Sorry. Not yet. Still in development. Keep following along. My next board fabrication order will probably include these boards for portable diagnostics.

And I agree that diagnostic tools is time and money well spent! If you cannot determine the cause of failure you cannot succeed.

This is indeed a problem. I knew this would be the case but couldn't think of a solution at the time of laying out the connectors. The issue with these connectors is that it is a very tight fit in the are of the board headers and there's not a lot of room for large boards. That's the X and Y axis. When you factor in the Z axis (the TIP102 and TIP36Cs) it adds another dimension (literally) of difficulty. I have since thought of a solution and will need to re-work the boards and have new ones made. This will add time but that's the cost of development. It's part of the reason why I created this thread so that people can see that development is not quick and easy and a reason why people just don't bother with this kind of stuff.

All feedback welcome. The more detail oriented the better. 80% of development is done in 20% of the time. The remaining 20% of development (the polish) often takes 80% of the time.

True and valid points. The power connector actually is a tap for +5V and GND. The diagnostic panel needs the GND reference for the switch side of diagnostics as well as for the state indicator LEDs. I can see two possibilities for accommodating your suggestion. The solution can involve both methods.

1) Provide a 0.100" connector for power - +5V and GND. This will require crimping a connector with the appropriate length of wires.

2) Provide turret test points for alligator clip connections for +5V and GND.

Say "Hi" to Welby for me. Haven't seen him since he stopped contributing to the Northwest Pinball and Arcade Show.

I tested these with a hard drive power brick last night and they provide the correct voltage. I want to test these in a machine to make sure. Then I need to discuss with my friend what he wants to do. These were made for him and I don't want to infringe on his turf. We'll be meeting up in the near future. Hopefully I'll have it tested before then but otherwise we could possibly test in one of his machines.

As a reminder I do provide fully assembled (complete) boards but there is lead time as I don't want to spend time keeping stock available when I could spend time working on the next project. I will stop and build boards if there is a request for one. For these boards I will need to update my spreadsheets with BOM information and then transfer the BOM information into a table in a document. This BOM is small enough I could spend some time trying to figure out Digikey or Mouser BOM lists.

Two weeks since the last update. Sorry. I did mention I'm not fast. Boards (hardware) is nothing like software. Software is good for rapid turnaround. The edit, compile, link and test cycle is quick. The schematic, layout, connections, fabrication and test cycle is not so quick.

Please remember that this thread is also about the journey. I will attempt to document the successes as well as the failures to provide an idea of what's involved in developing these things. It's not always straight to solution that you might think. This reminds me of my days past in the software debugger. People only saw the final determined cause of failure without seeing all the incorrect paths that had been taken before arriving at the correct path. It was often 5+ paths of failure before the 1 path of success.

So after the initial attempt for portable testing (with the coupled but not fastened together) I had a thought of a way to couple the two boards so they could be removed as a single board rather than two separate boards. This is what I came up. We'll call this the "double" system.

boards_081.jpg

As I was fond of telling people in my old job ... creativity does not happen on demand. Sure enough after a few days or so I found another (I think much better) solution. We'll call this the "single" system.

boards_082.jpg

I finally got those boards in. Last night I populated them and today I went to test them. I decided to use System 11 as the test bed this time instead of WPC but the principle is the same.

The power tap for System 11 is a little more necessary than the power tap for WPC.

boards_083.jpg

The solenoid drive. In this case it's solenoids 9-16. The double system actually doesn't work here. It sits too far below the connector and the interconnect connector interferes with it. The single system works fine so that's another point in favor of the single system.

boards_084.jpg

The special solenoid drive and trigger. The trigger is the double system - just for comparison.

boards_085.jpg

Closing the speaker panel to see how much space there is. Looks like there's JUST enough space with these prototype single boards.

boards_086.jpg

And this is how it will work with the speaker panel in place so that you can see the display as you're running the tests.

boards_087.jpg

And finally this is the system working. The diagnostic board is the old board that I made over a year ago. I have the newer diagnostic "panel" waiting for fabrication.

boards_088.jpg

So I think this system works and will work in WPC. I'm going to test it in WPC as soon as I can just to be sure but I see no obvious reason why it won't work. After that it's on to fabrication.

I have individual boards for each of the connectors that need a connection for diagnostics for solenoids, switches and lamps. It's a LOT of boards but the fact that each connector has a different key means that there needs to be a different board for each connector (for the most part). I will color code the boards based on the base color of the wires that connect to the connector. For example

- switch columns are GRN-XXX so the board will be green.
- switch rows are WHT-XXX so the board will be white.
- lamp columns are YEL-XXX so the board will be yellow.
- lamp rows are RED-XXX so the board be RED.

