Another OPP Carrier Board for MPF use

[cobra18t]

I put together a carrier board for two OPP PSOC boards. This is for new builds or complete overhauls, so I am not shooting for legacy system direct compatibility. I am looking for suggestions and feedback before I order some prototypes. It includes:

• Power filtering/switching
• 24 solenoid outputs (trying some new surface mount FETs)
• Up to 48 direct switches (can also use an 8x8 switch matrix)
  
• Raspberry Pi hosting option
• I2C and servo output via Pi
• OPP chain ports for expansion
• 4.25 x 7.75"
• Power connectors: 0.156"
  
• Switch connectors: 0.1"
  

All my lighting will be handled via FadeCandy, so no lighting support here. I tried to label all pins according to their MPF designation to make configuration easier. Forgive the lack of 3D models on the board. I am rediscovering KiCad after using Altium for years.

[Cadrion]

A few things that I see to consider- maybe onboard replaceable fuse for the HV.

Having a ground wire for each of the direct switch connectors may be pretty handy- not sure if the unlabeled pin a gnd or a key.

Heat dissapation/current draw going to be adequate for q26 with 24 coils? Just asking, don't really know one way or the other.

Heat dissapation ample for the density of the coil fets? Again, not sure just asking.

Maybe consider an OPP logo on there somewhere. I thought it was a good idea for my board- Hugh is awesome for giving us something to build with. (Thanks again Hugh)

That's all I see at this point. Looks great and cool idea for the pie integration.

[cobra18t]

@Cadrion, thank you for the comments.

1. I already have a fuse, both electronic and physical in the switching power supply I am using. Additionally, I plan on fusing groups of coils as necessary per game. I like how Gottlieb did it on the underside of the playfield on my 1985 Tag Team.
2. Right now the unlabeled pin was intended as a key, but connecting it to GND would add some versatility. Thanks, I will do that.
3. Q26 should be able to take the load. It is a 20A continuous part with ~50mohm on resistance. Heat dissipation should be okay with the planes that I have in copper and the fact that I will not be anywhere close to 20A average current.
4. The coil FETs are the biggest question...hey, I did call this a prototype. They are about 4 times as efficient as standard OPP FETs, so heat generation is lower. I do have a large plane on top for heat dissipation and lots of vias to also sink some heat to the bottom. What could go wrong, right? I will be running some tests with a thermal camera to find out. If anyone cares, they are Nexperia BUK9M24-60EX.
5. Has Hugh (@OPP) said that using his logo is okay? I would gladly add it if nobody considered it posing. I will send him a message if he does not see this thread.

[OPP]

1.  So looking at the board, it seems like it is essentially 6 solenoid wings and 2 wings that make up the switch matrix.  That all looks good, and looking at the numbering of the solenoids, and numbering of the input pins, I'm pretty sure that things are appropriately routed.

2.  U3, is that a 3.3V to 5V level translator for the Pi serial port?  If not, be careful that the Pi serial port is 3.3V and the PSOC boards are 5V.  While the PSOC will accept the 3.3V and do the right thing, the 5V coming from the PSOC will damage a Pi that can only handle 3.3V.

3.  Q26 looks like the P channel MOSFET for turning on/off the high voltage supply.  Is that type of control needed?  On your 3 pin J10 for selecting it, it uses either GPIO26 from the Pi or uses 5V0.  I'm assuming that Q25 has a Vth of less than 3.3V, otherwise, the Pi version might have problems.

4.  I like the little LEDs to show that the voltage rails are active.  I have had a few people building the bulk cap board, that don't populate them, but for safety reasons, I really like to visually see if the voltage is on.  You might want to consider a couple different LED colors to give people a quick indication.  (i.e. red means the high voltage is being shipped to the coils.  That means I should be a little more careful.)

5.  You've done a really good job laying out the planes.  Top notch!  Because of the wing design, it was always difficult to get a clean plane since I only had .8 inches.  You've done a much better job.

6.  There are thermal relief pads for the VIN/VDRV.  If there is a time to consider not using thermal relief pads, it would be there since that is where the largest inrush current will be to your board.  Removing the thermal pads will make it more difficult to solder because you need to heat up a larger part of the plane to get a good solder connection.  It probably doesn't matter, but I thought that I would mention it.

