Announcement

Collapse
No announcement yet.

BYO Pinball Interface/Output Boards

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • BYO Pinball Interface/Output Boards

    I can see this particular topic is going to warrant a lot of discussion, so I'm breaking it out into its own thread:

    Originally posted by pinnies4me
    The output driver board for the high power devices was sourced from a run of boards designed to be used with pinmame emulating a CPU for a game where that CPU was no longer available. The project seems to have completely disappeared, and we managed to secure the last two of the driver boards. Having said that, there was a design error in the board and it would have been problematic but for nuggy's insight into the issue. Reproducing something similar (without the fault!) is not out of the question at all, just a bit of cost to have a very small run made. We haven't priced anything at all at this stage as we've got one each and it was always the intention that we were each going to make our "dream game" (I can't mention the theme of the next game which will remain hidden until completion, although the development will be public just like CI, just keeping the toys and artwork for the actual release), so there has not been any pressing need to think about making a board.

    My brother in law (an electronics engineer) seems to think he could knock something up easily, and there has been discussion on RGP of some readily available devices that might be able to be adapted to the job as well. Nuggy might be able to pipe in tomorrow if he remembers to turn his PC on...
    Interesting. Did you manage to source a circuit diagram of the original driver board? That, and Nuggys experience of the design error and any necessary enhancements would make it a lot easier to have reproduced than doing it from scratch.

    Failing that, where would we find an 'plain english' description of what the board actually did and how it did it?

    David, if an original circuit diagram was available, is pricing reproductions/design enhancements something in your domain?
    My Projects
    sigpic

  • #2
    Originally posted by dmworking247 View Post
    David, if an original circuit diagram was available, is pricing reproductions/design enhancements something in your domain?
    Yes, that's right up my alley.
    "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

    Comment


    • #3
      I've been mulling it over some more (and talking with another AA member) and was thinking the output boards could be implemented as so:

      A PC interface board. This board has the USB / serial / whatever PC connection and a small micro on it. It communicates to the PC, decodes the output commands and sends that raw data to one or more driver boards. It may or may not have functions for effects built in.

      The driver board does the hard work of firing the solenoids. It just takes the raw data from the first board has almost no smarts on it. The only smart thing it does is disable all outputs if it looses contact with the interface board and maybe give an indication of any gross faults.

      Different driver boards could be made to cater for LEDs, larger lamps, solenoids, etc and all looped from the interface board. Each driver board could have 16 outputs. (data works best in multiples of 8) This way, people could use whatever combination of driver boards they need.

      Having the interface board separate would mean that changes in the interface connection type (serial, parallel, USB) over time (or for different O/S's) would not affect the driver board designs.

      I'm also thinking that optical isolation on the solenoid driver boards would be good and allow for use on systems where the solenoid power does not share a common ground with the logic, etc. Ground loops and high current do not make for a happy story so this may be essential IMO.

      So, that's where I'm at so far. I welcome comments and suggestions as this solution could be used for other pinball / arcade projects too.
      "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

      Comment


      • #4
        Originally posted by David_AVD View Post
        I've been mulling it over some more (and talking with another AA member) and was thinking the output boards could be implemented as so:

        A PC interface board. This board has the USB / serial / whatever PC connection and a small micro on it. It communicates to the PC, decodes the output commands and sends that raw data to one or more driver boards. It may or may not have functions for effects built in.

        The driver board does the hard work of firing the solenoids. It just takes the raw data from the first board has almost no smarts on it. The only smart thing it does is disable all outputs if it looses contact with the interface board and maybe give an indication of any gross faults.

        Different driver boards could be made to cater for LEDs, larger lamps, solenoids, etc and all looped from the interface board. Each driver board could have 16 outputs. (data works best in multiples of 8) This way, people could use whatever ombination of driver boards they need.

        Having the interface board separate would mean that changes in the interface connection type (serial, parallel, USB) over time (or for different O/S's) would not affect the driver board designs.

        I'm also thinking that optical isolation on the solenoid driver boards would be good and allow for use on systems where the solenoid power does not share a common ground with the logic, etc. Ground loops and high current do not make for a happy story so this may be essential IMO.

        So, that's where I'm at so far. I welcome comments and suggestions as this solution could be used for other pinball / arcade projects too.
        thats an absolute shite load of boards there!!! noise, wiring complexity,expense $$, limitation of non programmable (pergame) special effects, protocol will be a nightmare or are you going to handshake with these/this board ?!!
        i dunno


        home guys got to be able to do it like he can fit mags or boost controllers etc, capable people but not needing a masters degree? maybe i missed the fundamental point ?

        logistical (board count, interconnect,noise/psu requirements/supply splitting, back box space (!!!) etc) issues imo
        sigpic

        "Oh my mistake..you are wrong."

