(Topic ID: 271055)

NBAFB eject random firing after MPU no boot

By SunMonkeyAZ

3 years ago


Topic Heartbeat

Topic Stats

  • 18 posts
  • 2 Pinsiders participating
  • Latest reply 3 years ago by adalogue
  • No one calls this topic a favorite

You

Linked Games

Topic Gallery

View topic image gallery

F16CEEDF-D201-4C3D-B4EE-7933D266AD00 (resized).jpeg
6682F259-5588-429F-B6E3-71B4E10994AB (resized).jpeg
263DBE4C-EE2B-4A82-ABF0-429B351C8B4A (resized).jpeg
#1 3 years ago

Picked up a NBAFB a couple weeks ago. Everything was fine. In the coin box were the newest roms. Not sure why never installed by previous owner but popped them in last week. Everything ok for a few days then factory settings restored. ???? Checked voltage at U8 and low. Pulled batteries and a little green on the contacts like maybe a little past leakage but board looks clean. Sanded a little and now 4.5v at U8. Fired back up and went into settings to set date/time and it locked up and reset. Then dead. Disconnected everything from MPU and same. 201 and 202 solid. 203 off but would give a little blip here and there but rarely. Took board out to inspect again and put back and it booted up. Connected everything and ran for a couple minutes then froze and reset then died. After more attempts MPU booted with ribbons attached but no 501 on AV board. Touched ASIC at U9 and 501 lit on AV but MPU went back to 201 and 202 solid and 203 off. I ordered a tool to remove ASIC and did that yesterday. Re-seated and it boots every time now. Game fires up and plays but now the ball eject fires a lot randomly. Tried to find consistency in when. Seemed to happen most when defender motor moving or when a ball is in the left eject saucer and the timed backbox flipper shots are going. But happens other times as well. Sometimes it will fire all the balls out but generally I can feel the eject fire but not strong enough to get a ball out. They stay in the trough it just keeps going. Along with that I found that when "in the paint" was started the defender magnet would steal the ball or it would shoot. I couldn't pass. Some games last night were clean. "In the paint" was normal and no eject pops. But when whatever happens happens it causes the "in the paint" to not work and the eject fires. Swapped out trough optos with my WCS94 and same result. Not sure where to look now. Is this an MPU problem since it started after changes there? In test the defender optos weren't all registering. Cleaned and are good now. I noticed in switch matrix the defender optos are showing opened and close when the opto is triggered but they are listed as normally closed in the matrix like the other optos. Is that normal or is that a sign of something being off? It was also late last night so maybe confused but only defender locked opto had the X and the others were dots. Made me confused. Thanks for any help.

#2 3 years ago

Sounds like either switch issues or RAM issue, can you verify all your saucer switches are working in addition to right and flipper buttons that are used for passing?

Maybe try a factory reset?

Maybe roll back to old ROMs and see if they work better?

#3 3 years ago

I haven't hit every switch but I did run through ones I suspected last night. Flippers and saucers and ramp optos. All registered ok and no phantom switches. I plan to run through the whole matrix today after work. I slapped the cabinet and playfield a ton and nothing triggered. I did try the old 1.1 rom that was in initially and it did the same weirdness so put back in the 3.1 rom. I assumed if there was a rom issue I would get an error? But I swapped anyway just in case. I will try a factory reset when I get home this afternoon first. I hadn't thought of that. If it would give me an error at boot up that something was bad, ram or something I would at least have a start.

#4 3 years ago

You can have ROM issues without getting an error, but if you are noticing the same issues with different ROM sets, it's probably not the ROM. You can also have RAM issues without getting an error. My guess is it's something more simple (like switches), but RAM is a possiblity.

#5 3 years ago

So did a factory reset. Still the same.
I went through all the switches in the matrix. All register and no phantom closures of other switches. Still confused from what I posted earlier about the defender optos being normally open instead of closed but posting in the NBAFB owners thread to ask I f that is normal on others. All solenoids fire correctly in test.
In trying things I did find that if I pick up the ball in play the game never does a ball search. The announcer and things happen like a game is going on. So something definitely strange in what is processing.

#6 3 years ago

Defender optos should be NC, as is typical of all opto switches. Are you seeing any other optos as being NO besides the defenders? If you have the original printed manual, the switch matrix table in the back shows you which are NC, but this is hard to see in a scanned version.

This supports a switch problem. Do you have a switch matrix tester? If you disconnect all switch connectors from CPU board what happens in switch test?

#7 3 years ago

I do have original manual and that’s where I saw they are listed as N.C. but in the test they are reversed which led to my confusion. Never seen a N.O. opto but haven’t dealt with a lot of games. The defender is parked in the lock position breaking the opto but that one is showing closed. The other 4 positions are showing open. All the other optos are as expected. The 4 balls are in the trough so showing open. The two optos per flipper are open since the flippers are not engaged.
I don’t have a matrix tester. I will try switch test with the switches disconnected tomorrow and see what that gives me. Thanks for the help thus far.

263DBE4C-EE2B-4A82-ABF0-429B351C8B4A (resized).jpeg263DBE4C-EE2B-4A82-ABF0-429B351C8B4A (resized).jpeg6682F259-5588-429F-B6E3-71B4E10994AB (resized).jpeg6682F259-5588-429F-B6E3-71B4E10994AB (resized).jpeg
#8 3 years ago

I’d also try pulling all the balls from the trough so no optos are blocked and see what shows up in switch test, and move the defender (defender test?) so it’s not blocking an opto.