7.  Make sure you can get 9 pin 100 mil spaced connectors.  Many times, there are certain standard sized connectors that are significantly cheaper than other lengths.  I like @cadrion idea of adding a ground (because you will need on to complete the circuit anyway), but I also like your idea of having a knocked out pin to make sure things don't get misconnected.  When attaching cables to a machine, it might not be the best lighting conditions, and if the pins can get misaligned, or a cable can get placed in the wrong position, it almost always happens.  If 10 pins can fit, you might consider that.  It gives 1 pin for gnd, plus a knock out pin to make sure that a cable isn't connected to the wrong connector.

8.  I can't tell the width of the board, but if it is < 10 cm it can be significantly cheaper to get fabricated.

9.  Not 100% sure what R40 and R48 style resistors are for each solenoid.  Is D24 an LED and R40 or R48 a current limiting resistor for the LED?  It is just tough for me to tell.

10.  You can use the OPP logo.  I would be honored.  Here is the legalize.  OPP is a GPLv3 project (a copy left license).  As such, any works that are based on OPP stuff end up being GPLv3 projects also.  (That's the difference between GPLv2 and GPLv3).  Officially, the work must be publicly released so others use it, and base new projects off it.  Reality wise, it would make me happy if somebody ask for Gerbers for the board, they would be provided.  I personally put, I guess, years into the project in hopes that others would be able to base other projects and get use out of it.  When I see others using the work, it makes the time that I put into it that much more valuable which makes my small contribution more valuable.  That's why I started this paragraph with I would be honored.

Final note:  Do you want to base your design on the PSOC (which uses an EOL development board), or would it be better to wait for the STM32F103CT8 board.  That will probably involve moving one line from the left side of the processor to the right side since the NRST line can't be used.  It would also involve using USB as the serial port connection as opposed to chaining serial ports together.  Just something to think about, but if considering that, send me another email, and I can try and be more succinct in what ports will be used for what.

-Hugh

[cobra18t]

Thanks for great feedback, Hugh.

1. Yep, you got it. I opted not to use an incandescent config to get more solenoid outputs. 
   
2. U3 is a translator for 3.3 to 5V. I was concerned that the high VIH of the PSOC actually would cause it not to register input from the Pi properly. The resistors next to it are the voltage divider into the Pi RX.
3. I am not sure if I will use that PMOS control, but I like the option. Yes, all transistors I used react well to 3.3V inputs.
4. I was going to use blue LEDs for the power supplies and yellow for the coil outputs. I think red on the Vdrive is a good option.
5. Thanks. A wise mine once said to me..."You already paid for the copper, don't etch it all off."
6. I had already been wrestling with this one. These planes use a different thermal relief that lowers impedance from what is used elsewhere on the board, but I will look at it again to see if I want to strike the relief altogether.
7. I already bought the connectors. I am giving the Chinese KF2510 connectors a try, although Molex KK254 is the fallback plan. There was no significant cost break for the common sizes. I am going to leave it at 9 pins for now. I did connect the 9th pin to ground. You can either use it as ground or clip it to use as a key.
8. The board is bigger than 10cm. It is 7.75" x 4.25". Through JLCPCB, it will be $20 including shipping for 5 boards, so that is good enough for me.
9. One resistor is a current limiter for the LED, the other is a pull-down for the MOSFET gate.
10. Excellent and thank you for your contributions. I will add the logo. One reason why I decided to make this board in KiCad is that it seemed to be more commonly used in the community, so I could more easily share it than my usual boards made in Altium.
11. This is a stop-gap board. I need a board now to get my project going. I already had the PSOC boards. These 5 carrier boards may be the only ones I make for the PSOC, but I also hope to use it as a prototype for the next rev to use the STM32 blue pill. I will contact you for more details so I can get a jump on the STM32 version. Also, I only count 30 I/O on the blue pill if USB is in use (PA11 and PA12 appear to be shared with USB). How disruptive is this to your 8pin wing concept?

[cobra18t]

I just ordered the board. Now it's time to order parts and make a solderpaste stencil...

[Adrien A-G]

Cobra, I just discovered your OPP PSOC carrier board, I am amazed and interested. I am not sure to understand everything in the discussion as I am not really good in electronics. 

