(Topic ID: 242158)

DumbAss test and reproduction PCBs

By DumbAss

4 years ago


Topic Heartbeat

Topic Stats

  • 1,564 posts
  • 234 Pinsiders participating
  • Latest reply 6 days ago by Tophervette
  • Topic is favorited by 441 Pinsiders

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    View topic image gallery

    DSCF3080 (resized).JPG
    IMG_1440 (resized).jpeg
    boards_358.jpg
    IMG_1362 (resized).jpeg
    03_ua_mm_plaque.jpg
    02_ua_mm_package_inside.jpg
    01_ua_mm_package_rear.jpg
    00_ua_mm_package_front.jpg
    boards_357.jpg
    boards_356.jpg
    s11_cpu_c58.jpg
    IMG_2252 (resized).jpg
    20240112_212911 (resized).jpg
    IMG_0859 (resized).jpeg
    IMG_0858 (resized).jpeg
    IMG_0862 (resized).jpeg

    You're currently viewing posts by Pinsider crash.
    Click here to go back to viewing the entire thread.

    #223 3 years ago

    Sorry if this has been asked before, but is there any chance of a Data East Version 3 CPU board in the future? I'm not aware of anyone that makes a reproduction board. The particularly good thing about these is they cover a solid 7 years of games, since these boards are factory backward compatible with Version 2 and Version 1 games. So for example, your 1994 WWF Royal Rumble Version 3 board will also work in a 1990 Phantom of the Opera or a 1987 Laser War. That means you only have to worry about making a single type of board.

    From Clay's Data East/Sega repair guide:

    All DataEast/Sega CPU boards are downward compatible (so Version 3 could be used in a Version 1 or Version 2 game). However, it is not possible to use a CPU Version 2 board in a Version 3 game, since the reflexive circuitry used in Revision 3 does not exist in a Version 2 CPU board. Also, to modify a Revision 3 CPU board to work in Laser War (Revision 1), some modifications to the board must be performed. Check out DataEast service bulletin number 6 for this modification.

    From PinWiki:

    Version 1 and 2 Data East CPU boards are reflexive. Simply stated, reflexive CPU boards are boards where the 6 switched coil drive transistors were enabled by a switched coil switch input. These coils were typically pop bumpers and slingshots. Each assembly had its own switch, which was not part of the switch matrix, and was responsible for enabling its associated coil when closed. This implementation is similar to Williams System 11 Special Solenoid, however, unlike System 11 hardware, the MPU has no control over the switched solenoids (solenoids 17-22). Note that with the reflexive boards, the switched solenoids were limited to the slingshots and pop bumpers.

    Version 3 boards are non-reflexive. Non-reflexive boards use 6 special coil drive transistors, however, the switches responsible for enabling these coils are part of the switch matrix. Also, the addition of a 7407 chip at 11C enables the MPU to control the switched solenoids.

    Version 3, non-reflexive boards are backwards compatible, and can be used in a game which would typically use a version 2 board. Conversely, version 2 reflexive boards are not forward compatible in games which use a version 3 board. Note that with the non-reflexive boards, the switched solenoids were not limited to just the slingshots and pop bumpers.

    #235 3 years ago
    Quoted from DumbAss:

    As a follow on from the differences in the special solenoid configuration it might be possible to make a "bridge board" that bridges the Williams configuration to the Data East configuration and sits between the playfield connector and the board header (much like the special solenoid fuse board). This is a simple end user solution but the end user still needs to remember to install the board (just like the end user would need to set the DIP switches). Human input involvement = potential for error. Nonetheless ... is this a good compromise?

    I really like this idea. Here's a thought. You can install the bridge board before shipment only for customers that need it. Then list the one with the bridge board pre-installed separately. Have one listing with it pre-installed and one without it, and state in the listing what brand and games each one is compatible with out of the box. This will avoid the human element of having the end user install the bridge board. Plus, anyone savvy enough can remove the board and use the main driver board in another game of the opposite brand. That keeps the solution flexible and minimizes costs and logistical complexity on your end. Just make a run of driver boards, and include bridge boards in orders that require them based on what the customer buys. Win-win, and they are customer removable in case the customer wants to move the driver board to the opposite brand of game at some point. Heck, you could even sell the bridge board separately as well for people so inclined to convert their existing driver board.

    #238 3 years ago

    Oh wow!! How in the world did Bally/Williams and Data East not sue each other over design overlap on these boards?

    #254 3 years ago

    Whoa that's awesome! Glad to see someone besides Brett at X-PIN designing these.

