(Topic ID: 88959)

Bench testing Bally MPUs - 7th flash approach

By adalogue

9 years ago


Topic Heartbeat

Topic Stats

  • 20 posts
  • 7 Pinsiders participating
  • Latest reply 9 years ago by Zitt
  • Topic is favorited by 4 Pinsiders

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    View topic image gallery

    14020072376_45b816f1aa[1].jpg
    photo 1.JPG
    photo 2.JPG

    #1 9 years ago

    I'm bench testing a Bally MPU and can't seem to get the 7th flash without the cheat (jumper from U12-3 to U14-14). I am providing the board 5, 12, and 21.5 VDC at the correct test points, but no flash without the cheat, which doesn't let me test U14. PIA input from ZCD circuit is constantly low, making me thing U14 may be bad (Clay's guide).

    The whole "zero crossing detector" nomenclature makes no sense to me. The incoming voltage when installed in a game is 43VDC, voltage divided down to 21.5VDC at the test point. There is no zero crossing, it's a positive waveform after rectification on the rectifier board.

    In short, is it possible to boot the board up with 21.5VDC on the bench? Or is a bench supply too clean of a signal to allow the ZCD circuit to function properly?

    #2 9 years ago

    I've bench tested many Bally MPU's and frankly, getting to the sixth flash in generally a home run if you happen to have started with a dead board. I have never run into a problem with the seventh flash that would make me want to try and figure out how to bench test it. Why not just put it back in the game and see if it works and whether or not you have the 21.5v at test point 3?

    #3 9 years ago

    Good idea, can do. Was just hoping to check U14 on the bench. I suppose it could feed it logic on the inputs and verify the ouputs as a basic test.

    Also trying to get a better idea of what the ZCD actually does, or how it actually works.

    #4 9 years ago

    So.... Alternating current goes 120v high and 120v low at a set freqeuncy. This circuit watches for the zero cross pulse and uses it as timing for interrupts. Say the feature lamps. The game lights up a feature lamp and then uses zero crossing to determine how long to keep the lamp on. On solenoids, the game uses the zero crossing timing to determine how long to keep a solenoid pulsed for. The software waits, 10, 100, whatever times it "crosses" zero.

    To fully test this circuit you need to do it in a game that supplies +43v on j4. Then just follow the schematic. a 4049 at u14 is an inverter. So the output is always opposite the input. Bally was nice and wrote down what voltages you should see in this circuit so follow the schematic.

    #5 9 years ago

    The seventh flash is in fact zero crossing. The voltage comes off the rectifier board as unfiltered, it's dropped to 21.5 VDC on the MPU board for the test point.

    Sending a filtered 21.5VDC signal in to the TP will not work, it has to be an unfiltered "pulsing DC" signal.

    #6 9 years ago

    The 43V solenoid voltage is a positive signal, it doesn't cross zero (go negative), the bridge rectifier circuit on the rectifier board prevents that. So how does the ZCD circuit detect zero crossings when there are none? If it watches for changes in the voltage signal, that would make sense. But the solenoid voltage never swings negative, so there are no zero "crossings" to detect. It will be 0V 120 times per second, put never negative. "Zero crossings" are defined as a voltage change from positive to negative. If it was called "zero touching detector" I'd buy it.

    #7 9 years ago

    Thanks dewdude82, that is what I was after, very helpful!

    #8 9 years ago

    barakandl, FYI US wall voltage swings between +/- 160V-170V. 115-120VAC is the RMS value.

    Also, are you sure the CPU board uses the A/C voltage to time the lamps and coils? I always assumed it was software timed.

    #9 9 years ago
    Quoted from adalogue:

    barakandl, FYI US wall voltage swings between +/- 160V-170V. 115-120VAC is the RMS value.
    Also, are you sure the CPU board uses the A/C voltage to time the lamps and coils? I always assumed it was software timed.

    Yes I am. I have a very intimate in depth knowledge of these bally/stern games.

    even though it is a dc current... it is unfiltered so there is pulses. the software counts these pulses as a known specific timer to trigger interrupts. It runs it through that 4049 just as a spacer between the PIA. Probably buffers and keeps the PIA safe since the 4049 can handle higher voltages.

