Creature switch errors .. but why ?

Look at the switch matrix diagram and see if they are the last three in the string of switches, may be simply a broken wire, to have three not working is a common association. Being 41,51 and 61 check 21 as 31 is not used and there may be a broken wire. Seems to have all the signs of a broken common wire. Check White-Brown at switch 21 if and follow this wire as that is the common in that string.

Test the cpu with a jumper wire with pf sw connectors removed just to rule it out as a problem.

Dave, I have had a similar prob before and not all the switches in the affected row/column played up. Same as you I had 3 I sequence so I replaced the chip and it fixed the prob, Not sure why but I didn't care as it solved the prob.

Thanks for the suggestions so far guys .

@hotty .. yeah first thing I suspected and checked but tests indicate wiring is OK

@Firepower and JPD .. yeah worth a try to test CPU with a jumper wire .. I have noticed that switch 21 (slam tilt) is connected via the coin door interface board , not directly to the CPU so that is probably a big clue pointing towards a chip problem no doubt. Easy enough to check with Dave's suggestion

Now you mentioned the PCB for switch 21 is located on the cabinet. As the others stated jumper testing will eliminate a switch matrix board fault or play field fault. I do recall having a switch problem with the coin door switches on a BR and it was related to the connector on that board it was a real problem as I couldn't access the diagnostics. it may not be related but its worth a try removing it and look at the traces and possibly reflow the soldered pins and check the connectors. Fiddle with the wiring as this might show the problem in switch level, edge test on the DMD. Keep looking pretty sure you'll find it.

Gday all .

 

Switches register closed if I connect relevant pins on row and column connectors on the CPU board so that isolates the fault to somewhere on the playfield .

 

Switch 21 (slam switch on coin door) works fine so I'd expect no issue with the coin door interface board or connectors there . According to the schematic , switch 21 effectively terminates on the coin door interface board and is connected back to the CPU via J212 .

 

Will have to have another look at these 3 switches on the playfield itself . I guess next steps will be reflow connections on each switch and replace diode on each switch .. will let you know the result .. possible its a connector in the wiring loom somewhere too I spose

 

Cheers

Dave

Did you check the switch diodes in place of out of circuit mate? For 10c each they are a quick fix to rule them out.

Dave

 

Gday mate .. I was lazy and checked them in circuit .. I have about 95 new ones so just got to pull my finger out and replace them anyway .:)

 

Its just that I like knowing what caused the damn fault each time so I muck around too much probably

Have a look at the wiring diagram - by unplugging the pf connectors at the cpu you can properly test the diodes "out of circuit" so no need to replace them just on suspicion.

Have a look at the wiring diagram - by unplugging the pf connectors at the cpu you can properly test the diodes "out of circuit" so no need to replace them just on suspicion.

 

Very true Dave .. did not occur to me DOH!

I had another look at this and still can't find the fault .

Replaced the diode on each switch as I have plenty of them , but made no difference .

 

I am at a loss as to what is causing this problem.

 

Definitely have continuity on the row 1 wire (white/brown) between the 3 switches and the CPU connector .. weird

I know this won't be accurate as Creature is a Midway but I had a switch matrix issues similar to what your describing on a Williams Indiana Jones. On AA advise which was correct was to remove IC U20 on the CPU board, replace with a ic socket & replace the IC. This IC manages the switch matrix & cause all kinds of switch errors when faulty, Hope this helps. The I.C is a ULN 2803A (darlington transistor array) .

Yeah, CFTBL and IJ are exactly the same with boardsets - and often U20 is the culprit. But goodolddays has isolated the issue to the playfield by testing the board by itself.

 

When you checked continuity - did you check plugs? I think i had an issue like that once and the crimp on connector that the green wire (middle lug of switches normally) was not quite connecting most of the time. A real pain to trace, but maybe put the machine in switch edges and wriggle each connector (solder connections as well) I would start at either end of the daisy chain.

 

Also worth checking plugs at the board end.

