What's new
Pinball info

Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Strange Switch issue - Fish Tales

MadNat

MadNat

Site Supporter
Joined
May 11, 2019
Messages
1,054
Location
Warwickshire UK
Hi folks. Strange problem with my Fish Tales. In Switch Tests the bottom 2 rows have a high speed switch cycle. They all switch on and off, simultaneously across both rows. Looks like faster than the test is cycling.

Im guessing this is not a TIP issue - perhaps a diode or something worth checking out on the board?

Any help very much appreciated.
 
Remove your switch matrix connectors from the cpu board and see if the fault goes or stays.

and Check for battery acid damage on the cpu under the battery holder.

report back
 
myPinballs said:
Remove your switch matrix connectors from the cpu board and see if the fault goes or stays.

and Check for battery acid damage on the cpu under the battery holder.

report back
Click to expand...
Ok thanks Jim - feedback.

With both switch connectors (row/column) removed - no error. With switch connectors connected - error.
 
i had a similar error on a different machine. it turned out a stand up target switch had gone faulty it was shorted to be always on but it had done this in a way as to exclude the diode and this gave really weird switch signals especially on an opto switch in the same row. the opto switch just buzzed on and off very fast when in switch test mode. maybe gives a clue?


Sent from my iPhone using Tapatalk
 
MadNat said:
Ok thanks Jim - feedback.

With both switch connectors (row/column) removed - no error. With switch connectors connected - error.
Click to expand...

Ok so we know that your cpu board is likely fine and the problem is on the playfield somewhere. Next step is to check the switches in the rows in question on the playfield 1 at a time. If you work out the order they are in on the playfield for each row then it makes it easier to disconnect the row wire temporally to isolate the switches forward from that point.

Also if you have optos in the row it’s easy to disconnect those first to rule out and also the most likely is a standup or slingshot switch where the row and column tabs are shorted somehow

Good luck
 
myPinballs said:
Ok so we know that your cpu board is likely fine and the problem is on the playfield somewhere. Next step is to check the switches in the rows in question on the playfield 1 at a time. If you work out the order they are in on the playfield for each row then it makes it easier to disconnect the row wire temporally to isolate the switches forward from that point.

Also if you have optos in the row it’s easy to disconnect those first to rule out and also the most likely is a standup or slingshot switch where the row and column tabs are shorted somehow

Good luck
Click to expand...
Thanks - for the pointer. The fault stops when I disconnect the opto's from the board. So guessing thats when my problem is?

If it is here - my next question is how to determine which is causing the problem - the opto's or the board? The board looks a little cooked in parts:
A-13901-1.jpg

The opto's are here (could do with a clean)...
IMG_0199.jpgIMG_2395.jpg

I have some test kit (scope and a nice multimeter) - but would need help on what to test. Thanks in advance.
 
Last edited:
Does it stop if the optos themselves are disconnected but the opto controller board still plugged into the matrix, or only when the controller board is unplugged aswell?
 
Removing the connectors from the bottom of the cpu is not a definitive test that the board is ok. You need to be in switch test then disconnect all the connectors for the switches. Then you need to short each row to each column in turn to test each row and column without external connections. For the cpu to be confused it needs at least one input that will permeate the confusion.
 
myPinballs said:
Does it stop if the optos themselves are disconnected but the opto controller board still plugged into the matrix, or only when the controller board is unplugged aswell?
Click to expand...

Sorry - I wasnt very clear was I? When I disconnect just the opto connectors (i.e. 1 connector is in place - left hand connector J1), with J2 & J3 disconnected - the other switches work ok.
 
Last edited:
AlanJ said:
Are all the other switches working correctly with the optics disconnected?


Sent from my iPhone using Tapatalk
Click to expand...
Hi - yes, it seems that the other switches work as expected (under switch test) - when the opto's are disconnected.
 
MadNat said:
Sorry - I wasnt very clear was I? When I disconnect just the opto connectors (i.e. 1 connector is in place - left hand connector J1), J2 & J3 disconnecte - the other switches work ok with this setup..
Click to expand...

