I would not be too difficult to engineer a replacement MPU board for a Zaccaria
The board would basically be a copy of the original design (minus any hard to find parts that would need to be upgraded/re-engineered).
Then you would put the original EPROMS in, install the board and Voila...

Not undoable, in a nut shell.

Sure, it's possible.. it would just take someone that really likes Zaccaria and doesn't mind the somewhat limited audience.. at least in the US (compared to Bally/Williams homebrew boards).

---
http://www.pinitech.com - "Pinball Inspired Technology"
Kits, upgrades and test equipment for pinball machines

I took a brief look at the schematic and it actually looks like a really good case for modernization. Lots of 74 series style logic that could be wrapped up in CPLD or FPGA. The major down sides I see are the limited market and the oddball super rare CPU that's used. I couldn't easily find a 2650A CPU core for FPGA. I'm guessing there isn't one because it's pretty rare and would be in low demand. If a VHDL CPU core were available it would make the project a LOT easier.

Interesting idea. Not something I'd personally pursue.

Someone sell me the gerber files for any of these boards. I could probably make it work. I would have to assemble a few and see if it is worth it. If i had Gerber files for bally boards i know i could make that work. Same with WMS sys 7.

Then someone get on a Stern SB-300 right after that.

I'll do it. $10K/board including working prototype and gerbers based on original designs (no addition of CPLD, etc...). That may sound like a high number but honestly I would still be somewhat hesitant to not just sell the finished boards myself. You're never going to make that kind of money from Zaccaria, but you can be sure that a company like Alltek has made multiples of that selling the Bally stuff. I have finished designs in the can for Bally/Stern and Williams 3-7. What I don't have is spare time in my life to actually prototype them and probably won't any time soon.

Or just do it yourself.

That's what I would really recommend. If you're looking to work from gerbers in the first place, it's not tough to take the schematic, dump it into Eagle and create a board. The tough part is in the component placement and routing, but people seem to be happy with boards based on the layout of the originals so it's not tough to get it to that stage. I would be happy to give you advice for free on any of that stuff. I'm not the best circuit designer ever, but I've been designing PCBs for pinball for over ten years and I've done lots of boards for pinball in that time and come a very long way in terms of component layout and routing. I love designing the boards. I just don't love assembling and shipping them so most never see the light of day. I don't really see someone coming up with well designed gerbers for pinball MPUs that they're going to sell "cheap" unless it's not their original work because there is actually a lot of profit to be made in some cases (not Zaccaria, IMO), but who knows. Stranger things have happened in pinball.

Personally, I would not consider hand assembling them using through-hole components like the originals. That's going to take FOREVER and eat up all potential profit.

http://www.cadsoftusa.com/eagle-pcb-design-software/about-eagle/

Same thing I was thinking but it would probably be more than $10K. If somebody took time for the non-trivial effort to create gerber files - then they would already be motivated to selling the boards themselves. Getting a copy of the schematics is the easy part - going from schematics to debugged gerber files is quite an effort. Especially if you do it right with trace optimization, copper pours, etc. Toss in obsolete part replacement and you have a major time and effort expense.
A good PCB software suite is going to run you a pretty penny. It can be done with the free stuff but not easily. We invested more than $50K in our Mentor Graphics design software -- per user.

I like those "ultimate" board designs where the installed part is SMD but the thru-holes are next to it for conventional repair.

Just always looking for more ideas for projects. 10k... wow. I know I could do it myself if I devoted the time. I already have a full time day job and i do board repair 3-4 hours every day, and i am trying to sell components out of the house, and build nvrams ahhhh i gotta get back to work.

Easy answer -- quit the job

Zac (and Gameplan too) actually did copy the Bally design for most of their systems. The lamp drivers are the same on all 3 , coil drivers are the same on Zac's, The Zac displays are super similar including the power supplies. Other than going with the 2650 over the 6800 they are super similar

I like how bally pulls off there setup with only two PIAs. WMS used 4 and even more on later systems. Encoded lamps and individual scr worked out way better.

I've always liked the way Bally designed their system too. The way they multiplexed the PIA outputs of U10 is really an efficient use of the PIA.

