There's probably an open trace from the repair around U16.

It looks like the trace running from the top of resistor R137 that runs under the middle of U16 to pin 3 of U15 might be damaged too.

Can you post a picture from the back of the board around U16 and U15 so we can see what's going on there with the repair?

Dear o dear..

Question is are there any short circuits in that mess. The flux/resin makes it hard to tell. And we don't know what's going on under the U15 and U16 sockets.

Did you check this according to the schematics, or just that the soldered wires are making contact?

U16 is the only chip there that starts and runs the clock circuit, U15 is secondary and is not required for U16 to run. Can you remove U15 (incase it's putting some load on U16) and see if U16 starts pulsing.

The holes where the chips go through the board have copper connection between top and bottom around the edge of the holes - they connect traces between the top and bottom of the board. More than likely, one or more have been ripped out when the old chip was removed and connectivity wasn't reinstated.

I'm afraid you're going to have to check all connections to/from U16 according to the schematic. Also pay attention to any accidental solder splashes that might be shorting adjacent pins/traces.

A skilled solderer can repair this and make it neat but it can only be done with a proper desoldering station.

Yes, absolutely, they should be the same voltage since they should be connected together. You might have found the issue.

Connection is external.
I just checked one of my MPU boards. The trace between pin 7 and pin 12 is on the top side of the board underneath the U16 chip itself.
Note, pin 7 connects to pin 12 of U16 and this trace also connect to pins 9 and 10 of U15.

Do your best to try and solder a wire between pin 7 and 12. If pin 12 is too tricky, solder pin 7 of U16 to pins 9/10 of U15 instead.

Progress...

When you say the LED flickers, is it a very quick flicker the moment you power on, or is it a half second flash 1/2 a second after power on?

Presuming it's the former, then the board is having an issue testing the ROMs.

It's clear the ROM sockets have been replaced at some point.

Can you remove the U1, U2 and U6 ROMs and post some top side pictures of the sockets. Looking for any trace damage.

Oh the sockets weren't replaced, someone just resoldered them. On the one hand it's good because the board hasn't been further damaged, on the other hand these old sockets after 40 years are probably no good and need replacing.

Anyway, for now put the ROMs back in.
Test continuity of the pins between the ROMs. i.e. pin 1 of U1 should connect to pin 1 of U2 which should also connect to pin 1 of U6.

The only pins which will not connect to each other are pins 18, 20 and 21.

Yes these are correct, sorry.
BTW if I wasn't clear, the continuity check needed to be done from actual ROM pin to ROM pin, not on the solder side - i.e. testing continuity of the ROM pins through the sockets.

Are you testing this MPU board on the bench hooked up to a power supply or testing in the machine?
Remove the 3 ROM chips again, connect your logic probe and power on the board.
If the tip of your logic probe will fit in the ROM sockets, check that you have high/low/pulse activity on all pins except pin 12 which should be low and pin 24 which should be high.
If you can't fit the tip of the probe in the sockets, touch the probe on the unused socket points where U5 is located - note U5 location will have no activity on pin 21.

No worries.
To explain what we're doing: when the board powers up, the program starts by testing the three ROM chips and if it finds any issue it essentially stalls. When it stalls it reduces activity to the ROMs and limits what we can see is happening.

By removing the ROMs, there is no valid program to run at power on so the board runs off like a freight train providing activity on the address and data lines - this is useful to see that there is activity at all the pins on the ROM sockets and around the board.

Ok, sorry pin 19 of U5 also isn't connected to anything just like pin 21 and they both gave the same reading. However you also got the same reading on pin 13 which I expect should have behaved like pins 9, 10 and 11 - i.e. bright green and pulse light on.

With your multimeter, can you check if you've got continuity from pin 13 of the U5 location to pin 30 of the CPU chip at U9? This is a data line signal (D3).

Can you install your original MPU board in the machine without any ROMs and perform the same checks with your logic probe at the U5 location? Then we can compare results.

So on your original board:
Pin 13 at U5 is behaving normally.
Pin 18 with no lights is showing incorrect but I'll put it down to the probe not touching the pin properly.

Can you install the U6 ROM (leave U1 and U2 out) and probe pins 9 - 11 and 13 - 17 of U6.
Do the same with the replacement board.
We're now looking at the failed ROM stall condition.
Let us know the results.

Hmm, this is going to be tricky and made a little harder by the fact your logic probe has no "low" indicator. Your green "level" indicator only tells you when a signal is high. Without the low indicator you have no idea if a signal is floating open circuit/midrange voltage or if it is a logic low (zero volts).

See here: Logic probes typically have Low, Hi, and Pulse indicators.
https://pinside.com/pinball/forum/topic/terrybs-guide-to-logic-probes

BTW, sorry I should have told you from the beginning to set your logic probe to TTL, not CMOS. Some results will probably appear a little different. Essentially what we were looking for above was activity on those signals to the ROM bus to make sure there weren't any open circuits.

Let's go back and try the simple stuff. Install the U1, U2 and U6 ROMs into the replacement board. Maybe the action of removing/installing the ROMs will have swiped the internals of the sockets to give better connection and try it.
If no luck, then remove U7, U8 and U10 - they aren't needed to get past the ROM test.
If no luck, swap the U9 CPU chips between the boards.
If no luck, remove U11 and install the U10 chip you previously removed into the U11 socket (U10 and U11 are the same type of chip). We need U11 populated because it drives the MPU LED and gives us feedback of what's going on.

BTW2, what's the picture in the previous post about?

Agreed. If you can, try and solder the fracture first before reinstalling it. If you can't, you can't.
If the pin does break, then you'll have to get replacements ROMs, which we can then revisit trying to repair the replacement board.
It is possible that the U15 "repair" still has a missing trace that could be causing the ROMs to fail particularly around U15 pins 1,2,3,4,5,6. As a last quick check, make sure you are getting pulse activity on pins 3 and 37 of U9 with your logic probe. These come from clock output pins 8 and 11 respectively of U15.

Check that pin 6 and pin 3 of U15 are pulsing and also that pin 13 of U7 indicates the same.

Ok, so flashing "Level" LED indicates the probe is sensing "nothing". If you put the probe tip on TP4 of the MPU board (top right corner) which is ground, that will give you an indication of how your logic probe sees a low level state (i.e. zero volts) - the pulse LED should be off and I presume the Level LED will be off too.

Ok, that's it for now. Good luck getting U2 back in the old board in one piece.