Have you checked the high voltage with a meter to see what it's reading? If the new power supply follows the original with test point identification, it'd be TP2 you'd want to check on DC voltage. Exercise caution of course, since it's high voltage. And on that note, unplug the displays to avoid any additional damage to them while you're testing the power supply.

TP10 = ground (bottom left corner)
TP1 = +5 volts DC (top right corner)
TP2 = 170-190 volts DC (high voltage out, middle of board, adjustable via trim pot VR1)
TP3 = 230 volts DC (high voltage in, middle of board)
TP4 = 43 volts DC (solenoid voltage, top left).
TP5 = 14 volts DC (main logic power before regulation, middle of board)
TP6 & TP7 = 11 volts AC (lamp matrix)
TP8 & TP9 = 6.3 volts AC (General illumination)

Source: http://www.pinrepair.com/6803/#displays

It's not sounding too good with a crackle pop and dead displays AFTER they were working. Did they go super nova (light up almost pink) and then disappear? I don't see really dark segments, so they *could* be okay.. but it really seems to indicate there was a short somewhere or higher voltage into a component. You could pull one or both displays to see if you can find out what went crackle/pop. Usually if you hear it, you can SEE it too.. burnt mark on the board, cracked transistor or an IC that has a hole burnt through the middle.

Hey that's always a plus! Interesting, wonder what the crackling was. Any fuses blown on the power supply board or do those all look fine? With the game off you may need to check with continuity test across the fuse with a meter. Sometimes it's hard to tell depending on how they blow, especially the HV fuses which are often thin wire.

Of course if you've checked voltages at the test points you'll probably know if a fuse has blown as well.

Sometimes crackling can be a component or connector heating up but not actually popping.

Did you try swapping the original RAM back in? Easy enough to do just to eliminate that from the equation. On that note, is the NVRAM seated fully? Sometimes I've had people plug mine in and not insert it fully into the socket. You may have to remove the MPU from the machine and set it on a flat surface so it's not flexing the board too much. Just mentioning this since a bad connection there could cause some weird things to happen, since it's the memory for the machine.

Is the game booting into attract mode otherwise? From your initial posts it sounded like the game was working fine for a bit and then the displays went.. but then you mention the locked on flipper happening. So stepping back a bit, was the game working with the old power supply, displays and all? And did the displays then go out with the old supply, followed by purchase of the new power supply + nvram. Or did this all occur after installing the new power supply? Just trying to understand sequence of events.

Usually the best thing to do with these games, especially if in unknown condition, is to isolate things and work your way from the transformer/power supply forward. So you'd start with the transformer input going to the power supply, no other connections (or minimal to get some proper voltages there). Check voltages at the power supply. Then plug in the next thing. At the MPU you could leave some of the lamp/solenoid harnesses unplugged, just have the harness supplying voltages/grounds going to the MPU.. that prevents solenoids or lamps from locking on if there's MPU issues -- or blowing transistors if there's a coil problem. Doing it this way, you aren't throwing a new power supply into an unknown game and hoping for the best. It does take a while to step through things in this way, but then you know the last step you took if you do have an issue as you continue adding new boards or harnesses into the mix.

Definitely don't want to test voltages while leaving a flipper locked on..

First you need a working game/MPU that's not locking on a flipper coil, is getting 9 flashes, boots into attract mode and plays a game blindly. The suggestion of checking display voltages at TP2 on the power supply was assuming the rest of the game was working and should still be done at some point.. but that flipper locking on immediately is no good.

Sorry I think I'm a bit confused where the game is at or the sequence of events.

Did it go..

1. Game works with old power supply.
2. Displays die.
3. New power supply ordered & installed as possible display fix. NVRAM also installed.
4. Game now locks on flipper coil and may be doing other odd stuff.

Or did the display issues occur after the new power supply was installed?

That's a really good video, nice job explaining everything you've got going on and what steps you took so far!

I didn't realize you had an orange glow at the displays. I'm gonna take a quick look at the schematics to see if anything jumps out.

