That bottom bumper going off when it isn't supposed to could be the spoon switch itself, adjusted too close or shorting. One test for this is to smack the playfield with an open hand to see if the switch is too close. I had a similar problem on my F-14 playfield where a target was causing phantom switch closures at other locations. There was nothing I could find wrong with the switch, even after replacing the diode - but replacing the switch fixed it. It could have been the holes for the screws shorting somewhere in the switch body to the leafs intermittently, something like that.

There are some switch matrix troubleshooting docs and youtube videos online to help identify phantom switch problems.

Get the Bally test chip from pinwiki burn it and use its switch test which will tell you all the switches that are closed. (Install, get to switch test, and press and hold the credit button.... it will scan through all the closed switches on the displays). Then you can narrow down where your issue is.

First what I think is your easy fix. Pop bumper is probably a bad/missing capacitor. The cap helps eliminate switch "bounce" so if it is bad, the pop bumper can energize more than once. Could also just be a switch adjustment.

Now for the switch issues. Switch matrix issues are tricky. First, since you did some work on the start button, make sure that it's fish paper on the switch is properly preventing the switch from being grounded. If the fish paper on the credit button isn't there, you WILL have issues with the switches.

Next, go to the switch test in the diagnostics. From there you might find a pattern of all switches on a particular row or a particular column are giving you problems.

Next, visually check the diodes on the switches, including the pop bumper. A bad diode will cause matrix issues. Note that you can also test each diode with a multimeter, but to do it right, it takes a lot of time. You disconnect one end and then in diode setting test the the diode both directions. One direction you should get null reading, the other direction .4 - .6. So I wouldn't do the multimeter test to start with. We're trying simple stuff first, so just visually inspect the diodes on the switches for damage or being disconnected.

The difficulty is in determining whether your problem is in the wiring or on a component on the board. To determine that try these steps:
Make sure game is off and turn off all MPU DIP switches. Go to switch test. With this running, disconnect MPU connectors J2 and J3 (switch row and column signals). Connect an alligator clip to J2 Pin 1 (this is column 0 of the switch matrix). Then touch the other end to J2 Pin 8-15 (these are the rows) one pin at a time. Each time, a switch number should show in the displays. It should correspond to the matrix as diagrammed in the manual. Then touch the second end to pins J3 9-16.

Once you have done that, move the first clip end from J2 pin 1 to J2 pin 2 (column 1), and repeat everything above. Repeat this for J2 1-5 (all columns). This will work its way through all columns and rows.

From the above, if a row or column is bad, check the diodes on the switches for that row or column. Also, check the connector for that row. Again, you will need to compare this to the diagram in the manual to know which switches are on which row and column.

If you still can't find the problem, it may be PAI U10 (sends switch data). To test, you can swap U10 and U11 (hopefully they are socketed) and see if this clears the problem.

Yes, that test ROM will show you ALL simultaneous closed switches. The default Bally switch test will only show the lowest number switch that's closed without telling you about other simultaneously closed switches.
https://www.pinwiki.com/wiki/index.php?title=File:Sco_testrom.zip

The 1800 points and bonus advance are from the other three switches on the same switch return line as the start button (lower pop bumper = 1000 points, dragons den capture ball = 500 points, "C" target = 300 points plus advance bonus).

BTW when you start a game the A, B, C, D, E, F lamps should all illuminate. You've mentioned a few times the E and F illuminate, this is normal. Have you changed the capacitors on the C and D targets, the pop bumpers and the dragons den capture ball target?

Disconnect J2 (playfield switch harness) from the MPU board. Does the start button still misbehave? this will tell you if the problem is on the playfield or not.

LostWorld_SwitchMatrix.png

Yowzah, time to break out the Evaporust and break down that coin door.

If the start button is not insulated from the rest of the switch you could be introducing a ground to the matrix when you press it. I would replace it.
How does the rest of the coin door look? As bad as the start button? I would carefully tear it down. clean everything, and reassemble. For the record, I would not waste time on the coin door Molex plug unless you see visible signs of damage.

Looking at the schematic for Lost World on IPDB, it looks like the lines for the credit/start switch are through the J3 connector A4J3-2 & A4J3-14 & no other switches run on the credit switch lines?

Is it different in game to this schematic with the switches that are marked in red following other lines?