I have a limited choice of colors so it sometimes won't match but I will try to make a reasonable choice and also take into account overlap (solenoid boards will be RED due to overlap with lamp rows on System 11, special purpose boards will be blue and power boards will be black).

boards_089.jpg

Digital images typically don't do blue justice. It's often over-exposed and this phone camera doesn't really allow me to control any of the usual parameters available on a DSLR.

The board shown is the initial slave display board. I built it with SIP sockets to allow me to swap LED blocks in for testing purposes. An actual display slave board will look different (the digit blocks and the connectors are not present in a board used in a machine).

Some boards came in the other day (as well as those blue blocks above). I haven't done anything with them other than take a picture of them and pull the components needed. I have requests that need work work and those are higher priority. I will get to these ... eventually.

This is the WPC set. I have also the System 11 set but I haven't looked at those at all.

boards_090.jpg

I don't believe it's been in this thread. There was a second thread started the other day asking about this with the exit of Rottendog. For those interested in purchasing one immediately apparently Big Daddy bought the remaining Rottendog inventory and has them listed in stock on their website.

I hadn't given the Data East CPU board much thought. I know that Williams System 9 boards are also about to get more difficult to find as I don't believe anything other than the Rottendog MPU9211 serves this market. The Rottendog MPU004 serves the Data East CPU market and the board overlaps with Williams System 11C.

The Data East CPU board was the first big board that I cut my teeth learning repair on. I had damaged about 1/4 of the ICs on a friend's board and was more or less forced to learn how to repair it. It took me months of research and learning. In that sense I am familiar with the board and its layout. The prospect of reproducing this board is not so daunting to me.

I will need to pull a spare board that I picked up and repaired from eBay (about two years ago) and make an assessment. I believe the board I have is revision 03. I would like to understand the differences in the revisions. This pertains to differences between Data East CPU revisions and differences between Data East and Williams CPU boards as I will probably use the basic Williams schematic that I have already captured and work in the Data East differences. I don't know what path will be chosen until an assessment is made. When it comes to opening up big boards for modification or re-work I'd like to have my ducks in a row because these big boards aren't cheap to make (time and cost).

If a decision to proceed is made the board will likely not be available until the new year. There is development and testing time and I already have a bunch of other things going on (issues to investigate - your reported issue - and boards that are currently in the works that need verification).

A demand check or interest gauge (people posting interest) would improve confidence that time invested would be worth it and raise priority for any potential reproduction (actually of any proposed board not just the Data East CPU).

Thank you!

I started that board this year in early January when I was visiting family overseas. It's remained about 90% complete. The improvements on this board are to shift the ASIC to a daughter board so you don't have to wreck the board with hot air rework if you want to replace the ASIC as well as a 68-pin PLCC socket for the DSP. To make space for the ASIC connectors I removed the NBA Fast Break linking (really the debugging features of the board) to a separate daughter board with connectors.

I understand that some Rottendog WAV095-1 stock remains available. I see it in stock at K's and Big Daddy.

I do want to do this board for my own self satisfaction and can raise priority if there is demand.

I actually did have one of these made but never posted it here. The board is a unified (single) board rather than the master/slave (double) board system. I needed one for a Big Guns project I picked up without a display. I did a test fit in the speaker panel and it seemed to fit fine. The board works on the bench. The connector positioning is not identical to the original display board but it is compatible.

Note that this board is not the same board as the one used in Millionaire and early production F-14. I have a small number of this board (not the Millionaire board) available as of writing.

As a reminder I also have the full 16 block upper/lower alphanumeric display for most System 11B games (including Taxi and Police Force).

LED block colors available are: orange (amber), red, green, blue and white or any combination that you want. If you really want to be able to change colors then it is possible to install SIP sockets and change the LED blocks to your desired scheme. Amber and red are included in the base price. Green blocks are an additional $10. Blue or white blocks are an additional $20.

The quad display (DBU-PNL-QUD) is:

- Bare is $100 (includes the LED blocks as they are not carried by the big box electronic merchants).
- Complete is $160.
- Shipping is $15. It just fits in a medium size box.

Does NOT include:

- Power LED indicator unless requested. This is just light interference for mostly no useful purpose.
- Any light blocking foam. I have a lead to follow on this but I just haven't had the time to work through it.
- Mounting spacers. I have a limited supply of these. I encourage reusing your old ones. New ones can also be purchased from Pinball Life.

