I have designed an Arduino based pinball controller for use in Williams games from System6 to System11c.

This board will replace all CPU, Power driver and Music/Speech related boards - an interface for hardware extensions is also present.

At the moment this board cannot be used as a pure replacement as it cannot use the original EPROM software, instead it is meant for people who want to program their own games or add some hardware toys. I have therefore programmed some kind of pinball operating system which takes care of the hardware related stuff and offers an API for the game programmer to work with.

The board is quite cheap (well below 100€) and has no special components like FPGAs on it, so it is quite easy to build it by yourself.

There's a catch of course: I don't want to earn money with this project, which means I'm not going to sell boards or anything else to you, so you don't have much choice except of building it by yourself. However I put all the necessary data to my project page at GitHub, together with a lot of information to make it as easy as possible.

https://github.com/AmokSolderer/APC

It's still a lot of work to do the code for a new game and I'm some how fed up with doing it all by myself. Therefore my hope is to establish a small community of people who have fun designing new game rules and who just needed a base to start from.

So, if you are interested in this project, please give me some feedback.

The photo below shows the APC in my Pinbot

APC_Pinbot (resized).JPG

Hmm, mainly because I don't know any System3 or 4 games and just forgot them.

However, after a first glance at the schematics I think you're right and they should work as well.
I see a difference in connector 1J4. Apparently, the System3 and 4 games don't have a pin for the 'Memory Protection' switch and no GND pin. This would require some kind of simple cable adapter.

But I'm a bit confused about the 12V supply of the CPU board. In the schematics of the Flash CPU (System6) pin 9 of 1J2 is labeled with '12V unreg', but in the connector overview (attached below) it is mentioned that this pin has only 'unregulated 5V' (I guess it's a five, the scan is not very good).
In the 'Disco Fever' schematics it is consistent, schematics and connector overview state that there's only 5V unregulated present at this pin, but I wonder how they want to exceed the threshold voltage of the 6.8V zener diode they have there to release the Reset signal as soon as the voltages have risen sufficiently to ensure a stable operation of the 5V regulator.

If you have a System3 and 4 game at hand, could you please measure the voltage at pin 9 of connector 1J2? The APC needs 12V there to supply it's audio amplifier, so it would be good to know whether they're there or not.

Thanks in advance

Flash (resized).png

I found the schematics of a System3 power supply board and it looks like they spared the bridge rectifier, but used only TWO diodes to rectify the 9.3VAC coming from the transformer. That would mean the voltage at pin 9 of 1J2 should be in the range of 6.3VDC (half of System7) which is not enough for the APC.
The easiest way would probably be to just use a cheap 12V power supply and a cable adapter to feed this voltage to 1J2.

That's true, but the goal is to replace the sound card also and to be able to spice up the sound of these old machines a bit.

I found the schematics of a System3 power supply board and it looks like they spared the bridge rectifier, but used only TWO diodes to rectify the 9.3VAC coming from the transformer. That would mean the voltage at pin 9 of 1J2 should be in the range of 6.3VDC (half of System7) which is not enough for the APC.
The easiest way would probably be to just use a cheap 12V power supply and a cable adapter to feed this voltage to 1J2.

That's true, but the goal is to replace the sound card also and to be able to spice up the sound of these old machines a bit.

Just for your information: there is a vivid discussion ongoing in the german pinball forum (also with Jan) about how to add support for PinMame and the Mission Pinball Framework to the APC, which would require a PC (or Raspi) to be connected to the APC via USB.
With PinMame running on the PC it would be possible to use the original Williams EPROM software and MPF would ease the development of homebrew game rules.

Albeit this is mainly a software topic, I'm thinking about adding an analog audio input to the APC. With the audio output of the PC being connected to this, the audio amplifier of the APC could be used to play sounds generated by the PC on the pinball's loudspeakers.

I'll send an update when a decision is reached.

Additionally I'm going to add System 3 and 4 games to the list of supported platforms (under certain preconditions). Thanks to zacaj for pointing this out.

By the way:

This was bullshit of course (never drink beer while replying to technical threads). With 9.3VAC you have an amplitude of more than 13V and if you subtract the threshold voltage of the rectifying diodes you end up slightly above 12V.

That means, if I didn't miss anything the only adaption needed for System 3 and 4 boards to work with the APC is to add a small cable adapter to connector 1J4.

Update:

I have generated a Base Code which should run on all machines with minor modifications and is a good base for own game code.

My next task is to implement an API to control the APC via USB which would enable PinMame and MPF support.

Furthermore I'm not going to add an audio jack to the board. I've tried it out and the signal strength coming from the PC is too weak to get a decent audio quality. To improve it one would have to add more gain to the pre-amplifier which is just used as an active filter at the moment. But if you have to solder anyway you can also add the audio jack by yourself - or spend 5 bucks to buy a cheap amplifier board from china.

That means the board stays like it is and I'm about to get a quote from a german board manufacturer. When we know the price you can still decide whether you want one. Or does anyone know a suitable manufacturer in the US?

And as requested the Gerber files for the Mask and Silk Layers have been added to the Layout section.

The quotes from the PCB manufacturers have arrived. I have received quotes from the german manufacturer IBR/Ringler and from Seeed in China.
One big difference is that IBR charges 75€ for preparation (exposition masks and so on). If the board doesn't change afterwards you can order more without having to pay the preparation fee again, but with the first order you have to pay it. I'd add 5€ to each board to compensate for this. This way I'd be even after 15 boards.

For 5 boards the price at IBR would be for you:

127€ (Boards) + 7€ (shipping) = 134€ : 5 = 27€ (per Board) +5€ (preparation cost contribution) + 10€ (shipment to the US) = 42€

The same for 10 boards:

158€ (Boards) + 7€ (shipping) = 165€ : 10 = 17€ (per Board) +5€ (preparation cost contribution) + 10€ (shipment to the US) = 32€

Seeed doesn't charge any preparation fees, but shipment is more expensive.

5 boards from Seeed:

71€ (Boards) + 34€ (shipping) = 105€ : 5 = 21€ (per Board) + 10€ (shipment to the US) = 31€

10 boards from Seeed:

96€ (Boards) + 45€ (shipping) = 141€ : 10 = 14€ (per Board) + 7€ (shipment to the US) = 24€

Of course, there is a certain risk of a shipment from China getting lost. Additionally every now and then shipments like this are stopped by the german customs, which will add a lot more to the price. Furthermore I did some businness with IBR some years ago and I know they're delivering good quality.
I'd therefore prefer to order from IBR, which would mean a price per board of 42€ (roughly $38) for you if we just order 5 boards.

So please drop a line if you want a board for this price. The more people are interested the cheaper it get's of course.

In the german Forum someone came up with https://jlcpcb.com/
Apparently they charge only 32€ for 5 boards including shipping
Shipping takes a while, but he has ordered from them several times and the boards were of good quality.

