You may have wired the coil backwards.

^ if it’s like Taxi that’s not possible as the diode is on the auxiliary power driver board. Check to see if q75 is also shorted. It’s the pre driver for that coil

I’ll look at the one at work today I’m on the way there now

No.

Disconnect the special switch input at the cpu board, see if it still locks on. If it does, you have a board issue (74xx chip likely). If it doesn't, you have a playfield issue.

If it does it at bootup in game over mode, it's likely on the cpu board.
Predriver transistor, then the 74 series logic chip are likely suspects.

right jet is special solenoid #3, so U45

There's a sneak ground somewhere then if the transistors are good. It's on power up, not game start/self test?

I think you want to disconnect 1J18 and not 1J19. If you disconnect 1J19 the solenoid will never fire because you disconnected the drive wire (path to ground) from the solenoid.

Special solenoids are enabled by either the CPU or the special solenoid (playfield) switch. The controlling signal is combined (logical OR) at U45 or U50.

If you disconnect the playfield input this will isolate the problem to either the CPU board or the playfield switch (or supporting component). This is the best first diagnostic step. Replacing a transistor or IC on the CPU board will not fix the problem when the problem is a playfield switch.

If you disconnect 1J18 and the solenoid still locks on then the problem is on your CPU board. If the solenoid no longer locks on you have a playfield switch (or supporting component) problem.

Your pop switch is closed, possibly in a way that does not appear so (like mounting screw is not isolated inbetween the leafs). Grab your meter and probe the pop switch and see if it has continuity even when it appears open. Could be the switch diode as well.

Shouldn't be in game over with the connector removed that passes the ground signal back to the mpu.

isolated from the board, only possibility is a hard short to gnd, so you will need to probe the coil lug & look for any potential gnd interaction on either metal or anywhere a wire could be pinched.

i misread, my bad. yeah, you are isolated from the mpu

seems that way if it still locks with it disconnected

<disclaimer>

The following are my own opinions and are NOT authoritative. Please research the information provided before deciding to believe it or not.

</disclaimer>

For reference please see the attached image (below) from the schematic.

You have stated:

1) With 1J19 disconnected (connector with blue wire) the solenoid does not lock on.

2) With 1J18 disconnected the solenoid does not lock on.

3) You have replaced Q73, Q72 and U45 but the solenoid still locks on (on power on - does NOT need to be in active game mode).

The following can be concluded:

1) There is no errant contact with any ground reference in the solenoid (playfield) wiring. The solenoid ONLY fires when the ground is connected on the CPU board.

2) The playfield switch and supporting components are not providing an errant path to ground that would force the signal on U45-02 to be low (logical 0 or 0V).

3) Just because you replaced the components and they still test "good" does not mean they are now "good". An important point to note is that the solenoid locks on at power on. A previously replaced component can still fail (and be damaged) if the actual cause of the failure is not identified and fixed.

My logical thought process follows. All conclusions can be verified by measuring voltage (or using a logic probe) to check signal state.

The solenoid locks on at power on and that means the playfield solenoid switch signal input path (U45-02) is irrelevant. *ASSUMING* SS+EN signal (U45-03) is correct this would be high (logical 1 or +5V) that forces the output of the NOR gate low. This would cause Q72 to be in the "off" state. This does NOT exclude a failure in a component on this path that may be masked. If there is one you just cannot see it in the current state.

An unstable signal at any point in the solenoid control circuitry could cause a problem. You can check R106 to be sure that it is within specification. It can be measured in circuit. Pull U45 (you did put in a socket - right?) to isolate this path of the circuit. You can also check R105. This one typically measures lower than specification (2k versus 2.7k) but to be sure you would need to isolate it from the circuit.

Finally ... I have a TIP102 transistor that measures "good" with a DMM but when used in circuit it is shorted and locks on a solenoid (and subsequently blows a fuse). If a BJT transistor tests "good" it does not mean that it is "good". If it tests "bad" then it is "bad".

