Thought it was about time for another Taxi update.
Off and on I've been trying to resolve the issue of the lower pop bumper locking on.
I have to take a moment to thank Pinsider DumbAss -- not only was he a source for a new GI Relay board but he has also walked me thru/educated me on the topic of System 11 Special Solenoids. GRUMPY was willing to provide me with sound advice as well. Two GREAT pinball people with a wealth of knowledge that helped me solve (hopefully!) this problem.
Before I get too far into things let me say that my terms/explanations may not be 100% perfect. Sometimes I struggle a bit with the vocabulary but try to present a situation as my mind sees it -- hopefully in a way that makes sense.
If I backtrack a bit, you'll remember that the lower pop (or jet) bumper coil had obviously locked on at some time in this game's past as it had melted the coil sleeve inside it. I replaced the coil and made sure the actuation switch was set correctly but still saw no action as there were other components to fix on the CPU board. I started with replacing components related to Solenoid #21 -- specifically Transistor Q77 and Q76 that supply a path to ground for the coil when Q76 switches Q77 "on".
5J5A3723 (resized).JPG
After replacing those components the lower bumper would immediately lock on when the game was powered on. The Q77 Transistor was now able to send the coil current to ground, but was doing so when it was not supposed to. Looking at the resistors I thought R104 looked a bit toasty and replaced it even though it tested fine. . . with no change. Yes, I did not have the extra appropriate spec resistors so I borrowed them from solenoid #22 which is not used in Taxi.
5J5A3728 (resized).JPG
At one point DumbAss directed me to differentiate between a board problem vs. a playfield/switch problem. To do this I pulled the switch connector coming from the playfield off the CPU and powered the game on. The coil locked on immediately, telling me a problem was still present on the board. My next step was to replace U50. To be honest, my understanding of the current/signal paths gets a bit fuzzy once the chips start getting involved, but U50 is an integrated circuit NOR Gate. I think of it as a hallway with a bunch of doors -- some for current entering and some for leaving to/from other parts on the board. U50 is what signals Q76 to open Q77's path to ground. Instead of just replacing the chip I chose to put a socket in first, then a new 7402 IC.
5J5A3714 (resized).JPG5J5A3715 (resized).JPG
When the CPU was put back into place and once the game was powered on I thought the problem was fixed. . . but after a few minutes I was proven wrong. I was actually able to play a game or two but before long I noticed the bottom coil had locked on again. Something had caused our Q77 transistor to fail again. My next step was to replace the flyback diode (D20) on the Auxiliary Driver Board -- the original tested ok but it was an easy swap just in case it was wavering.
5J5A3722 (resized).JPG
Power the game up. . . play a few games. . . and then still the same result -- lower coil locks on.
At this stage I'm trying to decide if the issue is farther upstream on the CPU board or maybe, just maybe there is another issue on the switch/playfield area. I examined the switch contact spacing and made sure nothing was binding that could cause the switch to stick closed. It was after reading thru a number of archived Pinside posts that I directed my attention to the electrolytic capacitor on the actuation switch. I read that if the capacitor has shorted it can cause the coil to lock on, so I reasoned that if this one was near failure and it was only after it warmed up that perhaps the short was happening. So I snipped one leg of the capacitor free from the switch.
5J5A3851 (resized).JPG
Powered the game on. . . and proceeded to play for a half hour or more with no coil locking on!
I will order some new capacitors to replace the originals (22uf 100V). In the end I believe there were a number of issues with my game that caused improper function of the coil and it was very satisfying to work thru each area/component to find the solution -- even if it took a couple weeks of working on the game here and there.
Once again I cannot thank DumbAss and GRUMPY enough -- it just goes to show that you can't judge a Pinsider by his username. I'd say both are rather intelligent and quite pleasant to talk to.