    #259 3 years ago

    My experience with the box sockets shows that if you have a reproduction harness with the connectors fitted backwards, you will have to shave off the plastic notch with a razor. Or else the connector will not fit in the boxed socket.

    #265 3 years ago
    Quoted from DumbAss:

    I could guess that it's probably the same protocol used for the 128 x 32 but just quadrupled to 256 x 64 but I'm not certain. It looks like it uses a 14-pin ribbon cable so odds are good it's the same serial protocol.

    Most likely. Here's a trick I learned, you can connect a 128x32 game to a Sega 192x64 DMD and it will show the image with unused dots around it. Likewise, you can do the opposite and show most of a 192x64 game on a 128x32 DMD.

    #271 3 years ago

    It looks like people have been buying kits and building boards that way.

    2 weeks later
    #303 3 years ago

    Crazy attention to detail! I've never seen anyone make boards with that much information printed on them.

    1 month later
    #342 3 years ago

    Did you ever get a chance to trace down the logic issues with the WPC sound board?

    #344 3 years ago

    Awesome. I have a feeling the issue is RF interference on the address lines. It was much worse in my Dracula than White Water. And Victor couldn't reproduce the issue on the bench, but did reproduce it in one of his games. You also have the board I sent back for further testing.

    #346 3 years ago

    Could you put a scope on the data lines coming in to the sound board?

    #348 3 years ago

    Is that the board I sent back? If so I'm curious if the board I had is any worse than the others.

    #355 3 years ago

    I think you mentioned reproducing it in your White Water. You were just never able to on the bench.

    #362 3 years ago

    Good finds. Yes, all of our WPC machines have original CPU boards.

    3 weeks later
    #394 3 years ago

    That's awesome! I think you're the first person to manufacture a DMD using 16x16 LED blocks. It looks great. For a possible unified display panel, do you think the SMD LEDs would be more serviceable than the 16x16 through hole blocks? As if if one goes bad, how difficult would it be to replace vs. a block with potentially 32 different solder connections?

    #396 3 years ago

    So unified in the case of a single board (think Data East 128x16) instead of a display driver and display panel like with your master and slave boards? If so, will the master board be designed to replace or substitute an OEM display driver board?

    1 week later
    #411 3 years ago
    Quoted from DumbAss:

    The 8x8 blocks are 1.9mm in circumference and spaced 2.5mm horizontally and vertically.
    The 16x16 blocks are 1.7mm in circumference and spaced 2.5mm horizontally and vertically.

    Does this mean your displays that use the 16x16 blocks have a smaller dot size than the 8x8 blocks? I always preferred a slightly smaller dot size to the current 8x8 LED blocks, that more closely match the dots on the original plasma panels. Do you know how the 16x16 blocks compare to plasma in this regard?

    #415 3 years ago

    That looks awesome! Less grid lines, looks closer to original plasma dots, and is more repairable. These can also be used in Data East 128x16 displays if you ever wanted to do those some day.

    #419 3 years ago

    That's great! I believe the PinballSP DMDMK66 LED display uses both 5v and 12v to run the panel. I can get photos of this panel if it helps give you an idea on how this is achieved.

    #426 3 years ago

    Great history! Thanks for sharing. I don't think that has been explained in detail before.

    3 weeks later
    #444 3 years ago

    You are officially the new Rottendog. You've already done more than Jim has ever accomplished from what I can tell. Great work!

    #450 3 years ago

    Well these technically are prototypes since nothing is final.

    1 month later
    #472 2 years ago

    What do the little red displays on the bottom do? Never seen those before.

    #481 2 years ago
    #483 2 years ago

    I had issues with mine about 6 months ago. Im guessing the problems with the flaky address chips have been resolved?

    1 month later
    #515 2 years ago

    Glad the WPC sound gremlins are gone! Happy to help.

    2 months later
    #549 2 years ago

    Would you consider making a Data East/Sega/Whitestar GI smoother board? Craig Aker made the EnerGI Maestro a few years ago but doesn't appear to be persuing this anymore. Still plenty of interest:

    https://pinside.com/pinball/forum/topic/led-gi-smoother-for-data-east/page/9#post-6483071

    3 months later
    #629 2 years ago
    Quoted from dmacy:

    [...] There seems to be a few posers out there. [...]

    I mean if you're making pinball parts in the basement, I can see why.