Build and test time would be about two days. Board shown below is white (the yellow segments is the giveaway that it is white).

boards_091.jpg

Last night I spent a little bit of time looking at the Data East CPU board schematic to build a plan for the board. It's obviously 90%+ identical to the Williams System 11 CPU board but there are differences. I have ideas for an approach but want to solicit opinion because I can try to think of reasons one way or the other but the truth is that the more people who have thoughts and inputs the better the solution will be. I don't believe in "I alone can build it" but I do believe in "I alone will build it and be responsible for it".

I intend to use the System 11 schematic capture that I have due to the identical 90%+ electrical connections. There are few things that fall out from this.

1) Do I somehow try to make a "new" universal board that supports both Williams and Data East circuitry in a single board? This is the Rottendog approach for MPU004. It has pros and cons. One pro is a single board to support everything so no need to maintain separate board inventory or potentially component inventory. One con is that the EPROM size jumpers and address line connections are different between the Data East schematic and the Williams schematic. Another con is it will require configuration settings meaning that the user may be required to change settings and any time a user is involved there is potential for error. I know that MPU004 has DIP settings for the switch solenoids and this is obviously a problem because it's documented on the PinWiki page (https://www.pinwiki.com/wiki/index.php?title=Data_East/Sega#Special_solenoid_problems_with_RottenDog_MPU_board).

2) Do I make a separate Data East only board that supports the Data East jumpers (J1-J7) and the Data East special solenoid configuration? This minimizes confusion to the end user but means I have a dedicated board with separate inventory.

3) 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?

4) If I do go down a complete separate board for Data East do I keep the Williams physical layout that I already have and make modifications to the physical layout to support the schematic changes? Or do I scrap the Williams physical layout and reproduce the Data East physical layout? Keeping the Williams layout will make time to market MUCH faster because I don't have to re-layout 90%+ of the components and traces. Reproducing the Data East layout will make working on the board easier for those who only have experience with the original Data East layout rather than having to get familiar with the Williams layout and potentially translate between the two different layouts.

Hmmm. That's a lot for the reader to contemplate. First world problems. <sigh>.

Any input, feedback or comments are welcome. In the meantime I still need to work through the Data East schematic to identify and collate the differences between it and the Williams schematic to make sure I have a solid foundation to build on.

I don't have a store set up (yet). Send me a PM with what you're interested in and we'll go from there.

Interesting historical progression ...

Williams introduced the A/C switching (multiplexing) of the solenoid drive transistors. Williams initially assigned the A-side (relay de-energized state) to the solenoids and the C-side (relay energized state) to the flashers. In the event of relay failure at least the solenoids still worked and the game could be played (without the additional light show).

There was one issue with this in that there was no feedback to the software which side was energized. A software intent to fire a flasher and relay failure without C-side power engagement would result in a solenoid firing instead of the flasher. Williams fixed this with the additional 4N25 optocoupler on the interconnect board. Introduced after Taxi.

Data East on the other hand initially implemented the A/C relay by putting the flashers on the relay de-energized state and the solenoids on the relay energized state. Presumably to avoid any potential lawsuit about copying. The problem is that failure of the A/C relay causes the machine to just sit there and flash its flashers instead of firing the solenoids. The game cannot be played.

Williams introduced software controlled flippers with its "fliptronic" technology. This allowed the end-of-stroke switch to be changed from a high voltage normally closed switch to a low voltage normally open switch. The advantages and disadvantages are beyond the scope of this post.

Data East also introduced software controlled flippers with its "Deger" system. These are the Data East flipper boards. If you have one in front of you then you can witness something that still gives me a chuckle every time I see it. It is also shown in the schematic. Do a web search if you want more information.

ty-ffasi.jpg

Firstly ... thank you for the information. It helps tremendously. The following is argument for argument's sake not a criticism. As I previously mentioned I want to have all my ducks in a row before embarking on another big board. I did this with the other big boards that are complete but I didn't do it in a public "thought process" like this one.

There's actually a few more differences. The DE board uses a reset generator, allows for use of an 8MHz oscillator with a frequency divider, has a 74LS126 for address line selection and has different address line access to the high parts of the EPROM address space. The DE board supports the use of a fully addressable 27512 EPROM at 5C whereas the Williams board only support addressing up to 27256 for the EPROM at U27. Both boards support a 27128 in 5B or U26. The jumper selection is different and the DE schematic does not document J5a/J5b nor J6a/J6b. I will have to look at a physical board to figure out what those jumpers do.