        Comment


        • #5
          Originally posted by nuggy View Post
          thats an absolute shite load of boards there!!! noise, wiring complexity,expense $$, limitation of non programmable (pergame) special effects, protocol will be a nightmare or are you going to handshake with these/this board ?!!
          Shiteload? It's one USB interface and a small board per 8 or 16 solenoid outputs. Maybe the lamp board could have more outputs.
          Wiring complexity? Each board loops to the next with a short 10 way ribbon.
          Expense? I think it will be quite reasonable due to the modular design.
          No special effects per game? Isn't that handled by the PC anyway?
          Protocol? Serial (38400 baud?) over USB should be plenty fast enough.

          Originally posted by nuggy View Post
          home guys got to be able to do it like he can fit mags or boost controllers etc, capable people but not needing a masters degree? maybe i missed the fundamental point ?
          A custom pinball is not a project for the masses. It will require ability. It's not in the same league as making a mame or 48-in-1 cab. Comparing it to fitting mags is just silly.

          Originally posted by nuggy View Post
          logistical (board count, interconnect,noise/psu requirements/supply splitting, back box space (!!!) etc) issues imo
          Interconnects? Boards link via cheap crimped ribbon.
          Noise issues? If you mean electrical noise, good design should take care of that.
          PSU requirements? No different to any existing pinball machines.

          I expect the boards to be quite compact so I doubt space will be an issue. The solenoid boards could be mounted under the playfield if you really wanted to.
          "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

          Comment


          • #6
            You guys have the technical side of things brewing here, so I won't interfere, but I will mention that I'm a PCB designer for a PCB manufacturer here in Adelaide.

            I will be able to get some nice deals on bare PCBs when the time comes and perhaps a complementary run of prototypes...

            Comment


            • #7
              i have proposed a 1 boards solution, you have said/8/16 outputs per board, let me see, thats at least 2 X8's for solenoid control (and thats system 1 style ability!!) and at least 4 X8's for 32 lamps, hardly a comprehensive array of outputs for already 6X boards including master i presume(?), ribbon and loom (multiples of) do not make for an easy layout

              lets not forget that there will need to be transistor upgrades for hi current stuff then the seperate looming to each of those "boards" !!!!
              **** off,

              that is going backwards and are you going to have seperate mcu's on each board as stated ?! then that is mega expensive compared too, so no worries you go make the world as complicated as you like, i prefer the fundamentally sound and obviously cheaper 'kiss' principle....

              kiss my ass !

              [


              A custom pinball is not a project for the masses. It will require ability. It's not in the same league as making a mame or 48-in-1 cab. Comparing it to fitting mags is just silly.

              I beg to differ, it is not hard as per se, it requires technical organisation, something most are capable off

              I expect the boards to be quite compact so I doubt space will be an issue. The solenoid boards could be mounted under the playfield if you really wanted to.[/QUOTE]


              thats is poor mechanical engineering form right there, fine for prototype, no good long term.
              Last edited by Nug; 13th October 2008, 07:03 AM. Reason: Automerged Doublepost
              sigpic

              "Oh my mistake..you are wrong."

              Comment


              • #8
                Each driver board is simple. It has no mcu, just shift registers and driver bits. I don't know where you read that each has an mcu. The whole idea of this driver board system is that it's mix and match to some extent.

                The solenoid driver board has 16 outputs. It should drive any pinball solenoid. It's only about 3" x 4" so quite compact. No transistor upgrades should be needed as the logic FETs I've chosen are good for 14A. There's a slightly more expensive one that's good for 28A continuous.

                The lamp board will be an 8 x 8 matrix. (64 lamps) Depending on how the PC interface board design goes it may be driven from a different port. At the moment it's being designed with the same input as the solenoid board.

                Mounting the driver boards under the playfield is not necessarily a bad thing. It depends on a lot of factors but yes, first choice would be the back board area.
                Last edited by David_AVD; 13th October 2008, 07:15 AM. Reason: Pressed send accidently!
                "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

                Comment


                • #9
                  If this is the style of board you are using nuggy:

                  http://membres.lycos.fr/regismalt/PC...Schematics.gif

                  Then what David is suggesting is a very similar design, but with the option of a USB interface. There are no smarts on the pinmame HW board. The smarts David is proposing for the controller boards is primarily to handle the USB protocol. Some bonus features (with having a small controller on board) that may become part of the design could be a watchdog timer to prevent locked/burnt coils/flashers in case the host machine crashes...