For us this would mean that shipping the boards from Germany to the US would cost more than the boards itself, which makes no sense.
Therefore I propose the following: I'm going to order 5 of them for Germany only. Then we check whether the boards are OK and if they are I'd be very grateful if one of you would order 5 for the US. This person would then save the shipping cost and it would probably still be cheaper for the rest of the US guys. Additionally this company probably knows how to ship stuff overseas while minimizing customs trouble.
At the moment I have 3 ordered boards from the US (4 if Pinash confirms his interest), but I think there will be a lot more as soon as some of you have confirmed it to work.

What do you think about this?

Thank you Cheddar, I appreciate that.

I have just placed the order and for some reason they haven't charged the shipping costs, because I just paid $24.
I'll send an update when the boards have arrived.

By the way I'm just implementing the new USB API to control the APC via USB commands. So hopefully we will have PinMame and MPF support soon.

Update:

System11c displays and USB command mode are working.
Next task is to add support for System7 displays as our first machine to try PinMame on will be a Jungle Lord.

Still waiting for the boards.

IMG_0120 (resized).JPG

The boards have arrived and they look great.
Now it's up to you to decide whether you want to order now or wait for the german guys to clean the pipe.

I already implemented an additional minor HW change which makes the use of pre Sys11 displays a bit easier. It's just an additional resistor and a transistor. I still have to change the documentation accordingly, but the design data is up to date.
I got the idea from the discussion in the german forum and it is of course possible that more improvements come up when more people start using the boards.

However, it seems to be much more difficult to get the DIL version of the 74FST3244 devices in the states, so I'm thinking about changing the layout to an SMD footprint.

What do you prefer, should I change the layout or do you want to look for the DIL version? I think the german vendor also ships to the US, but of course it will be more expensive. I used this offer:

ebay.com link: 5x IDT74FST3244P Octal FET Bus Switch IDT
P1020494 (resized).JPG

I gave it a try, but the result always turned out to be quite messy. So I did some research about the replacement of the 74FST3244.
There's one direct replacement 74cbt3244 which seems to be still in production, but not in a DIL package.

The second alternative is the 74BCT760. This is not a direct replacement, since it's not a bus switch but an open collector buffer. Electrically this is IMO the better solution, as it provides a clear separation between the switch matrix and the special switches. This is the kind of device I had in mind when I designed this stage. The bus switch was just a second choice when I couldn't find an open collector buffer.
And even better: the DIL version of this device is available at DigiKey.

Hope this helps.

Hi Mark,

thanks for your Feedback.

I hope we will have first results witch PinMame soon. Luckily there is an expert living nearby who has just built his own APC board and is now going to make PinMame work. Last time we met he was almost able to run my Jungle Lord. Lamps, Coils and Displays were already working - just some issue with the switch numbering prevented us from starting a game.

Jan was also here and we tried to let MPF run the APC via the USB connection. The results were quite promising so I hope to be able to do my new Rollergames rules in MPF.

The main technical discussion ist currently taking place in the german forum, but I'm trying to keep you posted from time to time.

However if you want to join the discussion I could ask the others to move it to Pinside (at least partly).

Frank

If you wrote the german sentence without any help from your wife then you're only lacking german Umlaute on your keyboard to be perfect.
The link to the german forum is:

https://www.flippertreff.de/start/forum/topic/11356-arduino-pinball-controller/

You're welcome to join our discussion.

Frank

The USB interface is still under heavy discussion as we're currently implementing some changes. It's not yet complete, but you can see the latest dev version here:

http://docs.missionpinball.org/en/dev/hardware/lisy/protocol.html#protocol-reference-v0-09-rfc

Basically you could also use the APC alone to control additional HW - there is this 'Hardware extensions interface video' link on my GitHub page which is showing the use of an additional LED Controller and as you suggested, the USB interface could be used also. At a later stage, I might expand my SW to also support stuff like DMDs and so on, but at the moment most people seem to be interested in using MPF. Therefore they need to have an external master PC or Raspi and then it's probably easier to let this device control additional HW also.

In the end it depends on the demand. If some guys are starting to use the APC alone, I'm going to spend more of my time for supporting my own API, but at the moment I'm lucky to have very competent guys working on APC support for MPF and PinMame, so I'm doing my best to keep the pace.

You're welcome. Feel free to participate by pointing out unclear parts or gaps in the documentation for example. This is a work in progress and every helping hand is welcome.

Hi Lance,

ICs and resistor networks look good.

I think you can skip the lines 35 - 37 of your shopping list as they have a 2.54mm pitch and a lock.
The rest seems to be OK, but let's try to sum it up, just to be sure:

P1 -> 1x6 male pinheader with 2.54mm pitch
P2 -> 1x7 male pinheader with 2.54mm pitch
P4 -> 2x18 male pinheader with 2.54mm pitch
P5 -> 1x8 male pinheader with 2.54mm pitch
P6 -> 1x2 male pinheader with 2.54mm pitch
P7 -> 2x3 female pinheader with 2.54mm pitch
P8 -> 1x6 female pinheader with 2.54mm pitch
P9 -> 1x4 male pinheader with 2.54mm pitch
P11 -> 2x8 male pinheader with 2.54mm pitch
1J8 -> 2x13 male pinheader with 2.54mm pitch

That should be all with a pitch of 2.54mm, the rest has a pitch of 3,96mm.

3 times 1x4 male pinheader with 3.96mm pitch and locking mechanism
20 times 1x9 male pinheader with 3.96mm pitch and locking mechanism
1 times 1x9 male pinheader with 3.96mm pitch and locking mechanism (in the BOM this is still a 9 pin connector but your board will already have the 12 pin)

If you only want to use your board in Sys9 and Sys11 machines (like your Millionaire) you can skip the Sys3 - Sys7 connectors. That means six 9 pin connectors less and the 12 pin I mentioned can still be a 9 pin. You can see it on the board when there is an inner and an outer connector then the inner one is for Sys9/Sys11 and the outer one is for Sys3-Sys6. If there is only one then it is used for both. I'm going to add a table to my documentation to make it more clear.

The 74FST3244 is a real pain in the a*, your eBay link is about the QSOP package which cannot be used. Apparently the DIP version of it is available only in Germany. That's why I'm trying to get rid of it by using a 74BCT760 instead, which is avaiable from Mouser in a DIP package. The problem is that nobody is using it yet so there is a small risk that it's not going to work as expected. Mark is the first to try, so I would like to wait until he has confirmed it to work. I'm afraid it's up to you to decide whether to risk it or wait a bit. Your boards will probably need more than two weeks to arrive anyway.

Hope I didn't make it more confusing.

The BOM is linked in the Wiki on my GitHub Page.
I'm sorry, but I don't sell any boards. I ordered 5 to check my Gerber files and the quality of the manufacturer, but these boards are already gone.
If you order at https://jlcpcb.com/ like I did, the price for 5 boards is below 40€ including shipping, which IMHO is quite a bargain. Of course it would be great if you could offer the remaining 4 boards to others.

Thanks in advance

Frank

By the way, we're using the Molex 26-48-1245 connector (Mouser# 538-26-48-1245). I like this connector as it sells in a length of 24 pins and you can just cut it to the length you need. Additionally the clamping mechanism of these connectors is not that strong which is a big advantage when you want to change a board.

Thanks for your patience. If my documentation would be finished you wouldn't have to ask most of these questions.

