for multiple uses I could see someone hooking on up to the lamp matrix when it shouldn't be.

#24 7 months ago
Quoted from slochar:

Stands to reason that the ground runs in a chain as well, maybe one set it attached to something it shouldn't be. Williams loved to use those yellow jumpers for multiple uses I could see someone hooking on up to the lamp matrix when it shouldn't be.

Good point. Running 6.3V AC into TTL circuits would be bad as well.

#25 7 months ago
Quoted from Schwaggs:

So the solenoid fuse was out of the circuit and all you did was trigger the switch? No Solenoid fired?
.

Yes, that is correct.

When i ressurect the board again (hoping i can repair myself rather than sending it away!), i will try as you suggested and measure voltages either side of the switch. So if i am testing for a rouge 6.3VAC, i stil lshould pick this up by having one lead on the ground wire, the other on one side of the switch? Then also test both sides for DC volts.

I will have a good look at J3 wire coming from power supply board as well.

#26 7 months ago
Quoted from robm:

Yes, that is correct.
When i ressurect the board again (hoping i can repair myself rather than sending it away!), i will try as you suggested and measure voltages either side of the switch. So if i am testing for a rouge 6.3VAC, i stil lshould pick this up by having one lead on the ground wire, the other on one side of the switch? Then also test both sides for DC volts.
I will have a good look at J3 wire coming from power supply board as well.

No need to wait for the board to be repaired to test this. As long as the power supply and playfield are installed, you can check the special solenoid switches for improper power at the switches. In fact, I would probably troubleshoot this with the CPU/Driver board OUT so you don't risk damaging it again.

Right, measure DC and AC volts at both sides of the switch. Put one lead on ground and the other on the switch.

#27 7 months ago

I had a WMS game where the frayed not insulated GI stands at the end of a run would touch a spinner intermittently and only when the spinner was flipped up at its maximum point. Pain in the ass to find because you could not measure the voltage on the switch line because it only touched at full stroke.

A careful visual inspection goes a long way.

#28 7 months ago

OK, thanks for the advice.

Connected the playfield back up with no MPU/driver, only power supply board with all fuses intact. Measured the following:

Special solenoid switches (the ones with caps across them) - one side was 38VDC, the other side 37VDC, 0.3VAC on both sides between ground and each side

'Normal' switches: 0VDC, the white/colour wire had 8-10VAC, and the yellow wire 6VAC on each switch between it and ground.

A recap of the previous time it fried the CPU
Activating the normal switches was fine, it scores points and kept playing, however the moment i activated a special solenoid switch, it would take out the MPU. So is it normal to have the 38VDC at those switches? And if so, any other tests i can do apart from visual inspection? When closing the switches, there did not appear to be any change in AC or DC voltage

#29 7 months ago
Quoted from robm:

So is it normal to have the 38VDC at those switches?

No, and that's where you're problem must lie. Unfortunately those circuits are designed only to handle 5vdc, 38vdc is definitely frying stuff. You'll have to figure out where the wiring got mixed up or shorted on the pop bumper coil and/or switch(es) that's causing the issue. 38vdc should *only* be at the coil. 5vdc at both of the switches.

#30 7 months ago

If you removed a pop bumper, make sure you didn't pinch a wire under the coil holder. I've been following, but this is the only thing I can think of - you have pinched a wire somewhere that is shorting across to the switches.

#31 7 months ago
Quoted from robm:

OK, thanks for the advice.
Connected the playfield back up with no MPU/driver, only power supply board with all fuses intact. Measured the following:
Special solenoid switches (the ones with caps across them) - one side was 38VDC, the other side 37VDC, 0.3VAC on both sides between ground and each side
'Normal' switches: 0VDC, the white/colour wire had 8-10VAC, and the yellow wire 6VAC on each switch between it and ground.
A recap of the previous time it fried the CPU
Activating the normal switches was fine, it scores points and kept playing, however the moment i activated a special solenoid switch, it would take out the MPU. So is it normal to have the 38VDC at those switches? And if so, any other tests i can do apart from visual inspection? When closing the switches, there did not appear to be any change in AC or DC voltage

You found the problem! With the driver board removed, you should see 0V at both sides of the special solenoid switches. Trace the special solenoid ground wire from the PF, through the connector between the head and cab, to the power supply board. Make sure it isn't miss-wired at any of those connectors. It should end up on the power board, J3 pin 3 (black wire).

#32 7 months ago

Well thanks so much for all your help!

So i looked for the ground for the special solenoid switches and found that the j3 plug on the power supply board was in the wrong spot so the ground pin was actually on pin 4 (not 3) and therefore getting power not ground. So really frustrated that there are plugs that aren't keyed and they can go in the wrong position and also that I must have unplugged and incorrectly replugged that connector between swapping boards.

​​​​​​​Expensive exercise so now I just have to repair the boards again and it should be good to go.

Pic below is how the pin is now and should be but I previously had it one pin up

20200626_172810 (resized).jpg
#33 7 months ago

Damn, good catch. In case you're interested, connector keys can be added to the connector to prevent this type of thing from happening again. You'll just need a key/plug for 0.156" pins like they have here at GPE: https://www.greatplainselectronics.com/proddetail.asp?prod=15-04-0297-STRIP

#34 7 months ago
Quoted from robm:

Well thanks so much for all your help!
So i looked for the ground for the special solenoid switches and found that the j3 plug on the power supply board was in the wrong spot so the ground pin was actually on pin 4 (not 3) and therefore getting power not ground. So really frustrated that there are plugs that aren't keyed and they can go in the wrong position and also that I must have unplugged and incorrectly replugged that connector between swapping boards.
​​​​​​​Expensive exercise so now I just have to repair the boards again and it should be good to go.
Pic below is how the pin is now and should be but I previously had it one pin up
[quoted image]

Awesome job figuring it out!

That connector has a very unfortunate location for the keying pin - on the end of the connector. Even if your molex housing had a key, you still can accidentally plug it in one row up like you did.

Promoted items from the Pinside Marketplace
$ 349.95
Lighting - Led
Pin Stadium Pinball Mods
$ 11.00
From: $ 5.00
Cabinet - Other
Rock Custom Pinball
From: $ 9.00
From: $ 20.00
Cabinet - Other
Rock Custom Pinball
$ 14.00
Electronics
Yorktown Arcade Supply
From: $ 1.00
Playfield - Other
Rocket City Pinball
$ 33.00
Playfield - Decals
Metal-Mods
From: $ 5.00
Cabinet - Other
UpKick Pinball
$ 12.00
Cabinet - Other
Siegecraft Electronics
From: $ 18.00
Apparel - Men
Pinside Shop
$ 279.95
Lighting - Led
Pin Stadium Pinball Mods
From: $ 1.25
Playfield - Other
Rocket City Pinball
From: $ 20.00
Cabinet - Other
Rock Custom Pinball
$ 399.95
Lighting - Led
Pin Stadium Pinball Mods

Hey there! Got a moment?

Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside