Bally 742A - Summit MicroControl Questions

MIPS

Well-known member
Joined
Jun 4, 2017
Messages
457
Reaction score
381
Location
Kamloops, British Columbia, Canada
There was a 742A-192 machine not too far away from me that I decided to take a gamble (pun) on and buy from photos alone. The photos I received showed the exterior of the cabinet only. Included keys, all components, it "turned on and the lights worked" spare bulbs and a roll of nickels. Apparently it saw life at the Mitspah Hotel and the maintenance card indicated it was worked on as late as April 1982. A book of matches was found jammed in one of the door hinges. It ended up here about 20 years ago when it was inherited along with other belongings from someone who lived in Reno. The new owner was downsizing and was looking to find a good new owner for it.

CGS_7952.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7952.jpg

CGS_7979.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7979.jpg

I walked into this expecting it to be electromechanical and was told that it "used to work, but one day while the kids were playing it the reels stopped locking" and expecting it to be frozen grease. No it seems this machine has a microcontroller running the show.

CGS_7956.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7956.jpg

CGS_7960.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7960.jpg

CGS_7973.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7973.jpg

CGS_7975.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7975.jpg

There was a badly leaking battery on the main control board. Everything cleaned up okay in the ultrasonic bath but I was not prepared to work on this. Is there any documentation for this arrangement? There was no indication of voltage on the old cell. Does anyone know what a substitute might be?
Ragegoom has a few interesting photos and pieces of information on his site that might be useful but right now everything is loading from there very slowly. I'll try again tomorrow.
 
Last edited:
Spent this evening cleaning and reoiling all the pivot points and was able to get all the linkages to reliably operate. Also found what I guess used to be a rubber seal in the air piston had turned to goo and was jamming the piston. I can remove it and repolish the piston but where will I be able to find a replacement?
Also checked out the logic power supply and tried powering up with everything cleaned and reinstalled. I seem to be getting good DC power but all that happens is every relay, lamp, meter and solenoid in the machine bangs on and off at random. Perhaps it's not happy with no battery installed or I missed some battery damage?

I tried the page again this morning and was able to locate the maintenance manual which did a swell job of explaining everything. It seems fairly straightforward in my configuration however did not specify what the original battery was rated for other than tilts start to happen if it drops below 3.7v. Here is the board before I removed the battery.

CGS_7948.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7948.jpg

I'm about to dump the EPROM, then I'll set my sights on inspecting a bit harder for battery damaged traces.

Edited: Blast. Looks like I can find the documentation for the diagnostic ROM but not an image of the ROM itself.
 
Last edited:
The board had to be almost completely stripped to clean out the remaining battery residue. Along the way there was one bad trace discovered and Q1 (MPS3646 transistor) was found to be bad. Don't have a replacement yet as I am unsure of a modern substitute.

CGS_7980.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7980.jpg

I also took the chance to replace the godawful side wiping sockets that I know and hate well from Commodore's machines.
Also apparently you can get away with using a rechargeable 3.6v pack from a cordless phone as a replacement battery.

Also reassembled the air cylinder after carefully removing the old rubber seal and reinstalling the piston with the heaviest grease I could get my hands on.

CGS_7981.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7981.jpg
 
Replaced transistor on speculation that a 2N3904 should be close enough to the original spec. If that fails a 2N2222 should also work.
The battery has been substituted with a 3.6V rechargeable pack from a cordless phone. I'm seeing this done to other boards and should be sufficient. Did not stop the freaking out though. :(
I noticed that if I just pull the relay board I can shut everything but the three digit display on the front up and the display can turn out to be handy for diagnosing if it's still losing its mind or not. I should at the very least see a tilt code. Instead it just continues throwing garbage at me.
I dug through my box of junk and found an Intel P8035L microcontroller. It's electrically identical but has half the internal ram. My theory is it should start to do something and then fail because if the game code needs 128 x 8 ram space and should at least hang. No dice. Still freaked out.

At this point I've also managed to disassemble the game chip code which is straightforward MCS-48 assembly. So far at least it appears intact but unless someone says otherwise, I'm not posting it here for review.

The other option is the Diagnostic ROM. HMS electronics sells them still for $35USD each, which is probably why I couldn't find an image. If that fails well then I have no choice but to consider the game chip bad and have HMS make up a new one for $85USD....which will be fin in its own way because the form you have to fill out assumes the traditional fruits, bars and bells, This whole money value for lines is the weirdest thing I have ever seen.
 
Last edited:
Man there is a total lack of information about this system.
I sat down tonight and ran over the board and pulled up datasheets for everything. You can find ALL of them freely on google.

PCB%20IC%20Locations.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/PCB IC Locations.jpg

I also ran over the other photos of these boards and came up with a list of known compatible substitutions.

U1
-INS8248N


U2
-Any 2732 EPROM should work


U4
-HCMP 1824P
-CDP1824CD3
-CDP1824CE


U5
-74HCT373P

U6
-SN74LS145NDS


U7
-INS8039-6


U8
-ULN2003AN
-ULN2003A


U9
-ULN2003AN
-ULN2003A


U10
-INS8243N
-P8243
-M5L8243P


U11
-HEF4512BP
-MCM14512


U13
-CD40106BE
-MCM14584


U14
-HEF4512BP
-SGL4512BE


U15
-CD40106BE


U16
-MC14093BCP
-HEF4093BP


U17
-HEF4512BP
-SGL4512BE

If you have not noticed I am trying really damn hard here to assume it's not the EPROM that has failed... >_>
 
Decided the hell with it and stripped all the DIP's off the board so I can put them in sockets. With them removed however I took the chance to verify all the resistors, capacitors and diodes were all within spec. The reversal diode on the battery circuit was bad. There was another resistor I was having trouble getting a good reading on and swapped out but the tantalum seemed okay and all the smaller blue decoupling capacitors seemed okay. Also did not find more battery damage which is good. The resistor networks all seem to also be within their value range.
It almost certainly has to be an IC issue at this point, as if I ahve not beaten around this bush enough. Perhaps I can power the board on the bench and attack it with the oscilloscope.
 
Got it.

CGS_7991.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7991.jpg

CGS_7992.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7992.jpg

CGS_7994.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7994.jpg

When poking at it with the scope it became immediately obvious that some of the data lines to the EPROM were either too weak to give valid data or completely missing. I went back over and discovered roughly a dozen of the through-hole vias measured good continuity side to side but if I tested at the traces they were open. I had to put the board under a bright light and look with the loupe to find the joints where the trace connects to the via was either about to break or broken open. After repairing those I put the board back in the machine and it came right up and worked on the first play.

Now that it works I can keep a copy of the EPROM for safekeeping and the new battery pack is secured to the side of the reels with a piece of velcro where it can't leak and cause problems again.

CGS_7993.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7993.jpg
 
Now that I've had a chance to feed a few hundred coins through the machine and actually played countless games I have started to find the other quirks of this machine that I can't really explain.

CGS_7990.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7990.jpg

1) The 2N3904 was not really a suitable replacement for a 3646. I ended up going back and replacing Q1, Q2 and Q3 with 2N222N's.
Even then however, I still find that some segments will either be dim or flicker or can be brought back by smacking the cabinet. There has to be a loose connection somewhere.



IMG_0368.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/IMG_0368.jpg

2) For the first five minutes or so after I turn the machine on I get games interrupted by tilt code -4- after the first or second reel indexes and locks. It's non-existant according to the documentation, but *4- (where * represents one of the three reels) does and indicates that while indexing a reel spun backwards.

3) Also notice the TILT lamp is no longer on. The bulb reads continuity still but no power is getting to it.

4) self-correcting tilts (like the one above) according to the manual should release the arm to replay but it does not. I have to reset the CPU to release the handle.

5) None of the counters are working. I can play for a good 15 minutes and all the ram-resident counters will remain at 000 000. The ram and battery will however remember the last state it was in. or if it was in the middle of a payout. I can put three nickels in, turn it off for the night, turn it on the next afternoon and it will remember the coins inserted and be ready to play. Likewise the only mechanical counter in the machine is not working either, but I know it has previously back when the CPU was still randomly firing off outputs.

6) The payout odds suck ass. I don't know how it's doing the game electronically but in the last two days I've only had four payouts out of way too many coins and hours. There are lots of off-by-one-notch moments but the reels are indexed, so it's not like there's a bug somewhere on them. The machine would immediately tilt.
 
Last edited:
A popsicle stick with emery tape glued to the ends makes an amazing contact cleaning tool.

CGS_7998.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7998.jpg

I pulled the reel and hopper assemblies and cleaned all the contacts. This has seemingly not only resolved the issue with the display not working correctly but for some reason both the electronic and ram resident meters have started working. Likewise the payout odds have sharply improved, so I'm riding on a hunch that odds are at least in part being generated using the meters.

CGS_7997.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7997.jpg

I also burnished the bulb for the tilt lamp to no avail and then proceeded to track the wiring back to where the fault was. The wiring eventually terminates at K14 on the Slot Interface Board.

CGS_7975.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7975.jpg

CGS_7976.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_7976.jpg

This board is not like the other boards I can find depicted on the internet. This version uses the now discontinued OFA1202 solid state relay.
I traced the control side of the relay back to the CPU where it terminates on pin 3 of U6. I was able to easily see if it was changing signal states by making the machine tilt, and it was, so the relay was now suspect. Thankfully the interface board uses nothing but the same model solid-state relay so a simple location swap and the tilt light came back, confirming that at the relay was bad. I will have to do some research on what a suitable replacement might be.
 
Relay issue was resolved.
After discussing the situation, load and the control input it was determined that we should be okay to substitute the OFA1202 with a mechanical TQ2-5V and diode for a flywheel on this particular circuit. The contact rating is 1A which should be okay for the load of a bulb in the front and the candle. The relay is a double pole so they were bridged together. The relay can be made to be a drop-in replacement with a bit of perfboard.

CGS_8005.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_8005.jpg

CGS_8006.jpg

http://i11.photobucket.com/albums/a166/ballsandy/New Bucket/Slot Machine/CGS_8006.jpg
 
The system ran fine for the rest of the month and then this morning it began throwing those undocumented -4- tilt errors. Those could be cleared with a reset but now I got a -5- tilt code. This is not clearing with a reset. It too is also undocumented.

For reference:
-0- INTERNAL COIN COUNT METERS ERASED CHECK NICAD BATTERY AND/OR POWER SUPPLY
-1- COIN-IN JAM, SELF CLEARED CHECK COIN PATH, THEN PUSH HOPPER RESET AND CONTINUE
-2- COIN-IN SWITCH JAM CLEAR JAM, CHECK COIN PATH, THEN CONTINUE
X3- REEL X SPINNIN SLOW PULL HANDLE FOR REPLAY, NOTIFY MAINTENANCE IF EXCESSIVE REPEATS
X4- REEL X SPINNING BACKWARD
X5- CANNOT READ REEL X WHILE SPINNING
X6- REEL X FAILED TO STOP
X7- REEL X OVER SPEED
X-Y REEL X INDEXED EARLY BY Y STOPS PULL HANDLE FOR REPLAY, NOTIFY MAINTENANCE IF EXCESSIVE REPEATS
X-Y REEL X INDEXED LATE BY Y STOPS
-8- COIN-OUT SWITCH JAM CLEAR JAM, CHECK COIN PATH, HOPPER RST, PAYOUT CONTINUES WHEN DOOR CLOSES
-9- COIN-OUT JAM, SELF CLEARED CHECK COIN PATH, PUSH HOPPER RESET, PAYOUT CONTINUES WHEN DOOR CLOSES
--- EXTRA COINS PAID OUT NOTIFY MAINTENANCE, PUSH MASTER RESET TO RETURN GAME TO PLAY
--_ FREE HANDLE PULL, SELF CLEARED CHECK HANDLE RELEASE AND/OR HANDLE MECH, NOTIFY MAINTENANCE
 
Back
Top Bottom