The only other thing that I can think of at the moment is that the solenoid (back-EMF) diode has failed. The consequences of this are that when the solenoid is de-energized (either by CPU control or by cutting the power) the back EMF will damage the TIP102 (or TIP122). It often causes the transistor to short so that next time the machine is powered on the solenoid will lock on. It continues to stay locked on until you replace the transistor. If you replace the transistor (again) without replacing the failed diode it will damage the transistor (again). Lather, rinse and repeat until you run out of transistors. You cannot test a solenoid diode without removing it from the circuit. Any attempts to measure it will result in measuring the solenoid instead of the diode. These are so cheap that it's often easier to just replace it.

ss3_hardware.jpg

If you leave 1J18 connected you have not differentiated any of the ground paths.

- 1J18 connected (and locks on) = could be a playfield problem OR a board problem
- 1J18 DISconnected (and locks on) = very likely to be a board problem (must differentiate further with 1J19 connected or disconnected)

So what is it? 1J18 DISconnected - does the solenoid still lock on immediately at power on? If so ... likely board problem. If not ... likely playfield (wiring or component) problem but not necessarily. The playfield switch input becomes active when the EN+SS is set to low (either enter diagnostics or start a game).

IMHO: Don't bother doing anything until you have a potential direction to follow. If you shotgun replace parts that's no better than throwing stuff at the wall and seeing what sticks. Shotgunning components can potentially make things worse.

<this is NOT a personal attack>

If a person posts asking for assistance and advice for diagnosis is provided it is wise to at least do the tests and (accurately) report back the findings. Without the information the next steps cannot be made with any level of confidence. There's a reason the tests need to be done - to differentiate potential problem spots. If the solution is as easy as "just replace this part" then it's likely that would be stated by someone and that advice would have fixed the problem. If the obvious "fixes" do not fix the problem then it's not an obvious problem and needs more investigation.

</this is NOT a personal attack>

You can use any System 11 manual. This section of the circuit did not change through the years of the System 11 board set life cycle.

When you replaced the solenoid did you replace the solenoid diode as well? It's a fairly safe assumption that you did this as the diode is attached at the solenoid rather than on any PCB.

1J18 disconnected and 1J19 connected and the solenoid locks on at power on ... you have a board problem. You have isolated the playfield switch input by disconnecting 1J18. Focus on the board paths to ground.

1) Path to ground is provided by Q73 (TIP102).

2) Q73 base (the transistor switch) is controlled by Q72.

3) Q72 base is controlled by U45-01.

4) U45-01 is controlled by U45-02 or U45-03.

5) U45-03 is the EN-SS signal that shared across all the switched solenoids. It is initially disabled at power on.

6) U45-02 is the variable input. This is either controlled by the software or the playfield switch. If EN-SS is (correctly) disabled at power on then this input is irrelevant because EN-SS will disable this input.

7) There are supporting components in the circuit to either prevent digital logic signals from floating or over voltage.

You can shotgun replace some components but if that does not fix your problem then you need to either shotgun replace ALL the components or start measuring to determine where the problem is.

I would start by measuring the supporting components (see the circuit image above) as they are easy to measure to see if they test "good". Again ... just because a BJT TIP102 transistor tests "good" does not mean it is "good". Ideally components should be measured in isolation but often they can be in measured in circuit - if you verify against the schematic.

If you end up shotgun replacing Q73 (again) then make a note if the solenoid locks on at power on. Leave 1J18 disconnected at the very first power on to eliminate errant playfield switch input. The solenoid should not lock on. If it does then post this fact and direction will need to be reconsidered.

1) If it does not lock on then enter diagnostics.

2) If it locks on immediately on entering diagnostics then you likely have an errant input on U45-02.

3) If it does not then run the solenoid test and allow the software to fire the solenoid. If it locks on then you should replace the solenoid diode. Make sure that configure the diode polarity correctly.

4) If the tests all pass then power off and reconnect 1J18.

5) Re-enter diagnostics and trigger the solenoid switch manually. Verify that it works the same as software control.

I don't want to insult your level of understanding (I assume your understanding is VERY good as you are replacing PCB components). If you don't understand or need clarification ... just ask. I'll explain. I just don't know what your understanding is and what you want to get out of this repair.

I'm trying to help you. Please help me to help you.