It's efficient until you have something shorting on the switch matrix that causes the displays to output garbage. Discrete outputs for the PIA's make it super easy to diagnose a problem.

I change waaaay more individual SCR's that I do TIP42 transistors. I don't think I have ever shorted out a Williams lamp matrix while changing bulbs. The Bally's were superior when it comes to a non-booting or partially booting MPU board, but Williams ultimately did better with the rest of their system (other than the burning GI connectors that dogged them from 1980-1999)

The lamp matrix was a bad choice for wms in the early days. Having the CPU have to deal with the lamp matrix forced the software to be much simpler. Bally software / rules smoked wms until wms system 9 came out. The Bally hardware layout was clunky at spots (aux lamp board etc) , but really gave then much more flexibility than wms did.

I remember when I got my Farfalla it was missing boards. I think there was an Australian outfit that made a new MPU basically duplicating the original. I instead contacted Dave Gersic and he hooked me up with a set of rebuilt originals which run rock solid.

Look out everyone... I am getting dangerous. Starting with redesigning simple PCBs before I am getting more complex. 10k for a mpu layout.... zzz. I will get there soon

c8ca468e97030d38b1718620923e8130-11458_top (2).png

10621479eb8cd83583586af2ea5122e6-11403_top.png

Cool stuff! It's a good idea to add LED support to the lamp driver if you're planning to sell them. Here's a proof of concept layout I did of the Bally lamp driver board as an example of a different approach to that board. I used SMD SCRs but DIP would fit in the same space and basically route the same:

Aux_Lamp_Driver_Example.png

For the 5101-6264 adapter, the 5101 package will fit under the 6264 if you use the smaller pads on the 5101. That would make the board quite a bit smaller, relatively. I've given examples on the forum for combining the enables of the 5101 to add Gottlieb System 80 support and isolating the inputs and outputs of the 5101 with a 74LS244 to add Gottlieb System 1 support. It's worth considering adding that stuff to expand the potential market.

I've also been considering adding a jumper to my 5101 and 6116 NVRAM adapters for two banks of RAM. I think some people would like that. In theory, you could then run a switch to the coin door to switch between two sets of audits, high scores, etc... or just leave the jumper and switch it that way. That's something else to consider adding.

An MPU is a different universe than these small boards in terms of routing and prototyping, but it's not rocket science. You can definitely get there with enough time and effort.

A friend of mine has developed a replacement CPU for Farfalla - I'll see if I can get him to chime in here with details.

Thank you for you comments!

Things I had thought about for aux lamp driver board... Additional pads to support SMT and through hole. Resistor networks to keep enough load on the SCRs to latch them when using LEDs. Maybe a LED block to show on the PCB that each lamp circuit is working.

The 6264 to 5101 adapter, I didnt think the 5101 footprint would fit inside of the DIP28 sockets i have, so i staggered them. The hole size was chosen by the component library. I wanted them to be both the same, but I rolled with it. I will probably try and use fm1608 on these. I still havent figured out how to make them work with boards that use the 2nd chip select more than tying it to be always high with the power on reset delay. So they only work with WMS 6-7 and Bally/Stern -17 to -200 era. I will have to look up your posts.

Some issues i had was finding all the components. Eagle was better, but I will have to stick to KiCad to sell these things. Like the SCR i used wanted to drill mount holes in the lamp driver board and finding a 20 pin .156" header. I ended up butting together two 10pins. I will figure that stuff out in time. I assume there is away to make custom components. Time to watch more youtube videos... doot doot.

LOL

You basically have to design a lot of your own libraries / modules in Kicad.. yes there are ways, check out the library editor / module editor. It's not totally straight-forward, a bit quirky actually.. but eventually you catch onto what needs to be done to build anything you want. I've created many components not in the standard libraries this way out of necessity. You start getting used to having to look at the component specs (say LED dot matrix blocks for example) and using x, y coordinates to set the spacing between the pins. Because the standard libraries could never work for everything. It's tedious at first, but then you get used to using a combination of what's available and designing them from scratch.