I am looking forward to your first test of the board.

Good job! :¬)

[cobra18t]

I expect to get the boards and parts on Monday. It may take a few days to get around to assembling one.

Adrien, I do plan on making files available after I confirm there are no major issues. I would like to sell some OPP-based boards in the future, not to make a bunch of money, but to make them available to people with less electronics skill. Even if I sell them assembled, I expect to get the cost down from other ready made options on the market. That all will not happen until the next iteration of OPP since the PSOC board is now obsolete.

[Adrien A-G]

It sounds like a great idea. I think it could help people with no soldering skills or do not want to solder in order to use OPP. 

I do solder electronics (sadly, I don't understand all the inner working of electronics), but I'm more used to through hole than surface mount components. The FETs you are using seems to be so tiny, do you use a reflow oven? I'm not sure I would try to solder an unpopulated version of the board with my soldering iron.

I really love the integration of the Raspberry Pi to your design, seems so neat. I2C and Servos!

Do you think this board could work with high voltage being 12v (my mini pinball project use cheap 12v solenoids)?

Thanks!

[cobra18t]

For surface mount parts like those tiny FETs you mentioned, I use a lasercut polyester stencil to squeegee solderpaste on the board before placing the parts. Then I put the board on a hotplate to heat it to reflow temperature. I use a hot air rework station to touch it up if I need to. I have done this enough times that it is faster to do it this way than to solder through hole parts. I will take some pictures when I do the first one to give a better idea.

[cobra18t]

If you are interested in the play-by-play on this project, you can track it on my project post here:

https://pinside.com/pinball/forum/topic/re-game-1985-tag-team-pinball

I will only update this thread on MPF when the board is "done."

[Adrien A-G]

Oh great! I'll follow your progress there. 

Thanks for the details on your soldering technics.

:-)

[cobra18t]

I finally got around to testing the new surface mount transistors with actual solenoids. MPF made it easy to run a coil sequence in a show and just increase the speed of the show to make it a more and more difficult test. Based on the testing, I do not foresee any heating problems with these transistors. More info on the build log:

https://pinside.com/pinball/forum/topic/re-game-1985-tag-team-pinball#post-5274491

[cobra18t]

I have been using a hacked version of the original board for about 8 months and it has been great. It is about time to formalize some of the hacks and just iterate the design in general. Some of my friends have been asking for one of their own, so I also intend to have the surface mount components soldered at the factory.

Changes:

1. Officially moved to STM32 boards. I have been running them with adapters for months. The 4pin header on the STM32 board is supposed to be removed and replaced with a vertical header.
2. Removed RPi interface. With Gen3 of OPP being a USB-based STM32, the need for a separate interface faded away
3. Removed the power control switch. I left it on the entire time I had it, so why have it?
4. Added fuses and arranged power supply connectors to be separate. I have not had an incident requiring this fusing, but a good friend burned through at least 20 fuses while dialing in his PROC. Maybe I should better protect things when this goes out into the wild.
5. Added diodes to solenoid outputs. I generally use coils with diode on them, but if I forget one, the diode protects the FET.
6. Added 2x Neopixel level translator outputs. This is made possible by the great work that OPP has been doing to enable neopixel support and allow Solenoids on the same wing (NEO_SOL).
7. Changed the FETs to a higher voltage version. These are rated for 100V and 55A continuous.
8. So what can it do?
1. Power Filter. Good for 24-50V systems.
2. 24 Solenoid outputs (indicator LEDs on each output).
3. 38 direct switches <or> 64 matrix switches and 22 direct switches.
4. 2x Neopixel chains. Each connector is rated for 7A.
5. Is this not enough? Extend it with another board or a standard OPP-style wing setup.

[Benoit PETIT-DEMOULIERE]

That is GORGEOUS

[jabdoa]

Looks very nice and well done. One small remark from my experience building my machine: Consider adding two more power feeds for serial LEDs (or even more). For reliable operations of longer serial LED chains you often need additional power feeds. I run 20 chains of about 30 RGB LEDs each only and still feed them with power from both sides.

Jan

[MOSFET]

cobra, that looks awesome!!!

[mike wiz]

If I do another retheme I want to use OPP.

Do you plan to sell any of your boards?  I would buy 2 or 3.