This 74BCT760 is for connecting the special solenoid switches to the switch inputs as some kind of 9th row of the switch matrix. That means if things go south with this IC you're going to have switch problems. But I checked the datasheet several times and if I'm not totally blind then this thing will work. Worst thing that could probably happen is that slings and pop bumpers double fire occasionally, but this could then be fixed by changing the value of RR1.

Looks good to me. Don't be surprised when you find pads for an additional transistor and a resistor on your board. This is a new feature for pre Sys11 games so you can just ignore them.

Alas, there's no way for the APC to detect whether lamps, coils and so on are working properly - at least not without a huge effort in terms of circuitry. As Jan already wrote, there are several test modes available, but in the end it's up to the user to judge if everything is OK. However, the self diagnostics we all know from the old original boards couldn't do that either, as their self tests were mainly to check CPU related stuff like memories. This is something we don't need to care about any more, because today all this stuff is integrated in the Arduino.

Yeah, this is a problem, because at the moment you need some hardware and coding skills to use the APC. I wouldn't mind if someone would build the boards and sell them, but at the moment it's probably too early for that.

Wow, that was fast.

Have you seen that there's a step-by-step assembly instruction in the Wiki? Might save some trouble.

If you clone my repository you should get a directory named APC with all the .ino files inside. You have to compile APC.ino, the rest will be included automatically as long as it is in the same directory.
But for assembly step 1 you should start with the Soundcheck anyway. There is a directory APC_Soundcheck on GitHub in DOC/Software which contains two files APC_Soundcheck.ino and Sound.h. In the Arduino IDE the main .ino files has to have the same name as the directory, so open APC_Soundcheck.ino in your Arduino IDE and try to run it.

The picture is already online but not yet published. You can find it on GitHub in my working branch AmokWork -> DOC -> PICS -> APC_Connector.png.

No reason to apologize. It just shows me howto improve my documentation.

Congratulations, that means you have already completed the most difficult soldering tasks.

The sound quality should be quite good. That means if you hear some kind of noise or distortion than something is not right. A bit of static is normal of course, but it should only be audible when you turn up the volume without playing any sound. This will vanish if you later decide to use the digital volume control, because then the audio amp will be muted when no sound is played.
With the old boards you sometimes had quite some distortion depending on lamp and display activity - this should also be gone now.

Sorry, but I don't see it, can you post a close up of the connectors? Basically your Millionaire should have the same connectors as my Pinbot and I didn't have any problems.

You're soldering faster than I'm able to reply.
For the next step you need to use the normal APC.ino software instead of the Soundcheck, because the latter doesn't support the switches.

I'll check for the BOM issues. Thanks to you there will be a new BOM coming soon with all parts having Mouser ordering numbers, but until now it's just in the AmokWork branch. I'm going to merge it to the master branch soon, but first I have to implement the new version of the USB API.

Glad to see you're making so much progress. You have already been of great help for this project.

Could also be the old volume control knob. As you now have control of the switches, you could try to unplug the connector for the old volume control and use the digital one instead. For this you have to enter the System Settings then press Advance until you reach 'Speaker Volume'. The System should play the MUSIC.BIN file and you can increase the volume by pressing the 'Game Start' button.
When there's still some crackling take a look at your speakers. I didn't realize the speaker of my JL was damaged until I tried the APC in this machine. The speaker was still good enough for the old Sys7 sounds, but it sounded lousy when the APC was playing real music.

Ah OK, I thought you were referring to the connectors of the APC. The Sys11 Sound configuration was quite messy, because they changed it several times over the years.

When does this happen? Every switch being triggered should leave a message in the 'Switch Edges' test.

It has a temp sensor, so it should protect itself. Try it again when it has completely cooled down. Probably something is wrong with your digital volume Control, so switch the setting back to zero and use the knob for starters.

That's a bad soldering joint at U14.

Can you make a photo of this? It it only in the lower display row? Then you have to change the Display Setting to Pinbot.

This is probably not a big deal, but the problem is that I'm busy for the next 6 hours at least. Don't worry - we're going to fix this later.

Hi Lance,

No problem, I'm back, slightly drunk but hopefully still helpful.

That's strange. Like I wrote the IC has a temperature protection, so it won't die from overheating. That means if it's really blown than something has killed it and we should find out what it was before we put the next IC at risk. Most probably it's a short somewhere.
Let's first measure your speaker. Please unplug the speaker connector from the APC (10J2) and measure the resistance of the APC pins and the speaker cables to GND. None of them must have a resistance of less than 1Kohm to GND.
Then try whether the amp IC still get's hot with the speaker connector and the volume connector (10J4) removed and the digital volume set to 00.
Also check the voltage at pin 3 of the TDA. This is the volume control, which means the voltage at this pin should be 0V when the voltage control is set to 00 and then rise when you increase the setting until it reaches 5V at a setting of 255. After that, set the volume setting back to 00 and connect the manual volume control (10J4). When you turn the knob, the voltage should vary between 0V and 5V but it must never exceed 7V.
Also note that the manual volume control must always be unplugged when using the digital volume control, but even if it's not it shouldn't damage anything.

This is due to a wrong display setting. The default setting of the APC is for both rows having alphanumerical displays. That means the APC treats your lower display row like alphanumericals which leads to these artifacts.
It's the first entry in 'System Settings', the default is '4Alpha+Credit', press 'Start' to change it to 'Sys11 Pinbot'. After that proceed to 'Exit settings' which should write this setting to your SD Card. After that the artifacts should be gone (if not do a power cycle).

To be continued ...

That's good news, because it means that it's just your volume control causing the problems. Apparently the control voltage is stuck to the maximum volume which leads to distortion and overheating of the amp. Please unplug the manual voltage control (10J4), remove R70 and measure the voltage again. If it's still stuck to 5V then you have a short to 5V somewhere near C4, R76, C10 or even at pin 3 of the TDA itself. If it's close to 0V now, reconnect 10J4 and you should be able to use your sound with the manual voltage control. In this case check the soldering joints of Q5 and R68.

Look good to me. The lower display row of your Millionaire has only numerical displays and that's how it looks when the APC tries to show characters on them. But you're only going to see this in the settings, because in the game mode the lower row is only used for numbers.

The next steps involve the use of your logic probe, so let's first make sure that we're talking about the same thing. Your logic probe should have a GND and a supply cord, each with a clamp. A tip of course and two LEDs (probably red and green), one for high and one for low logic level.

Unplug connector 1J1 of the APC, then connect the supply clamps of the logic probe to +5V and GND on the power board of your Millionaire and power it up - both of your LEDs should be unlit as well as the Player 1 and the Player 3 display. Now try all pins of connector 1J1 with your logic probe. Both LED should light up which means that the pin is toggling. I expect that to happen at every pin except pin 7, as this is the one responsible for your leftmost Display column. If no LED is lit at this pin you have a bad solder joint at your connector, if only the high level LED is lit, the bad joint is probably at pin 2 of U14.

Try the logic probe also on all pins of the Solenoid Drivers (1J11, 1J12 and 1J19) with the connectors unplugged. Both LEDs should stay off, but as you said all coils at one solenoid connector were being activated I expect the low Level LED of your logic probe to be lit at these pins. It would be important to know whether this is the case for all pins of all solenoid connectors or just for some of them.

Cheer up - we're making progress. I'm confident you're Millionaire is going to work soon.

Frank

Edit: fixed some typos

OK, I've attached a pic of the TDA pinning, because we should be very clear on what pins we're talking about. I also added a picture to the docs which should help to identify the connectors:

https://github.com/AmokSolderer/APC/blob/master/DOC/PICS/APC_Connectors.png

For the following test you should unplug the speaker connector (10J2) in order to protect the amplifier.

As your problems started with the digital volume control it's probably a good idea to disconnect it. As mentioned above you can do that by removing R70 on the APC board. After that reconnect the volume connector (10J4 or 1J16 for Sys11), power up and look what happens to the voltage at pin 3 of the TDA when you turn the old volume control knob. Can the voltage be smoothly adjusted this way or do you read sudden jumps to 5V?
If this looks good you should be safe to try the sound again. For this you should adjust the volume to a low level (low voltage at pin 3) and reconnect the speaker connector. Now play some music and try if the manual volume control works.
Depending on your results we can decide how to proceed.

Good luck

Frank

TDA_pinning (resized).PNG

Sounds completely correct and I might be getting paranoid, but I've attached another picture anyway - just to be very sure.
Please check the GND of your multimeter also, the 'Solenoid GND' connector of the APC is a good place to connect it to. Pardon my scrutiny, but you can get all kinds of strange results if the ground connection is bad.

However, if all the above is correct and the voltage at pin 3 of the TDA is still as you've described then please measure the same at pin 2 of the voltage control connector (10J4). Can you now use the pot to control the voltage?

TDApin2 (resized).JPG

I guessed so

But the result of all of this: you have a break between both pins and it is probably a loose connection at R76 - Please check the solder joints.
To check the connectivity, disconnect the volume connector J4 again, do not power up the board and measure the resistance between pin 3 of the TDA and pin 2 of J4. When everything is right, the resistance should be 300K.

Just like this or did you resolder it?

That measurement is fine indeed.

But the sound has worked before, so we're dealing with a loose connection which means it can be OK in one second and then gone. You're close, it has to be there somewhere. May be try to apply some pressure to the resistor and remeasure it. Or just try to resolder it and then check again if you're able to control the voltage at pin 3 of the TDA with the pot (reconnect J4 of course

I don't think so. For some reason you cannot control your voltage at pin 3 any more. From the voltage readings you get it sounds like the pin is not connected and floating, which would result in random volume levels with a strong spike after turn on (which is what you see). As long as we're not able to adjust the pin 3 voltage the amplifier cannot work properly.

Could you please measure the voltage between both pins pf R76 with the same setup as above and see if the voltage changes when you turn the pot?

Oh yes it is, but that's not your fault. You just have one nasty problem there.

However, you should be able to mute the sound with the pot, so something's still not right there.

And let me rephrase my latest statement. Of course it is also possible that the pin 3 of the TDA itself is faulty and causing a current which spoils the voltage at this pin. But I don't just want to replace it because these pins don't die easily and I don't want to put the next amp at risk. We have to find the root cause first.

So please disconnect your speaker again, then try to unsolder pin 3 of the TDA and lift it until it has no electrical connection to the pad any more (I know this wont be easy). Then try the standard measurement again (adjusting the voltage with the pot), but measure the voltage at the pad now being empty, not at the lifted pin.

With pin 3 being lifted? That would mean the TDA is not damaged.

OK C10 is the last device left, please remove it and try again.

Do I get you right that with pin 3 lifted you get the correct voltages on the outer pin of R76, but the same strange voltage readings at the inner pin of R76?

Hi Lance,

it's still getting more strange. Either we have some misunderstanding or things are really weird on your board. To prevent the former let me just summarize what we did and what the results were. I have attached a picture of the circuit we're dealing with. As you can see this is nothing fancy.

At the beginning you removed R70 to disconnect the digital volume control. After that it's just the pot, a bit of RC filtering and the input pin of the TDA. For clarification I have named the input side of R76 (the pin closer to the rim of the board) as 'Node A' and the other pin of R76 (which is connected to pin 3 of the TDA) as 'Node B'.

Our goal is to use the pot to control the voltage at Node B, but for some reason you can only control the voltage at Node A, but not at B.

At first you checked R76 to ensure that the electrical connection between Node A and B is OK. You used a multimeter to measure the resistance between A and B and found it to be in the range of 300K which is fine. You even measured this resistance between pin 3 of the TDA and pin 2 of J4 to check the corresponding board traces also.

As the voltage at Node B was different from Node A and R76 was fine, it had to be a current through R76 causing a voltage drop at this resistor. But this current has to come from somewhere so we tried to exclude the possible candidates. First you lifted pin 3 of the TDA and after this didn't help you also removed C10.

However, when I got you right then all of this didn't help, the voltages at Node A and Node B are still not equal, right?

That means we still have a voltage across a resistor and according to ohms law this requires a current. But since you have removed all electrical connections except the resistor itself there should be no path left for the current to use and that means there has to be an unwanted current path from Node B to somewhere else.

Please check the pads of R76, C10 and pin3 for possible shorts. It might be a barely visible droplet of solder, so try to use a magnifier. And be sure to check the pad of pin 3 also on the top side of the board, may be something is hidden there. Try to measure the resistance between Node B and it's neighbours as well as GND, +5V. None of these values should be less than the value of R76, because this should be the only way out for the current.

Even though it is extremely unlikely, it might also be a board manufacturing problem. We know, that the design and layout is working, because several people are using it already, but it might be an issue with your particular board. Therefore please take a close look at the board traces also.

I'm afraid that's all I can do for you from the technical point of view. If all of this doesn't help we can try to setup a Skype meeting to look together.

Good luck

Frank

APC_Audio (resized).png

Yeah, but that's normal with the old pots. In the end I would recommend to use the digital voltage control anyway, but before we have to find out what causes your problems.

Like I wrote above, the critical point is that you have measured different voltages at Node A and Node B. This can only have two possible reasons, a damaged R76 or board trace (which is not the case according to your measurement) OR a current through R76.

Let's assume there is an unwanted current path to Node B and it's so well hidden that you're not able to find it. Then let's try to avoid Node B by building a new one. Just put R76 vertically on the pad belonging to Node A without the second pin touching Node B, do the same with C10, solder it vertically on the GND pad without the second pin touching Node B. Then simply connect the lifted pins of R76 and C10, power up and measure the voltage at Node A and the two connected lifted pins. If your problem has something to do with the board trace then you should now be able to control the voltage at the connected lifted pins with the pot, because we avoided the trace completely.
If it works you can also lift pin 3 of the TDA again to disconnect it from the pad and use a wire to connect it to the lifted pins of R76 and C10.

I have to study the schematics to see if that's possible. I never had any problems with my Pinbot though, which is the same generation as your Millionaire. However, 10V at the volume pin could possibly fry the amp, as the max voltage at this pin is specified with 7V.