Reading more of the posts and what you have done, since the game actually boots and starts correctly when you have credits except that it also scores 1800 and advances the bonus I think Quench is right in that pushing the start button seems to activate that entire row. If you are SURE that the diodes are all good and oriented correctly, I'd check to see if the wires got crossed somewhere. Physically trace the start button wires through the cabinet to make sure it goes where it is supposed to. Then check its connector (Quench has it above A4J3-2 and A4J3-14 as coin door switches on Bally went through J3 while the other switches go through J2), check the pin on the board to make sure that there isn't a solder bridge to a neighboring pin.

Curious, when you start a game, do any of the other switches work correctly that are not in the same row or column as the start button?

Caps don't matter their orientation, but the diodes do. Check to make sure they are oriented correctly. Physically trace the wire and see if you can find a place that it is being shorted (touching something metal, touching a lamp return line, etc.). Check the other switch wiring on that row also, particularly the bumper as it is easy for bumper wiring, due to its constant movement, to get messed up. Since it seems most everything is working, and you said you swapped the PIAs and the same problem, the issue has to be the wiring, the diodes, or the connectors/pins.

As for your other issue, if hitting that blue switch fires the LEFT sling, then you may have a problem with row 5 (I4). Does it also score points or award the D target?

Could you call me and we talk through this? Might be easier. Might help and might not. If so, I'll send my number and then delete the post.

Assuming those are pics of the four switches on row 6, the bumper and Dragon Den switches look good to me based on schematic. The start button hard to tell since it has different wire colors than the schematic says which isn't unusual for the coin door. As for the C Target, your wire colors don't match what I would expect. Looking at the schematic, you should have a white/yellow which you do, but then the brown/yellow the same as the others. Yours doesn't have the brown/yellow. Instead it has a brown/black. Without physically tracing it hard to know if this is on purpose or if something got miss-wired.

If the 1800 point problem still happens with MPU J2 disconnected, the problem is NOT on the playfield.
With the game off, test the diode on the start button switch (I know you said you changed the diode but lets see if there's some short there).

So long as no part of your body is touching ground, then it's ok. Better yet, make the start button switch contacts touch each other without you making skin contact. Your body can introduce unwanted noise.

The blue circled switch is just a rebound switch that scores 50 points and advances bonus.

The cabinet/front door switches are part of the same switch matrix as the playfield. Same switch strobe and return lines. The matrix switches I marked in red are in the cabinet/front door. The 32 DIP switches on the MPU board are also a part of the switch matrix, difference is they're only read once on power-up.

There's a problem with the switch strobe #0 line (which goes to the start button) somewhere that it's shorted to something. When you hold the start button, the short is causing the software to see those other switches on the same row as active - note this is a *phantom* effect, not a physical issue with those other switches.

I've got to head out but looks like you're getting help otherwise. Good luck!

With J2 and J3 connected on the MPU board:
The start button red-yellow wire is the exact same strobe signal as the white-red wire on the playfields outhole switch for example.
The start button blue-white wire is the exact same return signal as the brown-yellow wire on the playfields "C" target switch for example.

I've updated the switch matrix diagram above in post #8 to overlay the cabinet/coin door wiring details.

So technically if you jumper the outhole switch diode (banded side after the white-red wire) to the "C" targets brown-yellow wire, this will activate the start button using the same switch matrix signals on the playfield. Try this and see if you get the 1800 points or not. If 1800 points gets added, disconnect J3 from the MPU board and try again.

You've got some corrective work to do on the coin door.

LostWorld_CoinDoor.jpg

So..
When you remove J2 from the MPU board and press the start button with available credits, the game starts and you get 1800 points.
When you disconnect J3 from the MPU board and jumper a wire from the banded side of the diode on the outhole switch to the "C" targets brown-yellow wire with available credits, the game starts and you get 1800 points.

If this is all correct, it points to a problem on your MPU board because you isolated the playfield switch harness in the first test and isolated the cabinet switch harness in the second test with the same result.

The other two switch issues you mentioned in your last post are related to the same problem with the switch strobe #0 signal being shorted.

Remove both J2 and J3 from the MPU board.
With the game OFF, multi-meter set to resistance mode, black meter lead on ground, probe each pin on the MPU J2 pin header and let us know resistance each pin reads. J2 pins 1-5 and 8-15 should all measure the same around 56k ohms.

Measuring resistance/continuity with power on will give you false readings.

Post some very clear pictures of the outhole switch from side on. Also some clear high resolution pictures of the MPU board front and back.

Then move all 32 DIP switches on the MPU board to the OFF position. Does the start button still misbehave with the 1800 points?

The pics don't fully show J2 and J1 (upper 1/3 of the board).

