(Topic ID: 148577)

williams system 3 world cup switch matrix issue

By heypat

8 years ago


Topic Heartbeat

Topic Stats

  • 17 posts
  • 3 Pinsiders participating
  • Latest reply 8 years ago by heypat
  • Topic is favorited by 3 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

World-Cup-Switch-Matrix_(resized).png
World-Cup-Switch-Matrix_(resized).png
#1 8 years ago

I have managed to bring a world cup back from the dead but have not been able to figure out a switch related issue.

quick list of what has been done:

rebuilt power supply
rebuilt and fixed MPU
went through driver board all appears to be working properly (40 pin, reflow, sockets, upgraded resistors etc.)
rebuilt sound card

game passes initial board test, will coin up and play; however the switch issue makes it impossible to play a full game.

I did go through the switch test and there are NO switches closed.

I also went through all of the switches and 4 of them do not register in switch test mode, whats weird is 2 of the switches that do not read are duplicate switches in series (the other 2 switches in the series read just fine).

The following switches in the switch matrix do not read:

22 (right switch)
23 (left switch)
13 left ball thrower
25 spinner

I have checked the wiring and all diodes and continuity back to the board and all is good. I even replaced the board connectors just to make sure and no luck.

Issues with game play:

out hole switch 14 reads in test mode but does not register a ball drain unless you also press left eject switch 9

Also Left and Right Ball thrower coils do not activate when a ball is in either pocket. All coils do work in test mode so I know the driver board is working.

I also verified that the 2j3 switch row inputs are not strobing and the 2j2 switch strobes are all strobing. Which should indicate a playfield problem?

I though for sure I would find a bad coil or short but no luck so far, Any ideas?

Thanks for any help you can offer !

#3 8 years ago

IC15 or IC16 is prob. bad. i solder in sockets in case chips ever need to be changed again.

#4 8 years ago
Quoted from heypat:

I also verified that the 2j3 switch row inputs are not strobing and the 2j2 switch strobes are all strobing. Which should indicate a playfield problem?

2J2 is the switch send lines, which should be pulsing low. 2J3 is the switch receive lines which should be high unless a switch is closed and then will pulse low. So what you are seeing sounds correct.

Before you go replacing IC's I would figure out if this is a board issue or a playfield issue using the following procedure.

http://www.pinwiki.com/wiki/index.php?title=Williams_System_3_-_7#Testing_the_Switch_Matrix

#5 8 years ago

I checked the switch matrix via the board with a jumper and all 38 switches read just fine. Looks like I'm back to the playfield to try and figure it out.

I already went through most of the switches, so I will double check them and start looking at the grounds

I also noticed that the ball drain switch scores 2k the first time it is pressed and 5k each additional time you make the switch. And you also get points every time you use the flippers.

Thanks

Patrick

#6 8 years ago

During the restore process did you do any work on any switches?

Have a look at the following article. It's for WPC, but the theory part is the same. There's also some good info in the pinwiki article I linked above.

http://pinballrehab.com/1-articles/solid-state-repair/repair-guides/146-switch-matrix-theory-and-troubleshooting

2 weeks later
#7 8 years ago

Its been a while since i have had 5 minutes to look at this. Spent some time on it again last night, I had already went through that tutorial before my original post, when I check the switches in test mode none are closed. For current to flow backwards through a shorted diode and make another switch more than one switch has to be closed. I also went through every switch last night and checked the diodes and also cleaned and adjusted every single one of them. All switches are working as they should now, (the four switches listed above are now working fine).

current symptoms lead to a shorted diode but the switch matrix shows no closed switches and when i am testing i am only making one switch at a time.

current symptoms

start a new game with a ball in the outhole and the left eject coil followed by the R ball thrower coil fire then the outhole coil fires the ball.
Switch 32 (top rollover) causes a ball drain then game will not kick a ball out of the outhole until you make switch 9 even when the ball is in the outhole
Switch 28 (super ball advance) fires the R Ball thrower which should be activated by Switch 16
Switch 36 (left inside ball advance) fires the outhole coil every time it is made

I double checked every switch for the associated coils that are firing when they are not supposed to and even replaced all of the diodes even though they tested good.

Also all coils fire properly in test mode one at a time.

not sure where to look next any ideas?

#8 8 years ago
Quoted from heypat:

start a new game with a ball in the outhole and the left eject coil followed by the R ball thrower coil fire then the outhole coil fires the ball.
Switch 32 (top rollover) causes a ball drain then game will not kick a ball out of the outhole until you make switch 9 even when the ball is in the outhole
Switch 28 (super ball advance) fires the R Ball thrower which should be activated by Switch 16
Switch 36 (left inside ball advance) fires the outhole coil every time it is made

Can you do the testing in switch test and provide as follows:

Closed switch 32 and switches 32 and XX show as closed...

#9 8 years ago

I tried that already, when i test in switch test mode only the switch that i make shows closed, no other switches. I even tried banging on the cabinet, bumping it and banging on the playfield to see if any switches make and nothing.

#10 8 years ago

Correction to my earlier post, switch 32 does not cause a ball drain. I thought the machine was seeing a ball drain because everything goes dead. It is actually tilting when switch 32 is made during game play.

Is it possible this is a MPU issue and data lines are getting crossed some how?

It is obvious someone worked on the MPU board before i got it so there are a lot of unknowns.

has anyone ever experienced this before?

#11 8 years ago

I'm not that familiar with System 3. If in switch test you close two switches are they both shown? I know some earlier Bally's would only show the lowest number switch closed.