I wouldn't do that, because it'll cost you some audio quality. Just disconnect the pot by unplugging J4 and just use the digital volume control instead. It cannot cause any damage, as it applies 5V max. I use a volume setting of 35-40 and never got any problems. The worst thing to happen is that the amp gets too hot and mutes the speaker, but it's not giong to be damaged by this. If this happens, you can just apply a heatsink to it's back. I never had the need to do it, but that's the reason the amp is sitting at the rim of the board.

Concerning your solenoid issues. Are all three solenoid connectors (J11, J12 and J19) affected or is it only one?
If you don't want to use your logic probe as I described in the assembly help wiki you could also use your flippers to check. There is probably an extra two pin connector for pin 1 and 2 of 1J19 (APC connector 2J12) - this is for the flippers. The advantage is that they wont suffer from being acitivated permanently, so you can use them to probe whether the coil drivers of a certain connector are powered or not. Just connect the small connector to the one you want to probe and try the flipper buttons, if the coil driver is active then the flippers will work.

I have just finished the MPF Tutorial and it works well with the APC. We still have to update the display and sound commands in the API, but lamps, switches and solenoids are already doing well.

No Digikey, Mouser or Reichelt in Switzerland?

Depending on what pinball generations you want to use the APC with, you could cut the price by not assembling all Molex connectors. In the BOM some connectors are marked with 'Only needed for Sys9 - Sys11' or 'Only needed for Sys3 – Sys7', which means that you don't need these connectors if you don't want to support that kind of games.

Additionally you could check the shipping costs of Reichelt.de which is the cheapest electronics retailer I know of. This would have the additional advantage that you could use my Reichelt shopping list. I think Snux also used Reichelt, because they were still cheaper than the rest even when shipping to the UK. Alas, they don't sell all the required parts, so you still have to order the Molex connectors, the 74HCT154 and the 74BCT760 somewhere else. And the Arduino of course, but this you can order from china.

Concerning the boards you could ask the dutch guys from post #88 to sell one to you.

System7 works very well and if you intend to program the APC natively, there is even a game software available for Black Knight. The effects need some polish and you have to add the sounds by yourself, but the game itself is working quite well.
As mentioned on my GitHub page, I didn't use the original displays, but a self made alphanumerical type. This means that the text messages of the BK software won't be readable, but numbers like points will be shown correctly.

Do you know https://lisy.dev/ ? It's a replacement MPU for Gottlieb pinballs. Might fit better in your Tag Team than the APC.

I'm not familiar with Gottlieb machines, but the basic structure is probably the same. You have to pay attention to the details of course, because if the differences are severe enough to require a change of the HW driver then things might become cumbersome.

At the moment I'm in close contact with the developer of the Lisy boards, as he is currently adapting PinMame to be used with the APC. So I'm just going to ask him what the main differences between Gottlieb and Williams are and if he sees some major problems for your hack.

Williams does also strobe with a LOW signal, so this should work. Are there any special switches like the Special Solenoid Switches in Williams machines?
Alas, I'm unable to find a Gottlieb schematic, so I cannot look by myself.

That's probably a good idea, I plan on doing the same with an old Firepower playfield that's currently getting an overhaul. As you have probably seen on my GitHub page, I have designed a small LED expansion board for this purpose.
If you want to use the APC with MPF than you're probably better off with Facecandy, as it is already supported by MPF. But if you want to use the APC natively or with PinMame, then my expansion board might be worth a try. It's still in an early phase, but it is already implemented in the APC driver SW and can be activated in the settings menu. It does automatically take care about smooth turn on/off of the LEDs, because it hurts my eyes when LEDs are switching hard.

Yeah, we found out that we're living quite nearby and working on similar stuff. So we decided to use the Lisy_Mini to run MPF or PinMame on and control the APC via USB.

I'd planned to wait until the APC has become a bit more 'established'. By this I wanted to avoid people getting confused by multiple versions circling around.
What would you like to modify by the way? I'm always curious for new ideas.

OK, I think I'm going to add the KiCad files to the next GitHub update. This will be earliest in two weeks, as I'm on vacation at the moment, without access to my design data.

Update:
There will be a new HW revision coming soon. Snux found an issue with the direct switches in Sys11a and b machines. Nothing severe, it can be fixed with a single wire, but if you're thinking about ordering boards better wait a bit for the new version to come.
I'm going to post it here when the update is online.

And some good news. Except from the mentioned issue Snux was quite successful in making his MPF controlled 'F-14 Second Sortie' work with the APC.

That's the news I was waiting for.
Many thanks to Snux for his great work, he even overcame his fear of SMD soldering.

Only thing not yet tested with MPF is the sound, but at least I've successfully tested that the APC plays the requested sound and music files if you send the corresponding commands via USB. So it should be OK from the APC point of view.

As everything seems to work I have just released the new version, it contains HW V2.4 and SW V0.11. I also added the KiCad design files to DOC/Hardware.

https://github.com/AmokSolderer/APC

Note that the project page now features a status table showing which system generations have been tested and whether there's anything special to take care of. Below that table is a link to a changelog, which gives an overview about what has been done and why, also details to the new HW and SW versions are mentioned there.

BTW, I got several questions like : I want to change the rules of my machine, does that mean I have to reprogram everything?

Well that depends on the severity of the changes you want to perform. For small changes it might be enough to run your game in PinMame and just manipulate the interface. To understand what that means you have to know how the collaboration between PinMame and the APC works.

In the system settings of the APC you can select which game SW shall be executed. If you want to use MPF or PinMame you'd choose USBcontrol since MPF or PinMame have to run on an external computer (PC or https://lisy.dev/lisy_mini.html) which controls the APC via USB. This is handled by the USBcontrol.ino program which is part of the APC software.
To implement small gameplay changes you could write an own program which you select in the system settings. This program could act like a man in the middle and would normally just forward any event to USBcontrol.ino and back. But for certain events this program could act differently and introduce the desired gameplay change.
Of course this only works up to a certain extend, because you always have to be aware about what PinMame is expecting to happen.

It should work great for additional effects. Especially on older machines with no backbox effects you could put a LED strip behind the backglas and run a fancy lightshow depending on what happens on the playfield. If the machine has unused solenoid drivers you could even add flashers to such a machine (or someone has to develop an additional driver board for flashers and connect it to the HW extensions port of the APC).

I hope that helped, now have fun with it

Frank

I just released SW Version 0.12 to make the sound work with MPF.
With these changes the APC should be all set for MPF, the next task is to make PinMame work also.

You can find some more information in the changelog including a video showing my humble MPF (and filming) skills at work.

https://github.com/AmokSolderer/APC/blob/master/DOC/Changes.md

It's probably mostly changing the connectors. Of course it would be great if you'd also make your board data available to others.

Don't worry, all Arduino DUEs are operated with 3.3V.
Apart from being bus drivers, the 74HCT244 ICs you've seen in the schematic act also as levelshifters from 3.3V to 5V. It's one of the main features of the HCT logic family that you can operate their inputs with 3.3V signals while supplying them with 5V.

I also did some experiments with DMDs, because I thought about adding DMD support to the APC and a small DMD controller would be a good solution IMHO. You could control it via USB or the APC's hardware extension interface.
I used an Arduino UNO to find out how to control the display, but this was just a test as it lacks the necessary processing power to control the display and simultaneously receive the data via USB.