    2 months later
    #732 2 years ago

    Here's a tip. Don't show up at TPF on Friday, sell junk boards, and leave on Saturday afternoon. Not that we have this issue with Dumbass of course. Keep on rocking dude!

    #738 2 years ago

    And remember not to buy from some kids (or a grown man) in the basement!

    3 months later
    #820 1 year ago

    Maybe the subtle block can be rotated and positioned where the "bad" LED is along the edge of the DMD where it would be least noticeable. Can be used in a seconds or test DMD I guess, unless you're getting rid of blocks with minor variations.

    #822 1 year ago

    You're right, I forgot about that.

    1 week later
    #825 1 year ago

    Like putting your components through a cotton gin. Lol.

    1 week later
    #855 1 year ago

    I don't see a Williams logo...?

    2 weeks later
    #896 1 year ago
    Quoted from DumbAss:

    Very brief update. More to come once I get the boards I have been working on complete and in the mail system.
    Today I did a bunch of things in a batch. Verify some boards (in a machine) and grab a board from a machine. One of the boards is a new one that adds another to the opto testing arsenal.
    The Pinwiki subsection @ https://www.pinwiki.com/wiki/index.php/Williams_WPC#7_Opto_Board_Testing describes how to test the LM339. It's a fairly simple procedure and it works great. I already have a board that identifies problems with the opto transmitter/receiver pair but I don't have one that identifies problems with the LM339. Well ... not any more. This board plugs into opto testing system that I have and therefore requires the other boards and 8P8C cables.
    This is the new board. It's pretty simple. A bunch of switches that basically takes the place of the receiver.
    [quoted image]
    This is how it was tested and is supposed to be used in a machine.
    [quoted image]
    The instructions are in the silkscreen but essentially in switch test pressing the button should cause a response from the software indicating switch closing and opening.

    There's a typo, lol. "Detction."

    #899 1 year ago

    It's no big deal, at all. I wouldn't notice if I bought it to be honest, and wouldn't care the least bit.

    #909 1 year ago

    Speaking of modern components, do you know if the EZ-something 5v switching regulators draw less power on the 9v AC pins of the WPC power board secondary connector? I think those are discontinued but I bought a few of them ahead of time. I currently have a game with a factory LM323K and soldered in secondary wire harness and would like to fix that header/connector correctly without the 9v pins overheating again.

    #914 1 year ago

    Simple fix then! Thanks.

    1 year later
    #1473 69 days ago
    Quoted from DumbAss:

    For the final update (separate posts to keep things compartmentalized), I have an experiment that I have been working on. I started with a simple implementation for static settings (a string of LEDs). It then progressed to a more complicated implementation (a matrix of LEDs).

    It's a PWM (pulse width modulation) implementation. Essentially, LED dimming. The current popular product is too expensive for my taste. Plus, I love a challenge. This is the start of implementation of that challenge.
    The pattern can be programmed. It's a pattern of ON or OFF. The algorithm only cares about if the (eventual) target is ON or OFF and it adjusts the brightness of the LED over time to reach the target. This is actually different from the simple string which implements a scheme more like GI (8x fixed brightness settings rather than changing brightness over time). I have tried a few different patterns to test the algorithm and it appears to be working. The software logic is contained entirely in the ISR (interrupt service routine). The pattern programming logic is performed outside the ISR.
    The next step is to intercept the WPC signals and write the software that interprets the signals and assigns the ON or OFF state as a result of the input. The initial implementation will deal with the raw WPC signals. This will eventually target an integrated solution to the Power Driver board as an option. It is also possible to scan the output signals (J133-J135 and J136-J138) and use that information instead of the raw WPC signals.
    Eventually, I'll consider doing something for System 11 (3-9 as well) for both GI (relay control) and lamp matrix (either raw signals as an integrated solution or output signals).
    I honestly thought it would be harder. It was a pleasant surprise. The smoothness of the LED dimming (increase and decrease) is obvious in the video and (IMHO) produces a nice result.

    Very nice! What scan rate are you using for the matrix?

    4 weeks later
    #1511 40 days ago

    Awesome post. May want to fix your images though, the same success photo was inserted 4 times instead of the error photos you meant to include.

    #1513 40 days ago
    Quoted from Kawydud:

    look closely at the numbers, all pictures are different.

    Sorry, they are not on my end. Dumbass which photo shows the "INCORRECT U22" error? May just be Pinside?

    You're currently viewing posts by Pinsider crash.
    Click here to go back to viewing the entire thread.

    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/dumbass-test-and-reproduction-pcbs?tu=crash 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.