FWIW - flipper opto switches are not NC, they are NO, an exception to the NC opto rule - and apparently a typo in the manual. Press a flipper button and it goes from a dot to a square, not the other way around.

#9 3 years ago

So with all switches disconnected switch test showed everything open except for 24 always closed. Seems to me the defender optos being reversed in the matrix is normal. In motor test there are 5 boxes and they show an X in the box when the defender gets to an opto so that registers ok. I was still convinced it was something on the mpu board since it started after the no boot issues. So today I lifted each side of U4 and reseated a couple times. After that I played a handful of games and all was clean. I’ll try more tonight but hopefully that was it.

#11 3 years ago

My wife and I played for about 1/2 hour tonight and overall it went well. A couple times the ball eject fired weakly. One time I had a ball in the saucer and while it was doing the trivia game on the DMD a couple balls fired out. Other than that it was good. The in the paint defender action was correct. I could pass and shoot. So it seems to be close. Definitely playable if it stays like this. If it gets weird again I’ll go ahead and reflow the board and reseat everything and then maybe get a new cpu for U4 in case it’s going. Start with that.

1 month later
#12 3 years ago

To update this. Still same issues with random switch hits and trough eject random firings. Since mpu has u20 and u23 socketed tells me they’ve been replaced at some point. If it’s been an ongoing issue with this game based on obvious previous repairs I’m debating replacing the mpu with a rottendog. But if I get a bare board and move over the cpu and asic is it possible this weirdness is from them? What are the symptoms of a bad/failing cpu or asic? I cannot find info on that. But I also don’t want to just replace them first if I don’t need to. I’m so confused.

#13 3 years ago

Can you go into switch test, unplug the switch connectors, and see if you get any random triggers? That could help to rule out CPU. Also get a switch matrix tester from Siegecraft and test the switch matrix without the playfield connected to totally rule out CPU board having switch matrix issues.

My guess is the random coil firings are related to the switch issues.

#14 3 years ago

Where I’m stuck is I never have switch test show issues. With or without connectors everything is normal. It’s only when a game is started that things happen. I can go into switch test and enable high voltage and hit the flippers a hundred times (counted) and nothing is abnormal. Start a game and after a few flipper hits I’ll get a light show and sound like a ramp was hit. Ball eject firing I haven’t seen a pattern. So I’m wondering what’s different about when a game is active vs being in the menu?
If I had errors in test it would give me somewhere to start but nothing. I do get this extra dot that pops up when defender center spot switch is closed. Even shorting direct on the board. Haven’t figured out what it is. Posted about it but no responses.

F16CEEDF-D201-4C3D-B4EE-7933D266AD00 (resized).jpegF16CEEDF-D201-4C3D-B4EE-7933D266AD00 (resized).jpeg

#15 3 years ago

One thing that is different is the gameplay code is running and not the diagnostic menus. Not sure if they occupy different memory space or not. You tried a factory reset right? Latest version of code?

#16 3 years ago

Yes. Done a factory reset a couple times. I have the latest ROM and an older one. I know the ball eject firing happened with both. I will pop the old one back in to see if the ramp entrance activates with the flippers with the old one too. Also ball search doesn’t happen on ball one if I hold a ball. I narrowed those two things down since I put the newer ROM back in. I have the board out right now as I have a remote battery holder arriving today so I’ll probably do some testing later.

2 weeks later
#17 3 years ago

Update: I replaced the battery holder with a remote one and while the board was out I replaced U20. It was already socketed and popped in a new chip. Since then over the last couple of weeks I’ve played a few games or so every few days and no more trough eject firings and center ramp opto is no longer triggering with flipper hits. Maybe it was that chip or pulling the board and re-installing could have had an impact. Either way things have been running correctly now for a bit. Hope that’s it. Thanks.

#18 3 years ago

So it sounds like it was confirmed as a switch matrix issue if replacing U20 solved it. Nice work!

Promoted items from Pinside Marketplace and Pinside Shops!
From: $ 399.95
Boards
PinSound
 
From: $ 27.00
Boards
KAHR.US Circuits
 
$ 89.99
Lighting - Led
Lighted Pinball Mods
 
$ 45.95
Eproms
Pinballrom
 
$ 20.00
Cabinet - Decals
Pinball Haus
 
$ 18.50
Playfield - Plastics
Pinball Haus
 
From: $ 1.25
Playfield - Other
Rocket City Pinball
 
$ 89.99
Lighting - Led
Lighted Pinball Mods
 
4,500 (OBO)
Machine - For Sale
Robbinsville, NJ
8,500 (OBO)
Machine - For Sale
Bay Shore, NY
$ 24.50
Playfield - Plastics
Pinball Haus
 
$ 35.00
Cabinet - Other
Rocket City Pinball
 
$ 5.00
Playfield - Protection
UpKick Pinball
 
$ 9.95
Eproms
Pinballrom
 
$ 29.00
Cabinet - Other
Filament Printing
 
5,495 (OBO)
Machine - For Sale
West Chicago, IL
$ 29.95
Playfield - Plastics
Pinball Haus
 
$ 27.50
Boards
Pinball Haus
 
From: $ 9.99
Eproms
Matt's Basement Arcade
 
Hey modders!
Your shop name here

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/nbafb-eject-random-firing-after-mpu-no-boot 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.