You got a logic probe or better an oscilloscope?
If yes, with J2 and J3 disconnected, what does it indicate on the switch strobe lines pins 1 to 5 at J2?
If no, what voltages do you measure on those pins?

You need to get rid of that U10 AMI brand 6820 (known for having very high failure rate) with any other brand 6821 or at a minimum swap U10 with U11 assuming U11 is not an AMI.

Could even be the problem or part of it.

Finally we have a clue..

What activity do you see at the source of switch strobe #0 which is the PA0 signal at pin 2 of U10?

My guess is you have an open circuit from U10 pin 2 to the junction of diode CR43 and resistor R31 (which is under the red S33 activity switch). This results in R31 (which is a pull-up resistor) constantly pulling strobe #0 Hi.
Having said that it should also be causing lamp issues and player display 1 issues.

Provided that J2 and J3 are disconnected, then no.
If CR43 was open circuit, J2 pin 1 would indicate Low via the pull-down 56k ohm resistor at R132.
If CR43 was short circuit, then banded side would be pulsing like the non-banded side.

Your previous 56k ohm measurement on J2 pin 1 with respect to ground did not indicate any short after the banded side of CR43.

Please confirm 100% that all measurements since post #31, connectors J2 and J3 have been disconnected?

Remeasure the banded side of CR43 but this time flex the board while doing so, especially around U10. Any change in response?

How much resistance do you measure between pins 1 and 2 of J2 and also pins 1 and 2 of J3?

There's no pics showing the back of the J3 pin header. Can you see anything amiss there?

So it was still indicating HI on the banded side of CR43?

What variety in readings were you getting? On spec you should have got a total 112k ohms via the two 56k resistors at R132 and R111.
I can't see anything in your pictures or resistance readings to show the signal from the non-banded side of CR43 onwards being shorted to anything.

With J2 and J3 still disconnected and the board powered on, how much voltage do you measure at:
J2 pin 1 =
J2 pin 2 =
J3 pin 1 =
J2 pin 2 =
CR43 banded side =
CR43 non-banded side =
J1 pin 15 =
U10 pin 2 =
U10 pin 5 =
U10 pin 40 =

Unsolder diode CR43. What does your logic probe now indicate on J2 pin 1?

Below is the MPU schematic of the switch strobe #0 signal as it comes out from U10 pin 2 (PA0) signal. Orange trace shows its travel to the non-banded side of diode CR43.
Banded side of diode CR43 onwards trace in green - as you can see there is nothing from the banded side of the diode onwards connecting it high, all connections are to ground.

MPU_Strobe0.jpg

With power OFF and J2/J3 disconnected, measure the resistance across R132 (which is for J2-1) and then R111 (which is for J2-2). Both should be about 56k ohms. When you measure across J2 pins 1 and 2, effectively you're measuring through R132 and R111 in series and should get 112k ohms total.

Check them in circuit first (again, machine OFF and J2/J3 disconnected). R133 and R60 should both measure around 1.2k ohms.

At 2.02 volts, in TTL circuits this is a very marginal HI. In CMOS circuits this is neither HI or LO. If you set your logic probe to CMOS you won't get a HI reading.

Please replace CR43 if you have a spare 1N4148 or 1N914 diode.

Yes, go for it!

The caps in the circuit are too small to have this effect. BTW if you're touching the metal ends of the multimeter probes with your fingers, your body will affect the resistance readings.

If you're replacing switch matrix diodes on Mars God of War, then 1N4148 are not suitable. See the current discussion here, particularly the second half of the thread:
https://pinside.com/pinball/forum/topic/need-help-with-system-1-game-slow-response-time-

Just above the J3 connector, unsolder one leg of resistor R59 and lift that leg from the circuit so it's not making contact. Does this make any difference to J2 pin 1 reading HI with your logic probe?

Regarding CR43, you could temporarily try replacing it with a 1N4004/1N4007 diode if you have either.

Yes, you need a diode in CR43. If the original CR43 is still there that's fine for this test.

Changing Q55 didn't make a difference? What replacement part did you use?

The lamp symptom suggests Q55 is faulty. Can you swap it with another of the large SCRs on the lamp driver board and see if the fault follows the SCR?
BTW is it a typo that the part number is really a CR 106-6R ?

Alright, this suggests some resistive short between pins 1 and 2 of the J3 connector.
With R59 lifted, pin 1 of J3 is connected to nothing. So with the machine OFF and J2/J3 disconnected, if you do a resistance measurement between J3 pin 1 and J3 pin 2, do you get any reading?