Those wouldn't matter.

LTG : )

Original small boards and eddy sensor in your game ?

If so did you try reseating all the small connectors a couple times ? If that cures it, consider removing the connectors and soldering the wires directly to the board.

LTG : )

I test them with a logic probe / pulser. The parts on this board that aren't associated with the actual detection portion rarely fail; if I had to take a stab in the dark on whats wrong, I would say your that either you column pulse isn't getting to the board, or the row pulse isn't getting to the CPU.

The easiest way to check them is with a logic probe,
1) are you getting your low column pulse in?
2) is the row being pulled low when the column goes low?

My testing setup, just cause I like posting pictures. Not the same sensor, but the circuitry is similar enough. When the LED is lit, the row output should go low whenever the column in is low.
IMG_2280b (resized).jpg

Yep, just connect it to 12V and probe the column pulse on U2-4, D2, or J1-3. The red and black mini-grabber probes in the picture above is how I typically pull power for my logic probe. I just connect the mini grabber probes to my meter and across the input power capacitor then connect the logic probe to the meter. These are similar to the ones I am using: https://www.amazon.com/gp/product/B0002JJU3C/ref=od_aui_detailpages00

If you do find that the column pulses low and the row at U2-5 or J1-4 never goes low (assuming all other row6 switches are open), then the optocoupler is probably bad; but just like LTG, I'm still thinking its a connection issue.

Try switching your probe from TTL to CMOS and retest.

Typically you will use the TTL setting for 5 volt circuits and CMOS settings for circuits that operate at 5-15V, since the switch matrix is a 12V circuit you need to set your probe to CMOS in order to obtain the proper results. I personally forget which is which, so I use an auto detecting logic probe that automatically adjusts based on the voltage level you are using to power the probe

EDITED: 5:09PM CDT

Perform a resistance check between J1-4 and U2-5 as there is no reason for these points to have different values, for all practical purposes, they are the same point-of-test. If there is an error in this schematic, I will be unable to assist further.
pasted_image (resized).png

EDIT: According to your results checking J1-3, you have no column pulse, and show no changing value at U2-4, according to this information, you are missing your low going column pulse into the board. Without this pulse it doesnt matter what this board does, the row will remain high.

Edited to include J1 numbering

pasted_image (resized).png

no worries

Hmmm...there is some things that are odd here, specifically J1-3 which is your incoming column 2 pulse, this should be high most of the time and pulse low, if this doesn't happen the circuit cant operate properly. Are you comfortable momentary hitting the anode of D2 (U2 side) to ground? this would simulate a low going column2 pulse and if your board is working will either cause a closure of switch 26 or possible a ground short of row 6. If nothing happens U2 is likely bad. If you see the state change, you have a broken column 2 connection to your board. All of this is assuming that switch 26 is the only switch not working.

It shouldn't matter, J1-3 is your column pulse from the CPU, it shouldn't be effected by the state of the sensor.

Your blocking diode D2 is there to prevent the sensor from interfering with this pulse, similar to how the diodes across switches work.

This should have driven row 6 low IF the sensor was detecting an object and the light was on.

This is a grey area as these IC's are not designed to operate with an open input, if you had an open at J-3 I cant say with any certainty what you will read at that point.

I do have other ideas, if you put a ground potential on U-2, this could carry through the connection at J1 and set row 6 to ground, you should see this as a ground short on row6, alternatively you can measure the resistance between U2-6 and J208-7 or J209-7 on the MPU board and you should have nearly no resistance as they should be connected. If you read more than a couple ohms here, you have a break in the connection.

You can do the same for your column pulse wiring, you should have almost no resistance between diode D2 (J1 side) and its column 2 origin point on the MPU at J206-2 or 207-2 as electrically these are the same connection. If you read more than a couple ohms here, you have a break in the connection.

If both of these check out you have a bad U2. This is a very cheap part and replaceable; however it requires the right equipment and high soldering proficiency to replace without damaging the board.

pasted_image (resized).png

I did.

Maybe U2 is bad, maybe I'm missing something but the thing that's nagging at me is earlier I had you ground U2-5, if you have a good connection back to the MPU this would drive all switches in row 6 low, these ALL should have registered as closed, but you didn't see any switch closures or ground fault; this tells me that no matter what state U2 is in, its not going to be able to close switch 26. I must be missing something, but I'm not sure what.

If you have a 4N35 laying around go ahead and replace it.

If you have a chance, try your logic probe on J206 (12V logic) this is the same signal you should have seen on J1-3 ... it also bothers me that you didnt see a negative going pulse there.

Not sure why I didnt think of this earlier, but if you have a diode lying around you can always use it to jumper your Column 2 pulse onto row 6 by removing your J1 cable and sticking the diode between pins 3 and 4 (Green-Red) (white-Blue) wires with the striped side in pin 3. If this action doesn't cause a switch closure, nothing you do to the sensor board will matter.

Just be careful not to stick it in the wrong pins or you can short your 12V to ground.

Oh ... I never had you do this.

I think you have checked everything there is to check at this point. It looks like U2 will need to be replaced.

I do board work so I have a decent stockpile of parts, some of the most common high fail items that are good to have on hand are:

Micro switches
1n4004 diodes (these rarely fail but are throw away items when replacing a switch)
TIP102
TIP107
TIP36C
Bridge rectifier KBPC3504W
LM339 IC
ULN2803 (U20 on CPU)

I get most of my stuff from Mouser Electronics.

I doubt your CPU is the issue in this case, if you want to confirm this, disconnect J206/207, J208/209, and J212 then jumper J206-2 to J208-7; if the switch reads closed, then the CPU isn't the issue.

http://www.pinwiki.com/wiki/index.php?title=Williams_WPC#Switch_Matrix_Problems

You can accomplish the same thing with a wire...This is essentially the same test you performed on your processor board, only at the power/signal plug coming into the eddy board. The diode only prevents low going pulses (switch closures) on the row 6 from reactivating column 2; this wont be an issue during testing, but if you see phantom switch closures on row 6 or column 2 while testing, this would be why.

Your voltage to the board is fine, I cant see any other possible issue besides a missing output from your 4N35 ... I can think of nothing else that would cause this issue. I have to admit I'm a little set back by the replacement board not working.

On start up I get Switch 26 Eddy sensor error. However in game I believe Powerball is working correctly. I have a replament board for this installed. Could game being seeing an error but game still working right?