[cobra18t]

Thanks guys. 

Jan, I will take another look at auxiliary 5V taps. My original thought was that you would use a heavy gauge wire off these connectors and splice additional taps off the end of that, but more might be better. Maybe I could put another 4pin connector next to J10 if I shuffle things around a bit. I want to keep the right side clear for usb cables. I may also put some slots in the board to optionally secure usb cables with zip ties and a 3D printed or lasercut support. Unsecured micro-USB cables make me nervous in vibration environments.

Mike, I am making these initial ones for a few local friends...and myself, of course. I may have 1 or 2 extra that I could sell. Also, if the initial run looks promising, yes, I am interested in selling more down the line. I was thinking to sell in both a fully assembled form and just the boards with all the surface mount parts assembled. Pricing is far from being set. It will be more expensive than doing OPP yourself and less expensive than Multimorphic or FAST. That is the goal of this board anyway. OPP is too difficult for some people and Mulimorphic and FAST are too expensive for some people. 

I am probably a month out before getting the first boards.

[jabdoa]

If you want to add some docs about your board to the MPF docs let me know. Guess there will be some specifics on how to configure and use them.

Jan

[cobra18t]

Yes, maybe it would be a subpage of the OPP section. I will be thinking about that.

[cobra18t]

I made 2 simple changes tonight that took forever because of all the pushing and shoving I had to do to make room. There is now a second voltage output connector for additional neopixel power taps. I also added slots that can be used to tie down the USB cables. We will see how useful that is.

[Adrian DeGroot]

The board looks great! Glad you added the fuses! ;)

[OPP]

As always, feel absolutely free to ignore anything that I say, but here it goes:

1. Diodes for solenoids on the board.  I only know of one manufacturer that did that, but no others.  I wonder why?  It always seemed like a good idea to me, but then, what happens if the diode fails?  I have had to replace a fair number of diodes on pinball machines that did not work anymore.  Maybe this isn't an issue anymore, but it will be significantly more difficult for a user to replace a surface mount diode than a diode on the mechanical assembly.
2. 9-pin high voltage connectors.  Seems like there is a ground pin, but that doesn't allow keying to make sure that the correct connectors a put at the correct location on the board.  If they were switched for 10 pin connectors, the female connector housings could have a blocked housing pin position to make sure that the connectors plugged into the correct positions.
   
3. 9-pin high voltage connectors.  The ground pin is moving around in the connector.  That is scary because if the wrong housing is plugged into the wrong connector, there might be an immediate high voltage to ground direct short.  Minimally, I would suggest having the same pin as the ground on each connector to guarantee a high voltage to ground short doesn't happen if plugged into the wrong position.  Instead of a ground pin, it would be better to have a simple N/C that could be used as the key pin.
   
4. 9-pin high vs 10-pin connectors.  Sometimes it is much more expensive for certain size connectors.  For 100 mil connectors, last time I checked, 10 pin connectors were inexpensive, while 9-pin connectors were very expensive.  For the high voltage connectors, a long one can be bought and cut to size.  For the 100 mil connectors, it might be cost prohibitive for 9-pins.  You might just want to check Mouser/Digikey and see the price difference.
5. Grounds on low voltage input/switch matrix connectors.  I would rather have a keyed pin than a ground pin.   That would prevent housings being plugged in incorrectly.  (10-pin vs 9-pin could also help this).  Also, there is probably a ground wire or ground braid on the bottom of the pinball playfield to connect all the logic grounds together.  More than one logic ground connection isn't really needed because of the minimal amount of currents on the logic grounds.
6. Are the STM32F103 boards plugged in this board, or are they soldered into this board.  If they are plugged in, they will sit off the CobraPin board by a good bit, and it may be difficult to tie-wrap the USB connectors without strain.  If the board are soldered, are there any provisions for upgrading the STM32F103 using the ST-Link/V2?
7. Have you considered moving the USB pins from the board to a new more stable connector?  The little phone connectors are prone to failure.  Nearly any other USB connector would be less apt to failure.  (Of course, properly securing the cable so that the vibrations are minimal, and not plugging and unplugging it large numbers of times will make that less of a problem).
8. The silkscreen information is provided is fantastic.  That should make users with MPF configuration of the yaml very easy.
9. Providing two canned configuration files for the STM32s will be helpful for your users.
10. The LEDs showing the firing of the solenoids is swank.  I would have loved that on the solenoid wing, but couldn't fit it.  Nice!
11. High voltage LEDs showing high voltage leg has power is also great.  Scanning the board, I would immediately be able to tell that one of the fuses are blown.  Putting those LEDs directly next to the fuses would make it even more evident.