Like the .156 footprints.. if there was none in there, make a pad.. then put the next pad at the same Y-coordinate and 3.96mm off the 1st pad's X coordinate. Then do the same for each other pad.. and start saving them as MOLEX_156-2, MOLEX_156-3, etc. Personally I just build what I need.. so I don't have all the bases covered (ie.. I might have a 9-pin and 10-pin MOLEX 156 footprint/module but not a 7-pin.. till I need it). I can always delete or add pads to an existing module I created to get what I want.

It's fun creating new boards out of your own ideas. Not just copying what's already out there to make it cheaper or just using other people's ideas you find.. but actually thinking up something not out there yet, being creative.. and seeing something you didn't know was possible actually work out. Well that's the exciting thing to me with the boards anyway. The only downside is, the more you create.. and offer for sale the more it can take away from ability to design new stuff. So you kind of want designs that can sell in high volume (I'm still in search of one of those myself) and yet in the back of your mind know it'll mean you won't have as much "free time" for new designs.

Good luck and have fun creating

---
http://www.pinitech.com - "Pinball Inspired Technology"
Kits, upgrades and test equipment for pinball machines

I have made him aware of this thread....

Good ideas. The basic design I posted has the load resistors for LED compatibility as SIP resistor networks. I was also planning to add indicator LEDs to the SCR outputs. I've been designing a lot of these smaller boards lately using as much SMD as possible because I would like to eventually get a pick and place machine. SMD LEDs would be a lot cheaper and you can place them right by the SCR, but the DIP bar graph LEDs would also work.

The best advice I can give you is to learn how to make your own packages in whatever design software you're using right away. It's one of the most important things, IMO. When I do stuff for pinball now I end up just making most of the parts myself because it gives me the most flexibility.

Here's an eagle library with a 5101 using the smaller holes, as an example.

http://pinforge.com/Eagle_Example/5101_Example.lbr

I really like Eagle, personally, but it's the only one I've ever used so I have no frame of reference. It works well for me so I stick with it.

I assumed you were planning to stuff a 6264 NVRAM adapter in the 6264->5101 board you posted which made me wonder why you didn't just do a NVRAM-> 5101 board, though there's probably a small market for people who want to use a 6264 chip in a 5101 socket with batteries.

I wasn't easily able to find them. I'll repost later when I have more time to dig up the schematic examples.

Small market for sure on the 6264 to 5101 adapter, did something like that myself and there wasn't a huge amount of interest.

Shimoda, Woz let me know that you are interested in a replacement Zaccaria CPU board. I designed replacement boards for Bally and Zaccaria cpu boards a few years ago. The boards use a common microcontroller and Pinball Operating System, but the game code has to be written in C for each game. This is easily adapted from Visual Pinball or similar simulator programs. So far the system has been used for Bally Paragon (me), Zaccaria Farfalla (me), Bally Power Play (woz), and Zaccaria Pinball Champ (me, partially done). The idea was to provide a standard replacement hardware and software system for all machines made from 1977 to about 1986 (the 7-segment display era). The attached pdf is a bit dated but illustrates the concept. If interested, contact me.

Hi Shimoda,

Sorry for the delay, been a while and I am collecting the pieces. Attached are pictures of:

a) The Baby Zacc board installed in the machine
b) The Baby Zacc board with the Microcontroller Board attached
c) The Baby Zacc and Microcontroller boards side by side

To use this board you will need:

a) The 9S12D or 9S12X board (from technologicalarts.com, ~$100)
b) The Baby Zacc board (from me, or ExpresssPCB, ~$20?)

To improve my Farfalla program, or add programs for other Zaccaria machines, you will also need:

c) The P&E Micro Multilink Interface (from pemicro.com ~$200)
d) Freescale CodeWarrior development system (was free...)
e) 'C' programming skills
f) The source code and PbOS library from me, ~$0)

Andy

IMG_0027_BZ_Installed.jpg

IMG_0033_BZ_Boards_Attached.jpg

IMG_0035_BZ_Boards_Separated.jpg

Yes. I used the 9S12DP256 and 9S12DP512 before, but the 9S12DG128 would be fine.

As an aside here's Andy's board at work in my Bally "Power Play". Well.... it's the same tech arts board but with different code and an interface board to the Bally hardware.

Sounds like fun im in