I'm wondering why Bally used 120ohm resistor pull-ups on their solenoid driving transistors (TIP120 equivalent) and really why they did the pre-driving the way they did it. It seems like it's really just a waste of electricity when the solenoids aren't firing.

My understanding of the circuit is that all 19x transistors are theoretically ready-to-fire via the 120ohm pull-up on each circuit. The only thing that prevents them from firing is the 3081 chip that shunts the current from the 120ohm resistor pull-up to GND. So that means the majority of the time when solenoids aren't actually firing, 41.7mA of current is used per transistor circuit. 19 transistors x 41.7 = 792mA load on the Solenoid Driver Board, just to keep all the transistors from firing.

Without the 3081 chip in there (say you removed it) or if it's bad/shorted... the solenoids that it was keeping from firing (via the shunt of the NPN base to GND) will actually just fire as soon as the machine is turned on. There was a recent thread with that happening & some looking back at RGP history also revealed some discussion about how this all works.

So why did Bally choose 120ohm pull-up instead of something of higher value? Speed to activate? And on that note, why not have done some kind of inverter or pre-driver circuit here instead?

I'm wondering if there's a modification that can be done to these boards so they work "as original" but the current can be greatly reduced. 800mA just to keep coils from firing that are inactive most of the time just seems a bit ridiculous to me at this point. I know Bally engineers were smart guys, just not understanding why they did what they did here.. and why there can't be a better solution.

Cool! Exactly the kind of thing I was wondering (as far as mosfet replacement & higher ohm resistor used).

These Classic Bally/Stern machines don't get a ton of press for some energy-saving board mods like that. I wonder if you could then get the draw to a few hundred mA (or less) with mosfets versus the current 800mA. That'd help out the linear regulator quite a bit and subsequently reduce current draw on the 11.9vdc as well.

Is there a link with instructions somewhere? Or were you just talking about mosfet mods in general (ie. Williams Sys3-7, Sys11, etc). I hadn't come across anyone doing it for Bally's SDB.

I'd imagine temps do drop quite a bit.. hah.

I'm not too familiar with best practices with mosfet designs, but did find some RGP discussion from back in 2007 that related https://groups.google.com/forum/#!topic/rec.games.pinball/iVq1JLqMf2Q

Say an 22NE10L / IRL540N or equivalent N-Channel Mosfet was used in substitution for the SE9302 / TIP120. Would changing the 120ohm resistor over to a 10k & adding a 0.01uf cap in parallel to the 330ohm resistor (for "slew rate limiting" as mentioned in the RGP link) get the job done? Is the rest of the circuit okay to leave as-is?

I was reading up on the purpose of the 330ohm resistor, .002 cap & diode that are on the NPN base for each solenoid driver. Seems to do with slowing the speed for the solenoid to turn off. I'm not seeing this in the Rottendog schematic for their SDB that uses mosfets. Nor do I see a 0.01uf cap on the gate for "slew rate limiting". Ed mentioned Alltek handled the "slew rate limiting" a different way on their boards, so didn't need the 0.01uf cap.

I guess what I'm saying is, I'd like to try this out and ditch the 800mA overhead that an original board takes up due to this solenoid circuit. I think it'd be an awesome "bulletproofing" addition.. as most of what's probably causing 5v/12v headers to get stressed is related to load.. and half the load of the entire board system could be eliminated just by changing over to mosfet circuits on the solenoid drivers.

BOOM. 1.6A load on both the unregulated 12v & 5v linear drops to 0.8A in just this one change.