A few more thoughts for you Dave. Agree with Stuba to closely check how diodes are installed, should be

 

- diode banded end to switch common terminal

- diode non banded end to NC term and white (row) wire

- green (col) wire to NO term

 

some switches I've seen didn't have the NC term and they use a little bakerlite plate on the side with a lug to mount the diode.

 

Are there any optos in either row or col - if so I'd disconnect them for testing purposes

 

I'd also remove the cpu board and examine connector headers from the rear

 

I'd connect the col switch wiring to the cpu and use a paperclip and a jumper lead to swap the troublesome row across to another position and see what switch tests says.

Thanks for the feeback and ideas fellas.

 

@ Stuba , diodes look like they are the originals .. who really knows , but the ends were through the switch contact holes and then bent around

Agree it will sure be interesting to see what the damn fault is . Will try your idea of isolating each switch too, hadn't thought of that

@Gemini2544 .. hopefully I don't have to go that far .. currently I am doubting its a CPU board problem only because jumping the row 1 pin to the relevant column pin registers as a switch closure

 

@ Robm , yeah , I checked continuity from the plug at the board (pin 1 on J209 connector) to the last switch in the row and in fact to each of the 3 switches .. will try the wiggling suggestion but have reflowed connections on each switch when replacing dioides

 

@ Firepower .. yeah I'd better double check the orientation of each diode I guess , I would have thought that if one was wrong then I would have had a switch error for that switch long before now ?

 

No opto's in either row or column .

 

Will try your last suggestion too .. anything is worth a try

 

Cheers

Dave

Update please Dave, Are you getting any closer, Diodes??.

Update please Dave, Are you getting any closer, Diodes??.

 

 

Gday Jeff.. afraid you are all probably going to have to wait a while as we're going away for a month so doubt I'll be able to get back to this before then .

Gday Jeff.. afraid you are all probably going to have to wait a while as we're going away for a month so doubt I'll be able to get back to this before then .

Okey Dokey Dave, forgot about that:redface. Enjoy mate and have a great time:023:

Still trying to find the time to get back to this .

I did check the diodes are all the correct wway around last weekend when I was installing the Snack bar mod .

Also noticed that switch 41 error has disappeared but the other 2 switch errors are still there .

Also have a new (probably unrelated ) intermittent issue of 'pinball missing' when trying to start a game when all 3 balls are in the trough .

If it's not the diodes then it's time to consider the Switch matrix i.c, the ic while faulty will act normally under switch edge & switch fault tests. This i.c has a history of the faults you describe. Don't be too quick to discount the replacement just because it means removal & a replacement.

Gday all .. still looking for suggestions on this issue ..

Switch 51 and 61 not registering closed .. switch 41 has mysteriously come good (oh oh)

 

Also have a new more annoying fault which has made the machine unplayable !!

 

Right trough switch is registering open when another switch on the same column is closed (eg the other 2 trough switches or the outhole switch etc) .. this fault means you can't start a game because the CPU thinks a ball is missing . When all other switches are open in the same column this switch registers as closed (when there is a ball in that trough position), but as soon as another switch is closed in that column then this switch continuously alternates between open and closed (this is in single switch test)

 

I have replaced the switch because I had a spare and the original (Normally open contact) was not measuring consistently when operated.

 

I have also checked continutity is OK on both the row and column wires between this switch and the first switch in the row and column.

 

I have started to read up on fault finding WPC switch matrix issues but any help appreciated .

I still suggest to you #23 post.

I still suggest to you #23 post.

 

Sorry , missed that post .. thanks and will be considering that but have not done enough testing yet to prove U20 is faulty ,as I seem to have a row fault as well as a column fault ?? or maybe have two columns shorted together ?

As you probably know , U20 controls columns .. all this started after my daughter managed to sink 2 balls into the snack bar hole at the same time and they got jammed there .. so wondering if I have a short somewhere in the wiring loom as a consequence of that .

 

Done more reading on troubleshooting switch matrix faults since my last post which has given me some more testing ideas