However, I still consider buying a Teensy and write the necessary SW, but to be honest I don't even know if there's any need for something like this.
P1020627 (resized).JPG

Basically yes, but the 74BCT760 is an open collector driver and can only pull it's output lines to GND.
Therefore RR7 does also belong to the resistive divider dropping the 5V down to 3.3V. With the BCT760 I just added a 9th switch row to the matrix to handle the direct and special switches.

Yes, but my setup was meant for the old DMDs from the WPC era as I was thinking about controlling the old WPC DMDs with the APC. I've found several projects dealing with RGB LEDs, but I didn't find a small controller for the old DMDs. May be because it's too simple - with the UNO you don't even need additional HW. Or just because there is no need.

From my point of view the APC is all set to be used with MPF or to be programmed natively.
However after several questions concerning the population of the boards it has been become rather quite.

At the moment I know of only two people actually using the board in their machines, so I wonder what has happened to the rest? Is the population of the board such a big problem, are you waiting for PinMame support or what else is making you hesitate to use it?

I'm just curious so I'm going to start a poll about the problems you might have. Of course you're also welcome to post your opinion here.

For me this feedback is important to learn whether further support is needed from my side. Alas there's not much I can to to support the PinMame development as I'm lacking the necessary knowledge to be of much help there. The USB interface is set up, but it seems that Ralf is busy with Lisy and has no time to continue his work on PinMame at the moment.

Your guess is right, good to have you back.

I think you can probably make it work with 40 pins, but using a Teensy instead of a DUE would still require some work. I see the main problem in the SW as you would have to change the HW driver. In order to speed it up it is accessing the DUE's registers directly instead of using the Arduino commands.
Therefore I wouldn't change the controller unless I absolutely had to.

I'm also using the Lisy protocol. It has proceeded to version 0.09, so you have to to switch the MPF documentation to the dev branch to be able to see it.

Yeah, I expected that to become an issue. But to be honest I'm just not willing to solder boards in my spare time and then bother with shipping, customs and whatever might go wrong. For me it's just a hobby and I don't want it to become an obligation.
But since I don't want to earn money with it, I wouldn't mind if someone else would build boards to sell. Who knows, when the PinMame stuff works and boards are available for sale, one day the numbers will probably go up.

Unfortunately not, they're completely different, so I optimized the layout for minimum crosstalk and ground bounce instead of trying to adapt it to one certain system generation. For example, if you take a look at the pic of Ralf's Comet in my changelog https://github.com/AmokSolderer/APC/blob/master/DOC/Changes.md it looks almost perfect. So the APC seems to fit much better in Sys9 than in Sys11.

I got your point. Another way would be to make more use of the Arduino debug interface. This can send all kind of debug messages back to your PC. I have to admit that the SW doesn't support this feature very well at the moment. So one could also think about pimping the debug mode to send more meaningful messages to the PC.

Absolutely. And if it means that just some nerds like us are going to use this board then so be it.

Basically it would mean turning it into a professional product. This means a lot of non-fun work which would only make sense if I'd want to sell it.

Doing an FPGA might be a compromise solution. This could replace most of the components on the board while adding some additional functionality. Of course this would spoil the open source character of the project, but it would certainly fit to the product idea. With a hot air soldering system the FPGA could be easily pre-mounted on the board and would therefore reduce the soldering work for the end-user to a minimum.

This might be something interesting for long winter evenings, but when I look at the poll then I don't see any real demand, so I'm not going to pursue it with a high priority.

Exactly.

Also correct, but it uses the Lisy USB protocol (http://docs.missionpinball.org/en/dev/hardware/lisy/protocol.html) which only works with MPF (http://missionpinball.org/) at the moment. There is someone doing the PinMame implementation, but it is not finished yet (Help from PinMame developers would be welcome).

There's a catch I'm afraid: For me this is just a hobby and I'm not interested in manufacturing and shipping the boards. Instead I have put all the required information to order and populate the boards on my project page https://github.com/AmokSolderer/APC

It doesn't have to be one of the original core developers I suppose, just someone who knows how to deal with the USB interface.

Bontango wanted to do this and I guess he's exactly the right person, because he has already done the PinMame interface for the Gottlieb machines and his Lisy Controller. His first attempt with System9 was very promising already. Unfortunately he's quite busy with other stuff, but I'm going to ask him when he expects to have time for this.

I have to ask him anyway, because I'm thinking about some kind of Lisy-APC. This would be an APC with a Pi on board running the Lisy SW. With this you could run MPF and PinMame directly without the need of an additional PC or Pi connected via USB.
But as a joined Lisy and APC project it only makes sense if Bontango has the time to support it.

Hi ledbetter, welcome to our thread.
Before you order any boards, be sure to read the documentation. You're going to need some hardware and soldering skills to make it work.
We had some people here who have announced to have ordered boards or even started populating, but most of them went silent so I assume they couldn't handle it.

Which machines do you want to try it in? If you want to use with a certain machine only, you can spare at least some connectors.

My old APC gerber files also don't work any more, so I think it's a problem at JLCPCB. It seems to happen during the gerber processing, as it tells me that all of my files are OK, but it fails determining the board outline anyway.
May be their server is overloaded because smaller boards seem to work fine.

Let's wait until tomorrow, may be they get it solved.

OK, that means it worked for you?
I still cannot see my board.
When I click on the Gerber Viewer I get some strange error message about the tool not supporting files uploaded before July

May be they didn't notice that it's 2020 by now and that at the moment it's difficult to upload files AFTER July.

So it seems just the preview is not working.
Like bontango wrote, you don't need the dimensions when everything works properly as the previewer determines them from the gerber data. Let's hope the rest of their software still works.

Correct, but do you really want to use a lamp matrix? Would probably be easier with an LED strip.

I hand soldered everything, that's why I used large solder pads and components. The ICs still might require some practice, but you can buy a cheap hot air rework station if you don't want to solder them by hand.

That means you don't need the lamp matrix, so you can spare some SMD components.

Right, but with P-ROC there is already a working solution. However there is someone in the german forum planning to do an open source controller for WPC, so something might come up in the future.

Bontango is doing the PinMame stuff. At the moment we're planning to extract the WAV files from PinMame and put them on an SD Card. Then PinMame can just tell the APC when to play which file. This makes changing sounds an easy job, since you can just change the files. Of course they have to be extracted for every game, but this must only be done once and can be automated.
Actually we're going to try it this weekend with my Pinbot.

I know very little about PinMame, so I can just tell you how I understood it to work but I might be telling bullshit here.
As far as I know all the old pre WPC games had synthesized music and even the speech wasn't sampled in a format we could simply use nowadays (at least in machines up to System11). Memory was just too expensive to do that so they used synthesizer ICs which generated waveforms that sounded like music. Therefore PinMame has to emulate these old ICs and generate a WAV or whatever can be used with today's audio cards.
That means it's not enough to locate a certain sound in the ROM, but you must make PinMame play and record it as a WAV to be able to use it.