    #11 9 years ago

    You can use an $8 transformer from radio shack http://www.radioshack.com/product/index.jsp?productId=2102701

    and a bridge rectifier $5 http://www.radioshack.com/product/index.jsp?productId=2062584

    and give yourself the proper voltage for the last flash when connected to the TP3 test point.

    #12 9 years ago

    vid - yes, that would work. I have a variac that could provide the correct AC voltage, that I could rectify. However, I tested another approach that didn't require fusing another A/C source (a transformer hooked up to mains could source a lot of current) and it worked. Here's what I did:

    - Signal generator set to 120Hz, sine wave, output max'd
    - 5V to TP5
    - 12V to TP2
    - Rectified signal from sig gen applied to TP3 (resulted in 11.6Vpk signal, half wave rectified, frequency of 120 Hz)

    7th flash no prob.

    So this confirms (for me) that the ZCD circuit is not looking for zero crossings at all (see soapbox above and lack of zero crossings in the scope waveform), it's responding to the rectified 43VAC input and doesn't seem to be sensitive to the actual input voltage, a range of input voltages seem to work (i.e. 11.6Vpk/5.6VRMS half-wave-rectified signal at TP3 worked just fine).

    Attached are photos of the setup and results, in case they are helpful to anyone in the future. Scope probes were connected as follows:
    CH1 - TP3
    CH2 - U10 pin 18

    photo 1.JPGphoto 1.JPG
    photo 2.JPGphoto 2.JPG

    #13 9 years ago

    Put a frequency counter on zero cross input to the PIA. It is 120 times a second created from unregulated pulses on the the 43v line. The software uses it as a known timer for lamps, switch, solenoid interrupt routines. You can cheat the 7th flash in many ways, it doesnt care about the frequency, just that it gets any kind of pulse. Use any type of pulse generator on the zero cross input and I bet will 7th flash. The game will not operate correctly at all though and i would imagine the software would crash.

    #15 9 years ago

    I use a 9 volt battery in series to the 12Volt ATX power supply. Then attach the positive side of the battery to the test point you.

    #16 9 years ago

    My goal is not to get the board to function installed in a game with a certain bench test setup, simply to be able to test the ZCD detection circuit and U10 PIA on the bench and get all 7 flashes.

    If the ZCD circuit isn't picky about what waveform or voltage it receives, what makes any sort of bench test methodology invalid if it gets the 7th flash to show the ZCD circuit is feeding the PIA properly and the PIA is responding to the alternating signal and thus allowing the 7th flash?

    I don't see how this would result in a PASS bench test and a FAIL game test if everything else in the game is functioning properly.

    #17 9 years ago

    Zitt, have you gotten that approach to work?

    Based on this thread and my bench experience, I think we've established that a constant DC voltage at TP3 won't get the 7th flash, it needs to be a signal that changes over time with a frequency of 120Hz, output of an ATX power supply is regulated and therefore a constant DC voltage.

    #18 9 years ago

    Diagnosing the zero crossing is simple and i always do it in a game. 9 times out of 10 it is the voltage divide resistors. Every board i touch i installed half watt rated 2k resistors. They are always burned up.

    So when you put a board in the game, just put logic probe on CB1 of u10. If it is stuck high or low, and you have around 21.5v on TP3, then you just replace the 4049 after a visual inspection of resistors. I typically don't even bother with checking with a probe at the u14 4049 because the circuit is so simple. Failure of U14 is somewhat common. You can have a failure in that voltage divide that sends over voltage to u14 as well as that 2 watt resistors right below it gets hot and cooks that area of the board as well.

    #19 9 years ago
    Quoted from adalogue:

    Zitt, have you gotten that approach to work?
    Based on this thread and my bench experience, I think we've established that a constant DC voltage at TP3 won't get the 7th flash, it needs to be a signal that changes over time with a frequency of 120Hz, output of an ATX power supply is regulated and therefore a constant DC voltage.

    Absolutely. That's how I got the MPU board debugged for the Star Trek: Mirror Universe.

    What I do is leave the 21V alligator clip off. Get it's 6 flashes... then temp connect the alligator clip to the TP.
    Then I get the 7th flash. Works every time.
    Now; granted it doesn't really "test" the zero cross... but it's cheap and works most of the time.

    Do you need a schematic?

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/bench-testing-bally-mpus and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.