Connect with us

Pinball switch closure problem

Discussion in 'Electronic Basics' started by frenchy, Mar 22, 2005.

Scroll to continue with content
  1. frenchy

    frenchy Guest

    For years the pinball companies used a small .047 uf capacitor across
    some switch terminals for switches that could be closed and reopened
    very quickly like a target getting slammed particularly hard. The
    capacitor would increase the 'closure time' that the CPU would read the
    switch as closed, long enough to help it pick up the switch closure.
    These switches are just simple two blades and contacts, with one wire
    to the cpu, the other wire thru a diode to isolate it within the matrix
    of switches in the game, and the capacitor may or may not be across the
    terminals also.

    I'm having a problem with a particular swtich that the cap fix is
    working, but not completely. Some hardhits will still not register.
    Tried putting bigger cap on it with improved results but I've read that
    too big of a cap can cause 'ghosting' where it might start causing
    false closure reading on other switches in the matrix. So my question

    Is there a simple formula where one could increase the size of the .047
    cap or .1 cap or whatever is being used, but put in in series with a
    resistor, to lengthen the time that it is doing it's thing of
    lengthening of the closure to the cpu? I.e if I wanted to use a cap
    10x bigger, what resistor could I put in series with it to be sort of
    like the smaller cap without the resistor, but length of discharge
    would be stretched out? Or is this possible? I am pretty good at
    fixing pinballs, but electronics theory is not my calling. thanks for
    any assistance anybody can give me. Thanks!
  2. Active8

    Active8 Guest

    A schematic of the matrix and everything associated with it would
    help so I don't have to guess how to recreate it. Hint:

    View in fixed-width font, like Courier.
    -o/ o- ->|-


    created by Andy´s ASCII-Circuit v1.22.310103 Beta

    Putting a R in series with the cap will slow the charge *and*
    discharge and make it look like a bigger cap. The formula is T=RC
    which is how long it takes the cap to charge to something like 63%
    of the voltage applied or drop 63% from it's initial voltage to zero
    if that's where it's headed. Big R, big T.

    If the machine worked before, what could have changed that it
    doesn't work now? Did you replace the original cap with a new one of
    the same value?

    As far as that "ghosting" is concerned, the diodes should prevent a
    read of another switch from detecting the switch in question. The
    worst that could happen is that the switch acts likes it closed on
    the next read. Set the replay score higher ;)
  3. frenchy

    frenchy Guest

    This is a switch in a current new pinball model (Lord of the Rings)
    that has this common problem even when new (many owners report this one
    particular switch sometimes won't register as it is in direct line of
    the flippers and gets really hammered with fast shots). This company
    does not currently put caps on their switches. Other companies have in
    the past on troublesome switch locations like this. So I am
    experimenting with caps on this switch to solve problem. A .05 cap as
    used in the past did not completely eliminating the non-registering
    (but improved it). Looking to eliminate it if possible by increasing
    the switch closure seen by the cpu even more with bigger cap + a
    resistor if possible. Once pin company did do something along these
    lines but only on their non-cpu-controlled coils by putting a 22 uf cap
    and a 100 ohm resistor across the switch, but I don't want to assume I
    can just do that here as now I'm dealing with a switch matrix and
    another company and don't want to damage anything. Just trying to come
    up with rough estimate on, if .05 is the usual cap, what resistor would
    I use for a cap 3x as big, or 10x as big, for longer closure. thanks!
  4. frenchy

    frenchy Guest

    <<As far as that "ghosting" is concerned, the diodes should prevent a
    read of another switch from detecting the switch in question. The
    worst that could happen is that the switch acts likes it closed on
    the next read. Set the replay score higher ;) >>

    In the case of this pinball switch, it wouldn't hurt if it remained
    closed for the next scan, it is only going to score once anyway even if
    you held the switch closed by hand for the whole game. So getting it
    to stretch to the next scan would be great..Frenchy
  5. Active8

    Active8 Guest

    ..05 ?? please don't round standard values.
    If I read you correctly, the RC is a snubber to absorb the
    inducktive kick-back from the relay coil thus preventing arcing at
    the switch contacts. Otherwise T=RC so bigger cap = longer T. It's a
    simple proportion

    I'd still like to see the schematic of the switch/diode matrix and
    the way it's connected to the CPU/MPU/MCU so I'm not taking shots in
    the dark. I have to go back to your OP (original post) just to
    remember how you said you think it's lashed up. I can only guess:

    | CPU |
    | in outs |
    | | |
    | | |
    | o o
    | '\ '\
    | \ \
    | o \ o \
    | | |
    | | | <----- X
    | | |
    | V V
    | - -
    | | |
    | | |
    created by Andy´s ASCII-Circuit v1.22.310103 Beta

    If this is it (even if it isn't), I don't think a cap *across* the
    switch is good at all - the cap will pass spikes on every transition
    of the scan line.

    There also needs to be a pullup or down R in there and that's going
    to affect the time constant.

    If the outs go high to scan a switch, a cap from "X" to ground would
    hold it high after the switch opened. If the outs are active low and
    the diodes are reversed, cap to V+ .

    As I said previously, a series R will slow it's charge as much as
    it's discharge, but that assumes the source R of the outs is the
    same as the input R of the input. With a CMOS/NMOS CPU, the cap
    would stay charged for a very long time - MOS inputs require
    theoretically no current, thus no discharge path. Some MCU's have
    CMOS outs and TTL and Schmitt trigger ins - another situation.

    Maybe the best one size fits all is a R in parallel ( || ) with C
    and the time constant is T=RC. The cap will charge to full roughly
    5*RC (5*T) where R is the output R of the CPU scan out line(s). That
    R would be a low value and the cap would charge very quickly -
    probably draw excessive current too, and need a limiting R.

    On discharge, the R in T=RC will be the parallel R. After 3 T's (5
    T's is considered to be the practical time for full
    charge/discharge) the held high should now present a low to the
    input - a reasonable guess. You can fiddle with T. The oscilloscope
    is your friend if you have to fiddle with values.

    R charge limit
    ----|___|----+-----+------>|--- to in
    | |
    from switch | |
    --- .-.
    --- | | R discharge
    | | |
    | '-'
    | |
    created by Andy´s ASCII-Circuit v1.22.310103 Beta

    Determining how long you need to hold it high depends on the scan
    rate. If you don't know that, any equation is useless.

    You could just put the RC network at the input (after the diode and
    it would take care of all the switches. I like that better.

    I hope all this helps your understanding. A schematic would've
    helped a bunch. Next time I'll hint louder.
  6. frenchy

    frenchy Guest

    I don't know how to draw schematics good so I will just describe the
    various switch circuits:

    Switches - single switch, with one input going thru a diode first to
    isolate it in the matrix (if it's in one)
    ..05 cap - put across certain switches in matrix to lengthen time they
    are read. Cap is just placed across the switch. Non-polarized.
    Sometimes done by the factories.
    22 uf cap and 100 ohm resistor - the company that used this did it for
    stroke increase reasons, not for arcing or anything like that. Placed
    in series across the switch like the cap above. Was used on some
    kickers and pop bumpers that did not have to be cpu controlled (they
    just had to be smart enough to fire when they were hit). Cap and
    resistor lengthened the time the coil fired. Coil was still controlled
    with pre-driver circuitry which is what was being lengthened in
    activation time. This is a different application than some much older
    pinballs used occasionally putting a very large cap in circuir with the
    coil power itself to help prevent arcing on electromechanical switches.
    I will see if I can point to some schematics of this particular switch
  7. Active8

    Active8 Guest

    Come on man! look at the line under my ascii art and get with the
    program, pun intended.
    Again, A cap across a switch will pass spikes when the line is
    scanned. If the software debounce ignores it, great. I'll have to
    convince myself that it's a solution. For now, T=RC and all the
    stuff I said about it still holds.
  8. frenchy

    frenchy Guest

    In the two applications of switch caps in solid state pinballs there
    ever were - the .05 ufs across matrixed switches, and the 22 uf + 100
    ohm resistor on stand-alone pre-driver circuits for firing certain
    coils (both to lengthen how long the circuitry behind the switch 'saw'
    the switch as closed) - I've never heard of anyone reporting, nor
    encountered, false firings of the coils due to the caps from spikes.
    Maybe it's just the details of how pinballs typically are designed as
    per the switch reading that this is not a problem. Or the values
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day