Coin Counter Bypass Questions

FrizzleFried

Well-known member

Donor 2011
Joined
Mar 20, 2007
Messages
54,193
Reaction score
5,245
Location
NAMPA, Idaho
Star Castle and Armor Attack (among others) use a 3 lug switch in their coin doors that is attached to both the PCB and a coin counter. The NO connection goes to the coin counter and the NC connection goes to the PCB to give it credit. The PCB will NOT give credit unless it gets a coin counter hit first... I assume to make it harder to bypass the counter.

Anyway, I am trying to use a coin door "lock switch" which only has 2 lugs. With Star Castle, I can tie both the NC and NO connections and it will read both and will credit the counter and will coin up the game. It works great. Armor Attack is a different story. It will work as intended for the first credit... then after that it looks for a separate NO connection from the coin counter... which it doesn't get because it's tied to the other NC connection... so it sits with 1 credit...each time I credit up it will hit the counter but not coin up the game. If I remove the wires and short them independent from each other it works great. I also noticed something...and here is my question...

If I just short out the counter... it goes in to a "half count" ... but while it's shorted I can credit up as many times as I want. So I could ground out the counter... then the switch works great... BUT... is that mechanically stressing the counter? Is there some electronic device activated in the counter? I listen but I can't HEAR a coil or anything... will I burn out the counter using the method I propose? Is there some other way to bypass the counter?

Thanks!
 
It would help if you could post a small schematic of how it should be wired, and how it is currently wired. You say the original has a SPDT switch and your modification is using an SPST switch.

Perhaps the original had some sort of pull-up / pull-down on one (or both) of the lines. It sounds to me like the coin counter is unrelated to the problem but we'd need a diagram to understand.
 
Original ... one wire to PCB... one wire to counter... one ground. Wire to the PCB goes to the NO... wire to the counter goes to the NC and ground...

Basically a coin hit the switch...opening up the connection... which causes the counter click and send a signal to the PCB that it's been clicked... then when the switch closes back up again it sends the signal to the PCB that there is a credit there.

The PCB will not credit up without the counter telling it it's been hit.

I can ground the counter and the credit switch works fine. I am just worried I may burn out the counter (not that i really care... I just want it to work).

I am not sure how the coin counters work. If I leave it grounded, is it going to burn up the counter?
 
Ahem... no one knows how the internals of a counter works and whether I'd be jacking it permanently by grounding it out? Hell...I'm getting to the point where i'm just gunna do it and if it breaks... it breaks.

:D
 
Sorry I can't really follow your written description. I downloaded a StarCastle manual (from here) and have attached some screenshots. This is the connection to the coindoor:

attachment.php


I redrew that piece, a bit easier to understand:

attachment.php


The coinswitch indicates a coin by momentarily shorting 2:Brown to 3:Ground. The coin counter is depicted on "CPU Sheet 1":

attachment.php


So the coin switches connect at J3, go through the CPU board and a drive transistor, then out to the coin counter via J1. You can see the coin counter itself is really unrelated.

Notice on this sheet that the 1:Red / 2:Brown connect to a flip-flop. This looks like a hardware switch-debounce. The problem is that you've changed this from the original wiring so 1:Red no longer gets grounded when the switch is released. There's an truth table for "SR NAND latch" on wikipedia (but it won't let me add the link).

1. Red=0 Brown=1 then Q=1
2. Red=1 Brown=0 then Q=0
3. Red=1 Brown=1 then Q=no change

Your rewiring gets states 2 and 3 but never state 1, so Q goes 0 and stays 0. This is why you see it coining-up once and never again. You either need to keep it like it was originally wired, or come up with some clever workaround.

Note: I don't understand your description of what you've done involving the counter to "make it work" right now. I'm just trying to explain how this circuit is supposed to work. It's possible that you've got something shorted right now but you'd have to provide the schematic of your change.
 

Attachments

  • coin_switch.jpg
    coin_switch.jpg
    67.3 KB · Views: 59
  • redrawn.jpg
    redrawn.jpg
    8.3 KB · Views: 59
  • cpu_board.jpg
    cpu_board.jpg
    61.8 KB · Views: 58
Last edited:
To make it "work right" all I did was ground the counter wire... I can then coin up as many times as I want via the switch I installed in the coin door.

You're getting WAY too technical. BTW... Star Castle works fine by simply wiring both brown and blue wire to one end of the switch and black to the other... Armor Attack doesn't work... I only get a single credit. The only way I can get more credit's is by removing the two wires ... grounding the coin counter wire... then use the other wire to coin up as normal.

I suppose all I need to know is... will grounding out the coin counter so that it's "active" all the time going to kill it? If so... what can I do to bypass the coin counter so I can credit the machine with a single 2-tab switch (NO ... no NC tab).
 
Last edited:
FrizzleFried said:
Basically a coin hit the switch...opening up the connection... which causes the counter click and send a signal to the PCB that it's been clicked... then when the switch closes back up again it sends the signal to the PCB that there is a credit there.
This is not true, see schematic CPU sheet 1. CPU Board Q1 is the drive transistor out to the coin counter. There is no feedback telling the CPU that the coin counter has been hit.

FrizzleFried said:
I am not sure how the coin counters work. If I leave it grounded, is it going to burn up the counter?
Based on the description, more likely to burn up something on the CPU board. I offered a more detailed answer of what's going on. Your problem has nothing to do with the coin counter.
 
This is not true, see schematic CPU sheet 1. CPU Board Q1 is the drive transistor out to the coin counter. There is no feedback telling the CPU that the coin counter has been hit.

All I can tell you is that the PCB will NOT coin up (the Armor Attack) unless the coin counter is hit. I can post a video if necessary. It will stay at 1 credit no matter how many times I attempt to credit. If I hit the counter, I can then credit one time... and one time only (per hit of the counter)... if I ground the counter wire I can credit over and over.
 
Fixed.

:)

I don't know WHY it's fixed... but it's fixed.

I noticed looking at my Star Castle that the coin counter wasn't there anymore. Not sure why I didn't notice this before.

I figured...what the hell... and I disconnected the coin counter completely... and ... it works now!

So I guess the way to "bypass" the coin counter is to simply not hook the damn thing up.

As to why having the coin counter attached prevented the credits from increasing from 1... I have no idea... nor do I care... now.

:)

Thanks for the help though... I thought the counter was in the circuit "in series" between the switch and the PCB... I was wrong.
 
Back
Top Bottom