What's new
By:
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!

Switch Matrix issue.

T

tiltedjohn

Registered
Joined
Jan 3, 2020
Messages
37
Location
Banstead, Surrey
I have a very annoying issue in the switch matrix on one of my machines I would appreciate any advice / help.
When entering test mode I get an error message to check Switch 25 which is in row 5 column 2
When I activate this switch it triggers switch 45 (Row 5 column 4) switch 25 does not trigger at all!
If I check the colours of the cables connected to both switches they are correct.
I then disconnect the row and column plugs from the CPU and manually test switch 25 and 45 by shorting the appropriate pins on J207 & 209 both trigger correctly so I know it is not the CPU!
Next I reconnect J207 & 209 and try the same test shorting the appropriate pins to trigger Sw 25 but it show as Sw 45.
Clearly it is a wiring issue but where? I have manually run new cables to switch 25 but it still shows as 45! The diodes are fine and althouth 25 & 45 shre the same row, not other switch in that row is affected.
This is drivin me nuts I guess it has to be a wiring loom fault but I do not know where to start and how has it just started to happen (or maybe I never noticed it before)
 
When you say the diodes are fine did you take them out of the circuit to test them?

Are switch 25 and 45 diodes the correct way round on each switch (can you see the band)?

Is there a switch 15 and / or 55 ? They may need testing as well...
 
Pick Holder said:
What machine is it ?
Click to expand...
HS 2 25 (Left outlane) 45 (right bank middle) I have done more checks and with the connectors in place if I short various columns & rows there seem to be others that do the same . I am ordering some ULN2803 Darlinton Arrays just to be sure. They are cheap enough from Farnell (£1.55) What I dont understand is why it has suddenly happened?
 
I do not think its a ULN2803A causing this isssue. More likely to be wiring, diodes or switch wiring.

Make sure the diodes are on the correct spades of the switch and post importantly - check the direction (ie where the band is).
 
Moonraker said:
When you say the diodes are fine did you take them out of the circuit to test them?

Are switch 25 and 45 diodes the correct way round on each switch (can you see the band)?

Is there a switch 15 and / or 55 ? They may need testing as well...
Click to expand...
Diodes checked in circuit but they both check out and read the same as every other switch on the machine. They have not been touched so must be on correct polarity
Pick Holder said:
I do not think its a ULN2803A causing this isssue. More likely to be wiring, diodes or switch wiring.

Make sure the diodes are on the correct spades of the switch and post importantly - check the direction (ie where the band is).
Click to expand...
Everything was fine and I have not touched them all diodes check out for both forward and reverse polarity I have not touched the wiring so am not sure how it can be anything but the Darlington IC. I will get them tomorrow so will report back then Luckily the 2803 in socketted so is easy to change
 
In your first post you said the cpu was working correctly so no need for 2803.

Blown diodes can test correctly when left in circuit hence why I asked.

Put the game in switch test mode and use the matrix diagram to establish what exactly is happening.
 
Moonraker said:
In your first post you said the cpu was working correctly so no need for 2803.

Blown diodes can test correctly when left in circuit hence why I asked.

Put the game in switch test mode and use the matrix diagram to establish what exactly is happening.
Click to expand...
When I first did test with plugs removed from CPU They seemed to test OK However I tested again with plugs in place and several coloumn and rows seemed "connected". I think I an going to trace the wires back from the connectors as somewhere rows 2 & 4 must be connected. How can this be a diode? I dont understand but I equally dont understand how this can have suddenly happened without anyone touching the wiring, CPU or diodes. It can only be a faulty diode or CPU IC and I agree that the CPU is unlikely. How then would a "duff" diode cause columns 2 & 4 to short and if this was the case, why is it not happening to all switches on columns 2 & 4.?
If I am sounding ignorant, please forgive me but I am trying to understand the problem as well as fixing it
 
Last edited:
AlanJ said:
do the other switches on row 5 check out ok on the switch test?


Sent from my iPhone using Tapatalk
Click to expand...
Yes they do. It does not help that one is an "Opto" but I have found an excellent video on You Tube that really has helped me understand how the Diodes affect a Switch Matrix. and would explain how this has happened...
I have a whole bunch of switch Diodes so I guess I will start the hunt for the shorted one. I will let you all know how I get on as it may help others in the future I am beginning to feel stupid but I guess that is how we learn.
 
ok so if all other row 5 working fine, that would suggest its one of the diodes on a switch in col 2 or col 4. but probably not on row 5.
 
You need to test the CPU without connectors in place, you did and it was fine hence CPU ok and no need for 2803.

The fault is playfield side, so something shorted or faulty or possibly corroded MPU connectors. Thus the need to use the switch test to establish what the fault is exactly so that you can narrow down on where to look.
 
All Sorted! After watching the You Tube video and checking the "rectangle" of the switch matrix I discovered switch 85 had a shorted diode replaced and all is fine again. Thank you guys for all your help.
 
You must log in or register to reply here.
Back
Top Bottom