                  If people don't want a PC running their pinball, then a small MCU could easily talk via high speed serial to the controller board, skipping the USB interface components.

                  If the high speed serial interface is a concern for reaction speed e.g. bumpers (not sure how fast these need to react!) then like the early days of SS machines they could be reflexive (switch themselves directly, but register a hit to the pinball controller).

                  Its all interesting to me...

                  Cheers
                  Jacob

                  Comment


                  • #10
                    ummm (preparing to be shot down and flayed). do you need to build the highway to get from A to B? is it possible to just borrow a car and use an existing road?

                    I always thought a good way to tackle this is use an existing pinball driver board - like a pinled 'williams' 90's board. it would handle all of your driver and power management (you can just use an existing pinball transformer). all you got to do is talk to it with your own instruction set. given there is bench test equipment for these boards then the 5v logic that drives it would just become output from your controller (PC, micro-controller etc). HUO means no copyright issues..

                    your 'inputs' are direct to your controller anyway, if using a PC as the brain then it would be handling display and sound anyway. just a thought kill me quickly if need be.

                    Comment


                    • #11
                      Originally posted by stuba View Post
                      ummm (preparing to be shot down and flayed). do you need to build the highway to get from A to B? is it possible to just borrow a car and use an existing road?

                      I always thought a good way to tackle this is use an existing pinball driver board - like a pinled 'williams' 90's board. it would handle all of your driver and power management (you can just use an existing pinball transformer). all you got to do is talk to it with your own instruction set. given there is bench test equipment for these boards then the 5v logic that drives it would just become output from your controller (PC, micro-controller etc). HUO means no copyright issues..

                      your 'inputs' are direct to your controller anyway, if using a PC as the brain then it would be handling display and sound anyway. just a thought kill me quickly if need be.
                      The interface between the williams MCU and the driver board is pretty complex compared to the proposed design or the controller I think nuggy is using.... The replacement designs are not very complex, nor should they be very expensive. While HUO is the intent for now, what happens if other people want to play with one?

                      Comment


                      • #12
                        i modded the board so the last lamp output is used to toggle the watchdog so as to avoid the lockon due to crash
                        is done as simply and still no usb interface! (i disenable the board, not reset the PC but)
                        sigpic

                        "Oh my mistake..you are wrong."

                        Comment


                        • #13
                          Originally posted by nuggy View Post
                          i modded the board so the last lamp output is used to toggle the watchdog so as to avoid the lockon due to crash
                          Can you elaborate on this please? I don't follow what you mean. Thanks.
                          "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

                          Comment


                          • #14
                            OK, here's something else to think about. There are two ways to drive the lamps:

                            1: Matrix (say 8 x 8)
                            • Pro: Simpler and smaller PCB design
                            • Pro: Less wires from lamp area to PCB
                            • Con: One diode required on every lamp base
                            • Con: Matrices can be confusing to wire and debug
                            • Con: Wiring needs to be thicker


                            2: Dedicated output per lamp (say 64 outputs)
                            • Pro: No diodes required
                            • Pro: Easy to route wiring to individual lamps
                            • Pro: No issues with flicker due to speed of data updates
                            • Pro: No issues with blown lamps if a row or column gets stuck on
                            • Pro: Wiring can be thinner (offsets more wires to some extent)
                            • Con: Larger PCB design (biggest part is the connectors)
                            • Con: More wires from lamp area to PCB


                            I'm leaning towards solution number two now, as I can see how a matrix would really confuse a lot of people in both the building and trouble shooting stages. Can I get some feedback on this please?
                            "Everyone's always in favour of saving Hitler's brain. But when you put it in the body of a great white shark, ooohh! Suddenly you've gone too far!"

                            Comment


                            • #15
                              im considering giving my system away or open sourcing to a certain degree, i cant believe you gonna go and reinvent the wheel...

                              can someone, nerdy or other, tell me what is wrong with the way ive implemented the driver solution?

                              david, its a watchdog, these need to be toggled otherwise the 'dog' is not watching, it is toggled on the last lamp drive, there is no transistor for that particular output, clearer?

                              oh now i think i know what u mean.. there is a defeat on startup until system is booted
                              sigpic

                              "Oh my mistake..you are wrong."

                              Comment

                              Users Viewing Topic: 0 members and 1 (guests)
                              Working...
                              X