That's all I have.  As I said at the beginning, always feel free to ignore any of the suggestions.

[OPP]

Crap.  #2/#3 is not a ground pin, but a high voltage pin.  That removes some of the scariness, but have keyed connectors would be better.   Should have thought harder about what those pins had to be.

[cobra18t]

Thanks for the feedback, OPP! Here are my responses:

1. Yep, I view these diodes as backup diodes and would not rely on them. Their purpose is a last ditch effort to protect the surface mount FETs, which are harder to replace than surface mount diodes. I plan to use and will continue to recommend diodes on the coils themselves. For PROC PD-16 users, the manual declares that coil diodes are unnecessary because of the onboard surface mount diodes...feel free to continue using coil diodes.
2. The 9th pin is specifically for keying! That is why it moves around. I originally had it N/C until people requested that it also be connected to GND if they chose to populate that pin. In hindsight, that request was focused more on the logic side. I will change the 9th pin to a high voltage pin on the high voltage connectors. Again, it is the choice of the user to use it for keying or for power on the high voltage connectors and GND on the logic connectors.
3. <see #2>
4. I have been using KF2510 connectors for 0.1" and VH connectors for the 0.156". With those families, the price is proportional to the number of pins with no price penalty for "unusual" pin counts. 9pin is pretty common in pinball for keying purposes.
5. Again, these are for keying with the provision for GND for the people that wanted it. All the key pins have a silkscreen mark on the underside so that while you are assembling, you know which ones not to solder and pull out instead.
6. The STM32 boards are plugged in. My intent was to experiment with a lasercut or 3D printed standoff for the cable so that it was zip tied to the standoff, relieving the strain.
7. Yes, I toyed with moving to on-board connectors. I may still do it, but I am hoping this idea will pan out. I do like that if the connector goes bad, you just throw out the whole STM32 since they are cheap. That is much easier than desoldering a connector.
8. The silk on the existing boards have already paid great dividends. It is tedious work to get them just right, but it is worth it!
9. Yes. I think I will need 3 configs. Board 0 will always have the same config, but Board 1 can be configured for a switch matrix or just regular inputs.
10. The coil LEDs in my stress test video for Tag Team was my favorite. They were going nuts!
11. I originally had the high voltage LEDs next to the fuses. I moved them because I like the way it tied them to the bank of coils. I don't have a strong preference on this, anybody else have an opinion?

[cobra18t]

Jan, I added an OPP logo to the board in the last rev. This is also specifically designed for use with MPF, should I add an MPF logo to it as well? Do you have one for me to use?

[jabdoa]

We do not have a good monochrome logo currently. We got a flag and full logo here: https://missionpinball.org/about.html. I guess vectorizing the flag would work. I would do that in Inkscape and then export it for kicad.

Jan

[cobra18t]

I vectorized the MPF flag and the OPP logo again to make them more consistent and more likely to be printed correctly:

[cobra18t]

Got them...now I just have through hole soldering to do.

[OPP]

Nice.  Joe would be so weirded out by that fact that the OPP logo he did for me 8 years ago is now appearing on another random board.  That board should make using OPP a lot easier for some people who do want the hassle of soldering wing boards and such.  Congrats!

[cobra18t]

I finally got all the parts to solder a board together. There is an issue where the Neo-Sol wing (a wing with a neopixel output, 4 solenoids, and 3 switch inputs) always leaves SOLENOID 0 enabled that I need to work out with OPP. Until that is resolved, I am not going to hook it up to my Tag Team.

In the meantime, I rigged it to do some neopixel testing. Here is a 255 LED test pattern in all its glory...

[Hugh R Spahr]

Sorry for being out of the loop so long, but Summer has been busy.  I can't possibly write code while backpacking with the fam, so things slow down way too much, and I apologize.