It is largely the same board though. The EPROM addressing support and differences mean that a DE board will be usable in a Williams machine but a Williams board may not be usable in a DE machine. Hence why you're able to use the DE board in your Rollergames. Nice bridge or translation between the solenoid drive wires you made there. Impressive!

My System 11 board uses a reset generator. I got rid of that reset circuitry and reclaimed the space (along with the battery holder) to implement the "C" of the NLA SRC components.

After staring at the schematic and noting more differences I'm leaning towards doing a completely separate board for DE. This will take a little longer as I will probably do the layout the same as the DE but the biggest thing for me going this way is that the schematic in the DE manuals will be usable. There won't be a "where I can find the Rottendog schematic for MPU004?" equivalent. If the board uses the same schematic and same layout for all the critical circuitry function that question is moot. The boards will outlive me as long as the manual and schematic is available.

I would consider building a bridge board for the solenoid connectors to allow the board to be used in a Williams System 11C machine - but only System 11C due to the lack of sound requirement for those machines.

For some reason ... Friday seems to be board arrival day. Sometimes it's Thursday. Probably related to the day of the week I tend to get orders for boards placed. This latest batch contains 3 separate categories but this post will only go into one of them. I have some other outstanding work to complete before I can start focusing on the other boards. I like to know what the state of boards are because if they don't work then I start a background priority thread trying to figure out possibilities for reasons why they don't work. While those threads are simmering in the background I'll probably go back to the DE board. I suspect the DE board will be classic example of exponential growth (or progress). It will start slow as I chip away at it and when I have a plan and approach it will take off. Very quickly.

So what do we have this time. I should point out that these boards have been in the works for a while. I don't want it thought that these boards had higher priority than any other potential board. The reason they're here and now is because I got the boards made and they're in. My expiring mind wants to get them assembled and find out if they work or not.

The Indianapolis 500 Illuminated Target Board. That opto interrupter seems to cause some grief. The original board (still with mounting bracket) is on the left. The first reproduction with compatible 3mm opto is next. This board has been sitting around and I just put one together as there is a request for one. It passes on the bench and produces the correct results.

Ask and you shall receive. Sorry. That was being a little facetious. Here it is though. The two right boards use a 5mm opto and also isolate the opto from the switch matrix wiring. I had thought about putting an indicator LED but I already had to make the board a little bigger to support the transistor that isolates the opto and I didn't want to make the board any bigger. I made two versions. A "narrow" and a "wide". The narrow version has mounting bracket hole separation the same width as the original board. You should be able to re-use your mounting bracket. The hole to opto spacing supports a #4-40 machine screw head. The wide version requires a new mounting bracket. I'm working through this with another Pinside user.

boards_092.jpg

Closer image showing the new board differences. This will be cross-posted to other threads where interest has been expressed. I know this thread can be a tough read for people.

boards_093.jpg

And finally this board is the Position Encoder Board (A-20533 or A-20533.1-1). The board is used in Johnny Mnemonic and The Champion Pub. I know. You're asking ... why this board? Well ... the request for this board was made on October 24th (about 4 weeks ago). I just chipped away at the board and went through my normal process and it finally reached the complete reproduction stage. About 4 weeks or one month. Spot on my original estimate. This board also uses the new 5mm opto.

boards_094.jpg

None of them are tested but they'll get tested on my bench very shortly. I have very high confidence because I already built a prototype board and used the opto in said prototype board. These final boards use the same circuitry so there's every reason to believe that they will work.

Probably this.

For the LED the longer lead (although I'm sure you cut already cut the excess off) is the ANODE ("positive") and the shorter lead is the CATHODE ("negative"). The positive hole on the board is square. The negative hole on the board is circular (standard). The datasheet shows the dome has a dot or notch indicating the cathode end of the LED. That should line up with the "flat" end of the LED silkscreen on the board.

You can also check by doing a diode test on the LED. It should light up. The configuration of the leads from the DMM should confirm which is anode and which is cathode as you've installed it.

As you've discovered 3x of the 470 Ohm resistors are the current limiting resistors for the LEDs.

The other 6x of the 470 Ohm resistors are pull-down resistors. The signals are tied directly to ground. The lower the resistance the stronger the pull-down. The signals are fixed so it makes sense that the board still works.

Glad you got it figured out and working!