So try plugging in 1 connector at a time (J2 and J3) to see what happens. For reference changing all the optos and receivers is a good idea anyway for relialbity. Not expensive or hard to do.

Also, even though we seemed to have isolated the issue to bad optos, as Andy just mentioned, i should have told you to do a separate cpu switch row & column check, before moving onto the playfield check just to make sure a fault isnt also still on the cpu board. I forgot to mention this sorry.
 
pinballmania said:
Removing the connectors from the bottom of the cpu is not a definitive test that the board is ok. You need to be in switch test then disconnect all the connectors for the switches. Then you need to short each row to each column in turn to test each row and column without external connections. For the cpu to be confused it needs at least one input that will permeate the confusion.
Click to expand...

Will do this test this evening Andy. Just for clarity - this is to be done at CPU board level (J206 - Column & J209 - Row)? Are you advising I enter switch test and systamatically connect row 1 with column 1-8, then move onto row 2 and do the same, etc through to row 8 - column 8? Assume this is a simple jumper between row and column or do I need a diode between pins? Sorry for the basic question - don't want to make more damage!

Cheers
 
MadNat said:
Little vid showing exactly what happens when Opto's prugged into the board. Hope it helps
Click to expand...

My view is that its most likely a fault on the opto controller board itself if both optos sets are causing it. Its pretty unlikely for both pairs of opto sets to go bad, unlesss you'd not noticed this before and they've both failed over multiple years. If i was going to guess then i'd say the lm339 on the controller board is shagged, but its a guess as i dont have the board in front of me.
 
myPinballs said:
My view is that its most likely a fault on the opto controller board itself if both optos sets are causing it. Its pretty unlikely for both pairs of opto sets to go bad, unlesss you'd not noticed this before and they've both failed over multiple years. If i was going to guess then i'd say the lm339 on the controller board is shagged, but its a guess as i dont have the board in front of me.
Click to expand...
Ok - did the switch test on the CPU board - all was as expected - position 24 closed (manual says its always closed), the others all responded as expected - got the occassional shock when a test caused a solenoid to fire!!

While I was there - I reconnected the opto's, and checked the voltage on the transmitter side (white) as suggested by pinwiki (thanks @Moonraker ) - mine only read just over 1.3VDC - not 12VDC as the PinWiki suggests (https://www.pinwiki.com/wiki/index.php?title=Williams_WPC#Opto_Switches). Could a faulty LM339 cause this?

Thinking I should rebuild this board regardless - so popping a new LM339 in wont break the bank.

The pinwiki also outlines tests for LM339 - but not for the board I have (A-13901-1) - so not sure if that test will be appropriate for me.
 
The wiki is incorrect.

The opto transmitter is indeed supplied by 12v but you’ll only measure 12v across the opto itself if the opto component is faulty ( open circuit ). When the opto is good, then the correct voltage is about 1.3v. The rest of the 12v can be measured across the associated serial resistor if you’re interested.
 
Fishy swims again!

LM339 swapped and optos renewer. Thanks Andy for the swift service.

Slight panic when it didn’t work after fitting - all switches down from the catapult switch (which is reel optos and right exit ramp). Saw that the column wire came loose at the catapult switch. Perhaps the dodgy connection there was my problem all along - who knows. But all good now.

Thanks very much for your help all.
 
MadNat said:
Fishy swims again!

LM339 swapped and optos renewer. Thanks Andy for the swift service.

Slight panic when it didn’t work after fitting - all switches down from the catapult switch (which is reel optos and right exit ramp). Saw that the column wire came loose at the catapult switch. Perhaps the dodgy connection there was my problem all along - who knows. But all good now.

Thanks very much for your help all.
Click to expand...

Glad its fixed and we narrowed it down to the lm339. Pretty common fault so good for people to see the process we went through here.
 
You must log in or register to reply here.
Back
Top Bottom