Looking at the embedded picture, I see that there are only 255 LEDs lit.  (Actually I didn't count, I just noticed that the LED in the top right corner is not lit).  Did you know that by setting the number of LEDs to 0, it assumes that there are 256 LEDs, not 255?  That should make it so that you can light that last LED in the panel.  (Of course programming the configuration is a problem because I haven't fixed that bug yet.).  I'll switch over to that thread to discuss those problems.

[Benoit PETIT-DEMOULIERE]

Hugu,

keep backpacking away from code, no issue ^^

Family first, as you told me once 

Enjoy,

Ben

- -- --- -- - -- --- -- - -- --- -- - -- --- -- - -- --- -- - -- --- --

Le ven. 17 juil. 2020 à 15:53, Hugh R Spahr <openpinba...@gmail.com> a écrit :

Sorry for being out of the loop so long, but Summer has been busy.  I can't possibly write code while backpacking with the fam, so things slow down way too much, and I apologize.

Looking at the embedded picture, I see that there are only 255 LEDs lit.  (Actually I didn't count, I just noticed that the LED in the top right corner is not lit).  Did you know that by setting the number of LEDs to 0, it assumes that there are 256 LEDs, not 255?  That should make it so that you can light that last LED in the panel.  (Of course programming the configuration is a problem because I haven't fixed that bug yet.).  I'll switch over to that thread to discuss those problems.

--
You received this message because you are subscribed to a topic in the Google Groups "MPF Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mpf-users/rRanvON8aCo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mpf-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/CAD9bh02HyH0-%2B6O1%3DCfrmUvjheNE_ary2w0e%2BW86yebyc3dJvA%40mail.gmail.com.

[cobra18t]

I am with Ben. Enjoy your time with your family. I can always wait. I have plenty to keep me busy in the meantime.

I will change the config to 0 to get all 256 commanded.

[cobra18t]

I have been using the latest rev of this board in my Tag Team for a while now and have had no issues. Jan and Hugh have stabilized the neopixel support so it can drive 512 pixels now! Is anyone interested in a board for beta testing purposes?

[Benoit PETIT-DEMOULIERE]

Hi cobra !

I'd be glad to test one.

I have not enough time and some of my wings are not reliable.

As I'm still at the wiring level, that would be feasible to switch to that.

How can we proceed ?

best

Ben

- -- --- -- - -- --- -- - -- --- -- - -- --- -- - -- --- -- - -- --- --

--

You received this message because you are subscribed to a topic in the Google Groups "MPF Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mpf-users/rRanvON8aCo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mpf-users+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/f80fd62f-fe85-40cb-a923-f8b190f82f2an%40googlegroups.com.

[mike wiz]

I would be interested in getting a board!

[cobra18t]

PMs sent to both Mike and Ben...

[Hugh R Spahr]

Not sure where you are in your final process.  If you want to start selling them, I would suggest adding something to pinball makers website in the OPP section.  That would give people a way to contact you, and make them aware that the board exists.  You would probably want to add the specs for the board and what it includes (such as number of solenoids, the power filter section, number of inputs, etc).  At one point you mentioned it might have enough inputs/solenoids, etc to drive system 11 style games.  That might be interesting or give people a quick understanding of about the types of machines it could drive.  If you want, I could write up a small blog post with your contact info that it is available on the OPP blog.

Random other note.  Was checking the stern feature matrix for Avengers Infinity and had to laugh that one of their selling points was 25 RGB software controlled LEDs.  It made me chuckle.

[cobra18t]

Right now, I want a couple people to work through some beta testing with me. I will be adding some documentation to MPF Docs and can add it to Pinball Makers as well. Let's hold off on the blog post until I get a little more formalized information available. Thank you. Once I am comfortable with beta testing and have some documentation out there, I will gauge interest for a Kickstarter to do a small production run.

25! That is amazing!

[Electric Euphoria]

I am interested as well, I have 2 machines I want to revamp.

A gottlieb system 1 cleopatra and a williams flash

[nickdude]

@cobra18t, I would definitely support a kickstarter for this, I've been following MPF and OPP for a while, but finally have the room and time to start working on my project (thanks COVID).