True, the SW setup would be easier, but we'd loose flexibility in doing so. If we spend the (one time) effort to generate WAVs and let the APC use them, we can not only use these sounds with PinMame but also with MPF or the APC itself. The easiest would be to just replace certain files on the SD card to introduce your own music and effects. If you design your own game rules with MPF you could use the old sounds.
And as the APC features it's own pinball operating API you could even run your game in PinMame, but let the APC act as a man in the middle and introduce certain rule and effect changes and with the sound and music files at hand you have the full flexibility.

Ouch that's a lot of work. May I ask why you'd wanna do that?

That's the reason why it might take hours to extract the audio from PinMame because you have to make it play every pitch and tune and record it as an own WAV. You end up with hundreds of files and still have to know which file belongs to which sound.

Like it wrote we're going to try it this weekend. I'll keep you posted about the result.

Well, insanity seems to be a common disease among pinball enthusiasts.

The APC offers good sound quality without any additional sound HW, so if we manage to play the right wav at the right time we're done IMHO. The main problem appears to be the mapping, but according to Snux it must be feasible, because this Pinsound Studio seems to do exactly the same.
And as I already wrote we have the additional advantage that these sound files are accessible for MPF and APC games also. I think it's a great feature to run a game in PinMame and just add certain rules on the APC in C++.

Yeah, this is probably overdone, but these FETs are for the flipper fingers and I hate when those are weak. So I tried to keep the on resistance as small as possible, but I think the 540 will be enough.

The whole solenoid part is a bit overengineered, as I wanted to be able to conntect all kinds of fancy hardware to it without having to use a second driver stage as used in System11 for the high voltage coils.

Update: the first PinMame tests are very promising.
The lamps are working perfectly, displays and solenoids are having minor issues.
We still have to increase the emulation speed, but according to Bontango this is just a setting.

Navigating the test menu with the Advance and Up/Down switches doesn't work correctly. Do you know if the behaviour of these switches in PinMame is different from the original machines?
Except of that all switches are working fine.

We still have only partly implemented the sound handling. Up to now we just have 255 sound files which contain their their own numbers in speech, so Pinbot is telling me all the time which file he is playing at the moment. Mapping these file numbers to the correct sounds is one of the next tasks.

But all these are relatively small problems and I was already able to play some games.

I'm surprised by the high component prices. Did you order an Arduino DUE from China or did you pay the whole 40 bucks?

Yeah, the Mouser prices seem very high to me. I checked some parts and almost everything is less than half the Mouser price here in Germany. A simple 1206 Resistor is almost 5 times the german price.

That means, if I ever pull myself together and do a board version that can be populated by the board manufacturer it would probably even save some money for you US guys.

I'm curious about it.
Of course I'm especially interested to learn about missing or misleading parts of the documentation.

Yes, R80 has been removed. Have to delete it from the Wiki also.

I wasn't sure if someone except me would ever use it, so there is no documentation yet.
At the moment it is hardcoded to just control the LED board when the corresponding setting is activated, but the task to do it right is on my todo list … somewhere

No, that's a mistake. C2 is supposed to be a 100uF electrolytic capacitor, so it's number should be something like 710-860040272001.
Thanks for the hint. I'm going to change that.

Of course, any kind of help is highly appreciated.

It is already included in the current branch. It's part of the lamp section in the HW driver (TC7_Handler in APC.ino) and that's part of the problem, because this driver does directly control the HW and it is therefore the most sensitive part of the SW. Any issues here may lead to very nasty consequences.
So, give me some time and I'm at least going to add a command to send raw bytes to the HW extension interface. By this you don't have to deal with the HW driver directly, but you still have the flexibility to implement additional functionality as you like.

It looks like the USB communication is not working. Are you sure your APC is in USBcontrol mode? In SW rev. 0.13 the system comes up in USBcontrol mode by default, but not in 0.12. This version comes up in the basecode mode and you have to switch to USBcontrol manually in the system settings. This is shown in my MPF video

Compiler-wise 0.13 should behave the same as 0.12. When Git has pulled it you should have a directory called APC with all the .ino files (you need all of them). Navigate there in your Arduino IDE and open the file APC.ino (the other .ino files should open as additional tabs). Now press compile and it should work. If you're still having problems please post the compiler error.

Exactly.

At the moment I'm just using Sel5 as the LED_exp board is the only expansion board up to now. I think three select lines (four if you count the SPI_CS line which is also present at the exp connector) should be enough for most applications. Using it as an address is possible, but it would require additional logic on the receiver side.
If you want to play with the expansion port, please note that only Sel7 is 5V tolerant, the other selects are directly connected to the DUE which can just handle 3.3V. That means these selects need LS or HCT logic in the receiver.

I've added it to the LED_exp folder of the V0.13 branch, but be warned it's not exactly self-explanatory and far from ready.
Let me summarize what I remember:
- The APC sends one byte every millisecond.
- The first 8 bytes represent the status of all 64 lamps, after that it might send an additional command. Up to now only two commands are implemented.
- 192 is the optional color select command which is followed by three bytes representing Red, Green and Blue values. Every lamp that is turned on afterwards will light up to this color setting.
- After 20ms the sync command 170 must occur which triggers the LED_exp board to write the new lamp states to the RGB strip. Then it starts again with the first 8 bytes of the next frame.
- The LED_exp controller will fade the LEDs on and off in 5 steps to smooth the hard switching of LEDs.

For complex animations we would need additional commands. It should be possible for example to assign a certain color to a particular lamp. At the moment it's not possible to turn on two lamps of different color in the same data frame.

I've planned to do it when my Firepower is ready, because I want to equip it with an WS2812 strip instead of using the lamp matrix. Depending on the complexity of the lamp animations it may even be necessary to upgrade the LED_exp board to increase the data rate.
Or may be I'm just going to buy a facecandy board and control it from the second USB port of the DUE.

Yes, it can be powered from the USB port.

If you need more than 4 boards on the expansion bus then such an interface board would be a good way to do it. I'd probably go for a 74x138 but the outcome would be the same.

However, this bus can only drive boards which are located close to the APC board. That's because these ribbon cables are quite susceptible to distortion and will pickup all kinds of noise when being too long. Therefore you won't be able to place extension boards outside of the backbox which is a huge drawback for this LED strip driver as the LEDs will be under the playfield.
In my opinion Stern solved this problem in an elegant way by using shielded network cables as inter-board connections. So when we're thinking about an interface board it would probably be a good idea to plan for a parallel to serial converter which can then drive a network cable. The receiver would just need a shift register to get the parallel data back.

Yes, that's what I'm talking about. Why didn't you go for this solution?

I'm still not sure if the APC is really the best solution for a homebrew machine, as it is tailored to the old machines with matrices for everything.
When I got you correctly then you're not going to use the display part nor the lamp part of the APC, so why don't you use a modular solution like OPP?

OK, so what kind of configuration do you need?

- I guess the 8x8 switch matrix + 8 direct switches of the APC.
- The 24 solenoid drivers of the APC and 5 extra?
- Sound from the APC or external?
- What about a facecandy board for the LEDs? It has USB which is less sensitive to noise than a ribbon cable and it is already supported by MPF.

