I'd get some help. You are losing ground on your attempts so far.

LTG : )

Problem on the driver board.

LTG : )

Separate the CPU board from the Power board by using the tests. Use the switch test to check the cabinet opto switches. Use the solenoid (flipper) test to check the power board drive transistors.

Post a picture of the switch matrix state. You can get the state from T1 (switch edges) or T2 (switch levels).

Stick to T1. It doesn't cycle. I only want to see the state. Also ... don't touch any switches.

General warning to readers: do not disconnect or reconnect any connector with the power on unless you know what you are doing.

Now disconnect (pull) J208 and J212. That would the top left connector and the top right connector. Ordinarily I would always suggest turning off the power but for these connectors it is safe to pull them with the power on. If you don't feel comfortable then turn the power off before disconnecting them. You will need to re-enter the switch tests. Once you've pulled them post a picture of the switch matrix again.

Great. Your button problem is on the CPU board. There's no playfield input (you disconnected it) but the switches still register as closed.

F3 = Lower left flipper EOS (LLEOS)
F4 = Lower left flipper opto (LLFL)
F7 = Upper left flipper EOS (ULEOS)

They're actually not related. The EOS is playfield and the opto is cabinet.

01_flipper_direct_switches.jpg

The schematic. You should use a logic probe to determine where the signals are breaking down. If you don't want to or are unable to use a logic probe and you want to shotgun then the problematic ICs are indicated. Don't swap them all at once. Swap them one by one to reduce potential variables of change. Note also that Williams schematics are known to have errors so you should double check the schematic to board continuity. Odds are the schematic is correct but I have seen errors in them.

02_flipper_input_schematics.jpg

This is a tough call if you're looking at shotgun replacing without doing the actual electrical investigation. I strongly suggest you do electrical investigation to reduce potential damage to the board. It's your board so you get to choose what you want to do.

• The most logical thing is a single point of failure. This would lead to U7.
• It is also possible that some form of high voltage short can take out multiple components and cause multiple points of failure at the initial contact with the high voltage but does not progress further upstream.

The only way to distinguish between those two possibilities is to investigate electrically.

If this were my board (and I am proceeding without doing any electrical investigation - i.e. shotgunning) I would cut out the LM399s by the legs, remove the body and then remove each leg individually. Install a socket and then replacements LM339s. See what happens. If the problem still persists then it is most likely to be U7. Again I would cut the legs and install the socket, etc.

As I said earlier ... it's your board and you know your skill and experience. It's for you to decide how you want to proceed.

Swap the IC out before doing any work on the board. Always suspect prior work. I know that's sad to say but it's unfortunately more often than not correct in my experience. If the fix is easy as swapping the IC then you've saved yourself potential headaches. You can order SN74HCT244N or CD74HCT244E. The SN74HCT244N is generally cheaper.