The one comment I would make, is that since the transition to the STM32 board, the documentation is really confusing as it's in a state of transition. There is still some stuff out there that really only makes sense for the old control board on the pinball makers wiki. I would make sure the new docs are up to date, and even just a section that talks about experimental or potential roadmap features would be really helpful. 

As I get started down this path, I'd like to contribute back, if someone could let me know where to star, I could suggest updates to the wiki as I run across something that doesn't line up.

Thanks for this project, it's really amazing!

[cobra18t]

That is great, Nick. We are trying to catch up on the STM32 documentation, but it is still a work in progress. Jan was able to merge some changes I made to the MPF docs (dev only) to include some info on the CobraPin board. Here is the main page:

https://docs.missionpinball.org/en/dev/hardware/opp/cobrapin/index.html  

[Nicolas Seritti]

Thanks for the update, this is very helpful.

Out of curiosity, with the new controller, am I correct in assuming that chaining is no longer a thing? (Not specifc to the cobrapin)

--

You received this message because you are subscribed to a topic in the Google Groups "MPF Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mpf-users/rRanvON8aCo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mpf-users+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/f83739cd-d1ba-4649-b582-6acaa7856240n%40googlegroups.com.

[cobra18t]

Correct. Every STM32 is its own chain and requires a USB cable to the host.

[huesito_williams]

wow, nice project @cobra18t!

it's possible to test the pcb? or to buy one? 

I started an opp pcb for my table a couple weeks ago, but now I've seen yours and it looks very promising :-) I've been using opp since a year and I'm very happy, but I want something more compact than the wings option. 

Thanks!

[cobra18t]

Thank you!

I have shipped out all the beta test boards that I had available.

I think there is enough interest to organize a batch build of the boards. This will probably be in the form of a Kickstarter. I have used Kickstarter in the past and it was a good way to handle the logistics of preorders for a batch build. I will draft the Kickstarter page over the next week so that people can preview it.

[Dan - aka BorgDog]

thanks cobra18t, interested in trying this out as well.

[cobra18t]

OK, I have a Kickstarter drafted for this board and it is approved for launch! I intend to launch this Friday (or Saturday depending on how far you are from Los Angeles time). You can sign up to be notified of the launch here: https://www.kickstarter.com/projects/1862357036/cobrapin-pinball-controller

You can also preview the campaign here:

https://www.kickstarter.com/projects/1862357036/cobrapin-pinball-controller?ref=502j4c&token=922c3a91

I started a page on Pinball Makers to start documenting it, but it is in very rough draft form:

https://pinballmakers.com/wiki/index.php?title=CobraPin

[cobra18t]

Just to follow up, I did launch the Kickstarter for the CobraPin board last Friday. It has already met its funding goal! Thank you all.

If you still want to check it out or get a board yourself, here it is: http://kck.st/2KjE8B1

[pinballnugget]

excellent work, have been following this closely.. 

regards 

christian

[Gabriel M.]

I'm in for one. Hopefully we can get these as an available product for years to come

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/c5b5911e-c9a1-4d07-9c48-c2aefa6c2326n%40googlegroups.com.

[Krayon]

Hi cobra18t,

Great work, it looks amazing :D

I was the person who wrote to your Kickstarter FAQ asking where one can find the schematic and related files. I was saddened to read your reply stating that you were not currently making any PCB files available and didn't know if you would in future. Unfortunately having been bitten before, I cannot support a project myself that is closed in this fashion. I decided to post here just to bring your current position to the attention of anyone else choosing to back your project (you don't appear to have added my question to the FAQ). The other reason is that unless you have a specific license with OPP/Hugh, you may be breaking the license by not sharing this information as outlined earlier in this thread [1] so thought I'd bring it to your (and Hugh's) attention.

Anyway, great work again, I love it.

Side note to MPF and OPP: Thanks so much for your amazing work. I've only just started using them but am blown away!

[1]:

> 10.  You can use the OPP logo.  I would be honored.  Here is the legalize.  OPP is a GPLv3 project (a copy left license).  As such, any works that are based on OPP stuff end up being GPLv3 projects also.  (That's the difference between GPLv2 and GPLv3).  Officially, the work must be publicly released so others use it, and base new projects off it.  Reality wise, it would make me happy if somebody ask for Gerbers for the board, they would be provided.  I personally put, I guess, years into the project in hopes that others would be able to base other projects and get use out of it.  When I see others using the work, it makes the time that I put into it that much more valuable which makes my small contribution more valuable.  That's why I started this paragraph with I would be honored.