When testing in switch test are the balls installed?

Quoted from heypat:

Switch 32 (top rollover) causes a tilt [edit] then game will not kick a ball out of the outhole until you make switch 9 even when the ball is in the outhole
Switch 28 (super ball advance) fires the R Ball thrower which should be activated by Switch 16
Switch 36 (left inside ball advance) fires the outhole coil every time it is made

It would help if you could correlate the solenoid action to which switch number would cause that (as you did with switch 28).

#12 8 years ago
Quoted from terryb:

It would help if you could correlate the solenoid action to which switch number would cause that (as you did with switch 28).

I got it (although I did guess at which tilt). Here's what we know so far, which gives us our rectangles, but they don't line up as would be expected. Are you positive column 2 had a strobe on it?

World-Cup-Switch-Matrix_(resized).pngWorld-Cup-Switch-Matrix_(resized).png

#13 8 years ago

I just double checked and all columns have a strobe and all rows are held high. I went into test mode and confirmed no switches were made (removed ball from game). System 3 switch test mode shows the last switch made, if no switches are made it will not display anything. when I make a switch in test mode I expected a phantom switch to show after i make the switch but nothing. I have tried it at least 50 times and no change, I also left it in test mode for some time to see if a phantom switch would make but nothing (even bounced it around and banged on the playfield).

I also made switch 32 and probed the switch rows to try and figure out which Tilt is being made but the only row that strobes when switch 32 is made is row 8 which is the proper row for switch 32.

Switch 28 (super ball advance) fires the R Ball thrower which should be activated by Switch 16
Switch 36 (left inside ball advance) fires the outhole coil every time it is made which should be activated by Switch 14

Another phenomenon to mention, When a new game is started or switch 9 is made the 3 coils below always fire:

Left eject coil fires every time switch 9 is made (this is the proper switch to fire this coil)
Outhole coil fires every time switch 9 is made which should be activated by switch 14
right ball thrower fires every time switch 9 is made which should be activated by switch 16

Also wanted to Thank you for your help and input on this issue.

#14 8 years ago
Quoted from heypat:

Left eject coil fires every time switch 9 is made (this is the proper switch to fire this coil)
Outhole coil fires every time switch 9 is made which should be activated by switch 14
right ball thrower fires every time switch 9 is made which should be activated by switch 16

Those symptoms point in a completely different direction than the other data.

I don't suppose you have a scope? The original switch problems point at column 2 or a diode issue. Since the system only reports the last switch closed it makes it pretty tough to narrow down a diode issue. Column 2 could be pulsing at a low voltage (say 3 volts) which would pass, but not be enough to work properly in the game.

You could check all of the columns with a DMM and see if there are any voltage differences.

With the limited data I'm really running out of ideas other than checking every diode for being reversed, shorted or the diode lead shorted to ground or a switch lug.

World-Cup-Switch-Matrix_(resized).pngWorld-Cup-Switch-Matrix_(resized).png

#15 8 years ago

I do have a scope and pulled it out and blew the dust off it tonight. I have not used it in over 6 years so I have some relearning to do. It is plugged in and charging and i should have time this weekend to start probing around.

I agree everything points to a diode issue but I checked them 3x, and even replaced ones i though might contribute to the issue even though they all tested good.

i will start by probing the rows and columns to see if anything looks funny. If you have any suggestions other then that I'm all ears.

Wish me luck and Thanks Again!

#16 8 years ago
Quoted from heypat:

i will start by probing the rows and columns to see if anything looks funny. If you have any suggestions other then that I'm all ears.

Make sure you've got a good solid 12 volt pulse on each column. One of the problems with a logic probe is it doesn't tell you the quality of the pulses. You could have low voltage or funky (technical term) pulses.

If you've got a 2 channel scope, put channel 1 on any of the strobes and then watch the row on channel 2 as you close switches.

The 8 columns are pulsed across and then repeated. So if 1 switch is closed then you will get 1 pulse on channel 2 out of the 8 on channel 1. If two switches are closed then 2 pulses out of 8, etc.

One other option but it would take wiring up a couple of adapter cables. You could swap all the row and column signals. So column 1 would instead go to row 1 and row 1 to column 1 and so on. Due to the blocking diodes none of the switches should work in this configuration. If one does work then you know the diode is shorted or reversed (or touching one of the other switch lugs) on that switch.

I'd go with the scope first, but at least there is another option.

4 weeks later
#17 8 years ago

well I have other problems now. Haven't had time to look at this in forever, life has been crazy lately. Any way fired it up to check it out tonight and now its completely dead. won't even get into test mode, symptoms change every time i remove and reseat the rom chips. Roms were in it when i bought it, the software was upgraded but the chips are old and looking rough (one of them still has gold legs). Board has had the scan bees replaced with sips and whoever changed them did a nice job.

Any one in the metro Detroit area of Michigan have a rom burner that can burn me some new roms?

Promoted items from Pinside Marketplace and Pinside Shops!
$ 170.00
Displays
Digipinball Shop
 
$ 11.00
Electronics
Yorktown Arcade Supply
 
$ 27.95
Eproms
Pinballrom
 
$ 10.00
Playfield - Protection
UpKick Pinball
 
From: $ 9.00
$ 9.95
Eproms
Pinballrom
 
$ 12.00
Playfield - Toys/Add-ons
UpKick Pinball
 
$ 12.50
Lighting - Led
RoyGBev Pinball
 
From: $ 90.00
Tools
Pincoder Store
 

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/williams-system-3-world-cup-switch-matrix-issue 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.