Looks like segment data for the displays comes out of the U7 6821 PIA on the MPU. There's also momentary solenoid, continuous solenoid, solenoid bank select. If you had a logic probe you could try probing pins 2-9 on U7. I'd imagine that chip to see if there's activity while the game's in attract mode. Those U7 pins seem to feed the data to the 74HC373 chip that's on each display.

The flipper thing is odd.. which also has me wondering if the U7 PIA is to blame. It would kind of make sense if that chip was dead, why the displays weren't lighting and why you can activate a flipper since it looks like PB16 (pin16 of U7 on the schematic) is involved with the flipper enable relay.

If you have spare good PIA chips, throwing one in there might not hurt. I'm not sure what swapping the U7 and U8 PIA might do, so don't know if I'd advise that.. on the older Bally games you could swap PIAs and if it created different issues you had a pretty good idea what the cause was. You could try it, but results may vary

Just FYI there's a handy little tester at http://www.neoloch.com called the "Inquisitor".. that has both RAM testing blades and also a PIA testing blade. If you're fixing up a lot of older pinball machines (90s on back), most of them use those PIA chips and having a way to test them easily is nice.

I keep going back to the thought of U7 being involved here at this point... there's a chance it's +5v not making it to the display, but with a glow at both of those displays & the same issue taking them out, it seems like it's further up the chain as I'm sure you're suspecting as well. So if it's display data related, I'd say it's less likely for it to be the BCD data involved in digit enable.. and more likely it's the PIA feeding the segments data to the displays. Plus that weird flipper thing happening.. and for that PIA to be involved with flipper enable..

Just a guess based on symptoms (no displays, flipper issue) and U7 being involved there.

That said, looking at the manual, the 6th flash happens after the software tests U7. Not sure how extensive the test is. It was just a thought if you had a PIA to throw in there, since U7 is in a socket it'd be an easy enough test.

As far as testing the displays further, well, there's test points on the back for 180VDC and 80VDC.. if you had alligator clips you could measure voltages there, but definitely need to be careful hooking things up on the back of those and don't want to handle the displays while you're testing them. High voltage sucks.

There's no +5v test point on the back of the displays, but that you could test with the POWER OFF to the game. You'd have to find something that 5v is feeding on the display (ie. one of the logic chips on the display) and then use continuity test on your meter to a 5v test point on the MPU to see if you get a beep.

At that point you'd rule out any voltages not getting to the displays and be left with display signals. Now if just a single segment didn't reach the display, or just a single BCD line.. you'd probably still get something out of the display. But if it's a latch/strobe or all the segment data isn't hitting the display.. then it wouldn't light up. With both displays affected, and no blown components on the back of the display, and an orange glow out of them.. and an MPU that looks pretty decent, I was just thinking U7 looked like it was involved.

This is then where a logic probe would help.. since you could probe a few things back at the MPU to see some of the low voltage signals are being generated for the display (digit enables, segment data, latch, etc).

If you haven't already, try reseating the display connector on the MPU.

I hate to throw too many things out there, but I also came across this post talking about D23/D19 going bad..
https://pinside.com/pinball/forum/topic/need-a-bally-6803-exper

D23/D19 are diodes in the flipper enable circuit. There's part of the schematic shown in one of the posts.. maybe check those diodes? They should be near the relay at the top right of the MPU I would think. With the flipper short having occurred, maybe back EMF did affect something in that flipper enable circuit and got back to U7.

There could still be something to that whole back EMF think & diodes in the flipper enable circuit. If your meter has a diode test (usually only the more expensive ones do) then you could check those with the machine off. On diode test, you should get a reading one way (maybe .6v or higher) and no reading the other way.. usually meters beep continuously if it's shorted. I'm not sure how failed diodes there would affect an otherwise working PIA.

Might be time to break out the logic probe, or get a better logic probe if you don't think yours is up to snuff. Then you can check the lower voltage signals being generated by U7 for the segments is pulsing high/low (PIA pins as noted in an earlier post). You could also check the BCD data for the digit enables is pulsing high/low.

Unfortunately seems you're at a point where you've got to go "deeper" into the technical side of things to see where a signal or voltage is being dropped. That is, if you're getting 180v and 80v at the display test points. If you aren't getting the correct voltage at those test points, that's a problem.