I am very new to this group and am so grateful I have. I was wondering if anyone is willing to offer their Kicad skills to help me recreate an old logic board based off of a schematic? I have no clue where to start in reading schematics but am hoping to create an old pinball board that is no longer being manufactured and is nearly impossible to fin. I’m trying to bring this poor machine back to life in any way I can. Anyways, If your interested, we can discuss payment. Please take a look at the schematic and see if it can be done. I would need the logic board and sound board. thank you for you time! Please feel free to PM me or repy here. thank you!
That looks like it could be a fun project for an aficionado of classic pinball equipment! (I suppose there are some aficionados who insist that “classic pinball machines” use electromechanical relays, rather than integrated circuit logic gates.)
After flipping through your document for about 90 seconds I have the following comments and observations:
The original printing, and subsequent photocopying, have compromised the document’s legibility. I didn’t spot any cases where the original text or graphic can’t be deduced if you squint hard enough and use a little magnification . . . but there may be a place or two where this isn’t the case. (Or, worse, the smeared pixels resemble something very different from the original intent, such as a " 3 " that looks for all the world like an " 8 ".
Component obsolescence needs to be evaluated. The original 74xx series logic circuits were outdated even in the mid '70’s, have since passed through “obsolete”, and are now nearing “antique”. Other parts may have long since been discontinued by their OEM. The components on that board need to be checked item by item to see which ones are still available from OEM stocks. It’s true that the galaxy is littered with 74xx logic chips but gray-market parts have a reputation for marginal performance or low reliability.
I’d estimate there’s at least half a day’s work just to verify availability of components, and perhaps a couple day’s work to identify truly suitable replacements.
If you can accept a “functionally equivalent” product rather than an exact reproduction of the original circuit there is a good to excellent chance that the board could be redesigned using modern IC’s. This could add anywhere from a day to a week to the task.
Without counting them, I’ll bet there are at least 200 total parts on the board - about half of them are IC’s, and half are other components (resistors, capacitors, transistors, etc). I’d probably spend a day or two just copying the schematic into KiCAD, and verifying my work…
That board is huge! Eyeballing the placement diagram on page 13, it must be at least 10" x 25". I’ll let you use the “Estimated Cost” calculators on some of the PCB fabricators’ sites to estimate what it’ll cost to get some made, after the layout is complete.
Interesting project! Was there a parts list in the original doc? That would help a lot identifying obsolete components. I found a bit more information at http://www.ipdb.org/machine.cgi?id=2864
Cheapest PCB cost from pcb shopper was $220. I’d guess 1 week of design work, if you found a cheap online freelancer could be around $1500, a commercial service considerably more.
Being more of a software guy, I would replace all the logic with an MCU. That would have the advantage of a cheaper PCB, and be more reusable for other applications. I would probably retain the sound board for authenticity, but otherwise the user should not notice a difference. I’m not surprised to find people already doing this type of thing already http://missionpinball.org/
Thank you guys for the advice and considerations! I want to send you two more links. One has the pdf to the catalog parts. and other has photos of the original board. I would not care if I had an exact replica regarding the logic board. I would just need it to be a plug in play. I will also consult the missionpinball.org. If this does seem like a realistic build, then I will buy the physical schematic and scan it to you myself. Take a look at these additional links and let me know what you think. thank you so much!
Or you could look into an infineon xmc4xxx or xmc1xxx. They are not quite as cheap put infineon dave is a really good development tool. (stm-cube is not bad but it only allows you to easily configure the hardware. Dave can do a lot more.)
All the above mentioned mcu’s are programmable over jtag/sw.
If you buy an infineon development kit you get a free programmer. (on most of them connected via mouse-bites)
Otherwise the segger edu programmer is not that expensive. (42€)
For a one-off, that consideration is quite irrelevant. The key factor is development time. In this case, I would just buy an Arduino Mega 2560, job done. I would say it is actually a no-brainer!
The time consuming part would be reverse engineering the control logic. On balance, reproducing the original circuit would probably take less time. From the links, it appears all the components are still available.
I was wondering if the board really needs to be so big, it looks quite sparse. In order to re-use the original looms, a similar form factor would be needed, or maybe split into two boards with edge connectors. The board cost isn’t that horrendous compared to the development time.
The Atmel needs very little around it and runs off 5V, so more compatible with the original 74xx design.
Development tools are free/cheap too, I have been paying about 22€ for clone AtMega2560 boards
I dread to think what the original design would cost with expensive BOM and large PCB
I think for this application you are right. It does not matter what mcu is used. Nothing in this application needs a lot of computational resources. So who ever does create the hardware should use whatever mcu they already know.
But why would i use the mcu’s i mentioned?
Both the stm and infineon mcu’s have 5v compatible gpio’s. (as long as they are not used as analog pins. But if you use analog signals you should have one precise voltage reference anyway.)
Yes they need a 3v3 supply. (but if everything else is 5v you get away with an ldo for this supply)
I don’t think that software development for this cpu’s is more work. (especialy since there are very good tools that help you. As mentioned infineon dave and stm-cube.)
This tools are free to use! (But only work for mcu’s of the company who developed the tools)
In my opinion atmel 8bit mcu’s only survive because people are afraid to try other mcu’s. Yes before tools like infineon-dave and stm-cube came along it was a lot of work to get a bigger mcu to run. (Luckily this is history now.)
Sure, I’ve been using ARM CPUs in embedded systems since 1994…before that used Intel 80186, 286. I’ve programmed multi-million dollar mainframes and 20 pence PIC’s, and most things in between.
Recently I have been using Infineon XMC4300 and DAVE, I like it a lot. I also played with XMC1200 which runs at 5V.
I think that has to be balanced against ongoing maintenance. Arduino has just a lot more knowledge around. I find the IDE unusable, but with the Jantje plugin for Eclipse it’s a lot nicer.
Anyway, just for fun I started on schematic capture, I created a repo https://github.com/bobc/bigleague. I have done a couple of pages, quite straightforward but there are a few legends that are too illegible, some I can guess at.