Yeah, I have to update this part anyway when I'm doing my HW_exp interface rework. But the normal lamp matrix refresh needs just a few clock cycles, so not much to gain here.
Are you planning to keep the APC in the backbox or do you want to place it in the cabinet? When you're doing a complete rewire anyway you have the choice and since you're not going to use the display part there's not much in the backbox to connect it to.

You're welcome.

For the additional solenoids of your Scared Stiff you could use use something like the attached schematic (is the name solenoid expander board a good choice or should I call it solenoid extension board?)
This would be an easy solution if the APC and this board are kept closely together. If you wanna place one board in the backbox and one in the cabinet we should probably go for a USB solution. The Arduino DUE has one additional native USB port which we could use for that. That would require an Arduino Nano or something similar to control the extension board.

APC_Solenoid_exp.pdf

Don't rush. I have to do my rework of the HW_ext software first to make the APC SW support it. So you have enough time to bring your APC board to life first, because I'd like to have this PinMame stuff completed before I start new tasks.
I'd also etch a board and test it, not only to test the SW but I'd also like to try these IRL530 transistors as a possible alternative for the APC board. This would spare the UDN2981 levelshifters and all those SMD resistors at the transistor gates. Plus they are available in an SMD package and can be automatically populated in China.

Sounds to me like you'd rather need a small controller with HV tolerant inputs and a solenoid driver. I did a shaker controller for a friend of mine. It features an Arduino Mini (or Nano) with 16 inputs which are 100V tolerant. That means you can connect any input with a solenoid output or some lines of the switch/lamp matrix of your pinball machine. By this you'd be able to program an event that can be triggered by certain patterns of the lamps, switches or solenoids.
You can find the schematic attached and we even have done a board but it's also not tested yet. I think it won't take long to do it, but like I said I want to complete the PinMame support of the APC first.

Talking about PinMame support - it is almost done. We still need to extract the sounds of the second sound channel and put them on the SD card of the APC, but we did it already with some of the channel one sounds and it works fine. I already spent some hours playing my Pinbot with Lisy_mini and APC controlling it and apart from some sounds still missing it works perfectly. So I hope the next APC SW release will come soon and after that I hopefully have a bit more time for other tasks.

Shaker.pdf

Did you use them in an APC board?

There's one thing you should be aware of:
The APC (as well as the solenoid expansion board) does not have any free-wheeling diodes on board, because the pre-WPC machines had them soldered directly to the solenoids or on the aux power driver board.
This changed with WPC when these diodes were moved to the normal power driver board.

That means if you want to use the APC in a WPC or any homebrew machine you have to make sure to place them somewhere or your power driver transistors will fry eventually.

That's the idea, but the design was done just based on the datasheet of the transistor. I still have to order some parts and do some measurements.
The 47K might change as it is a trade-off to be able to connect the inputs to the switch matrix, as well as to the solenoid drivers. The switch matrix requires low currents in order not to disturb it. For the solenoid drivers the current needs to be higher as the parasitic GS capacitance of the protection transistor might produce a current spike at it's source and we don't want the resistor to transform it into a voltage spike.

But the solenoid drivers are the tricky part, so if you just want to use it at the switch matrix then you could give it a try.

Interesting. I don't know Gottlieb very well. Did you have to change a lot to make the APC work with it?

No problem, the input dividers (RR7, R77 - R85 and RR8) are reducing the input voltage to 3.3V.

Williams is also using a low pulse at the strobes (Sw Matrix Drivers).

Are you planning to use an LED Board or do you want to use the original light bulbs? The latter would require some kind of adapter at the HW_ext interface of the APC.

I just have a System80 schematic, but if your system is similar then you could use the original driver board from Gottlieb and just add an adapter board for the HW_ext interface of the APC to it. From the first glance I think for the lamps this adapter board could be quite simple, a single 74HCT154 would probably do the job. This way you wouldn‘t have to populate all these lamp driver transistors at least for the prototype.

I only tried it with the original displays in my Rollergames setup which is part of the Github project:

https://github.com/AmokSolderer/APC/tree/master/DOC/Software/MPF/Rollergames

This is the setup I used for my MPF video

I don't know much about Smartmatrix, so I'd stick to the MPF documentation http://docs.missionpinball.org/en/latest/hardware/smartmatrix/index.html

Did you populate the display related parts? If yes you might consider using your BK2K to make sure the board is working correctly before starting with MPF. You can adapt the Basecode.ino to the BK2K as described in the 'Software Preparation' part of the main Wiki page https://github.com/AmokSolderer/APC/wiki
When the Basecode runs you can enter the test mode by pressing Advance with the Up/Down switch in down position. This way you can test the HW without having to bother with MPF's configs yet.

Yes, for some reason Williams changed the polarity of the display signals when they introduced the 2x16 alpha displays. That means all your characters are inverted (you can actually read it with a bit of training )

The display type can be changed blindly by doing the following:

- press the Advance button. This will make the display stop scrolling.
- slowly press the game start button repeatedly until the display becomes readable
- lock Up/Down in the down position and press Advance to get to „Exit Settings“
- press Game Start to confirm. If an SD card is iinstalled the new setting is stored

First of all: did it work, can you read your displays now?

You're right I should probably state in the comments that DisplayType and so on are just pointers to APC_settings.
If a settings file has been found on the SD card then APC_settings is read from there, otherwise the values from APC_defaults are used.

Same here, but I think I wouldn't be able to resist. It's just some seconds and if no meetings are imminent

Remember to change DisplayType back first or you might get some unexpected results.

Of course I'm not trying to tempt you, but theoretically you could change the first byte of APC_defaults to 3

Thanks. When are you going to populate you first APC board?

The current challenge with PinMame is to extract the sound files automatically. We have decided to just extract the files, put them on the SD card and let the APC play them. The APC offers good sound quality and it's easy to replace the files with own sounds if desired.
If the files are stored on some internet server, then this work must only be done once for every machine. Bontango is currently extracting them for my PinBot. The results are very promising, but I think it's still mostly manual work. Up to know we have roughly half of the files for one of the two audio tracks.

Hard to believe as it looks so familiar. Try to finish step 3 and press 'Advance'. If the result looks like the picture below than we know what it is. The other picture is the same with the right display setting.

Sys11c_display (resized).jpg

Sys11c_display2 (resized).jpg

This doesn't look like a bad solder joint. Then you would expect certain display segments or whole digits to be missing.
So let's look for the weird stuff:
Is your ribbon cable connected correctly? It's very easy to have it shifted by one row and the result could be similar.

Another idea would be to press the 'Highscore Reset' button after bootup (while the display is scrolling). This invokes the blanking which should make the display go blank. If this doesn't work and all segments are lit instead then check the jumper JP1. For your type of display the center pin of the jumper should be connected to +5V.

I'd like to track this problem down systematically, because any additional soldering might just further damage your board.

OK, but the damage is done so we have to make the best of it.
When I got you right, you don‘t want to use the display part anyway. It seemed like a low hanging fruit to put the board into the BK2K and test it, but you can still skip the display part and test your board via USB.
You could try to connect it to MPF or you could use a terminal program to send the USB commands directly.