I apologize for the newbie post. Our TOTAN that we've had for about 2.5 years has a ground short on row 4. It happened in the middle of playing a game. The game just rebooted in the middle of a game and came up with this fault.
We've figured out our previous machine issues, but I don't seen to be making progress with this one despite searching around online.
The ground fault does go away if I remove connector J212 on the CPU board or J3 of the coin door board.
I have checked the connections to the switches listed in the switch matrix for row 4 and I don't see anything touching ground. The exception being the ball through, which doesn't seem to have a white/yellow wire.
I've played with the multimeter, but I don't know that I've done anything terribly useful with it.
I guess I am hoping for a next step or two to further narrow down the issue. Am I right that a row 4 ground short means that the white/yellow wire or one of the switches it connects to is grounding somewhere? If I could at least be sure about that, it would be helpful so I can concentrate exclusively on those wires and switches.

Thanks for any help.

Disconnecting the plum bob doesn't make the ground short go away, nor does disconnecting any of the connectors on the coin door board, except for J3. I took the coin door board out and inspected it for damage, or burns, but it looks fine.
By "unhooking" the switches you mean de-solder them, right?

Thanks for the responses.
So I tried disconnecting the coin door again, and it doesn't change anything.
I tried closing the coin door, and yes, I suddenly have a ground short on row 2 as well.

Thank you Zaza. When jumping J206-2 to J209-1 I get the whole row 1 grounded. I will buy a few U20 chips. Thanks for the help!

New chip arrived today. I replaced it and there is no change. All switches are activating a whole row in the T1 test. Seems like one switch grounded would not cause this. Is this a board issue?

Mini oscilloscope should be delivered tomorrow so I can check it out. Thanks!

I didn't realize the circuits were high voltage. The oscilloscope I got is low voltage. The readings were off the screen.
I put together a crude voltage divider and was able to get readable input. The chips were giving 50v, but the image below shows much lower voltage because of the voltage divider. The U23 and U20 outputs all look like the attached image. Just for camparison I tried the house 110v AC and got a similar dirty looking waveform. I tried the DC calibration point on the oscilloscope and got crisp square waveforms. I'm not sure what all this means.

20170722_221817 (resized).jpg

I replaced U20 and U23. Switch matrix tested normally. Everything seemed fine. Started a game, and within a second of shooting the ball it reset. Then when it came back displayed:
Check fuses F101 and F109, J127 and opto 12V supply
I couldn't find any issues. Beat my head on it for a while. Took things apart. Put them back together, still had error. Decided to start a game and manually trigger playfield switches to see if one in particular blows things up and nothing broke. Played a whole game just fine. And the error disappeared. Seems fine, for now.