Regards,

Krayon

[Gabriel M.]

Being based off OPP...

I thought this project would be open as it is a carrier board based on the open pinball project.

Otherwise we get fked like fast controllers and ben heck's boards

Tell me this isn't so

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/44acc9cf-d55b-44fe-a833-c4cad7d23b87n%40googlegroups.com.

[jabdoa]

Afaik the firmware is the original OPP open source firmware and Cobra also contributed to that. Same for the MPF part. The board layout parts currently appear not to be open. Not sure if they have to as that is complimentary. I guess the schematics would be unproblrmatic as they are quite similar to OPP wings. Sharing the Gerber is another story.

Jan

[Krayon]

> The board layout parts currently appear not to be open. Not sure if they have to as that is complimentary. I guess the schematics would be unproblrmatic as they are quite similar to OPP wings. Sharing the Gerber is another story.

It's certainly against the spirit of the license (GPL). I'm not a lawyer, but the GPL appears to be applied to the entire OPP project which would include the schematics and board layouts. To my understanding, as this is derivative, this would therefore also be under the same license. All the OPP contributors could of course still agree to license it under other terms outside of the GPL to cobra18t which may indeed have happened.

Krayon

[Hugh R Spahr]

I can only respond for OPP, but everything that is in the OPP repository is GPLv3.  That includes firmware, python scripts for regression testing/programming, schematics in KiCad (although many revs back from the current version, somebody did just ask me to verify I can open them on linux and the latest version of Kicad and it opens them properly), Gerbers (stored in the repository so Kicad isn't needed to reproduce the boards), OPP pinball framework (who the Hell would use that?), and even Qt code to match pinball players in leagues to maximize the different number of players you play to get a more accurate/less random rating, etc.  GPLv3 is a copyleft license.  (My other post about GPLv2 vs GPLv3 was actually incorrect.  I was trying to compare GPL versus non-GPL licenses such as MIT or certain versions of CC (creative commons).  I should never write posts like that without going back and re-reviewing my notes on licenses.    Sorry for any confusion.

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/b50ded4c-0b28-4130-87c8-624ce83cb868n%40googlegroups.com.

[cobra18t]

Hello everyone,

CobraPin schematics and gerbers will be released after completion of Kickstarter fulfillment. The reason I am waiting until then is to focus on delivering and supporting the people that backed the Kickstarter. Also, I will not consider the design complete until after further testing and verification--I am doing a board revision right now based on feedback from the Kickstarter and beta testing. I will add this to the FAQ of the Kickstarter page to make it clearer for people.

I look forward to our continued collaboration. This has been a fun project.

[jabdoa]

One last remark on the topic: GPL (no matter which version) is a license for source code (explicitly). It is not adequate for documentation or hardware layouts. There are licenses such as creative commons for those kinds of things. (Same with MIT which we use for MPF; MPF docs is also CC licensed).

Jan

[Hugh R Spahr]

I am not a lawyer, nor do I play one on TV.  I ended up using GPLv3 because of this wikipedia page on Open Source Hardware (OSH).  https://en.wikipedia.org/wiki/Open-source_hardware.  "Openhardware.org recommended in 2012 the TAPR Open Hardware License, Creative Commons BY-SA 3.0 and GPL 3.0 license."  I take that sentence to mean GPLv3 is sufficient (but maybe it has to be all three of those licenses?)  The original software/firmware I wrote was GPLv2.  Then, noticing that it was insufficient for hardware, I ended up upgrading (the so called GPLvX or later) to GPLv3 to mark my intentions for the information.  While Creative Commons licenses were released at the end of 2002, I didn't know about them.  I was more aware of MIT, BSD and GPL licenses.  Some of the original firmware that I wrote predates the released year of the CC license, so it wasn't an option.  Promoting the firmware from GPLv2 to GPLv3 seemed the easiest and most logical option.

Well, that's the history, and why I chose what I chose.  After writing the above paragraph, I suddenly think to myself how much of a loser I was even in the 90s.  Maybe it was because I was more idealistic and less cynical.  Yeah, that's the ticket.  If somebody can definitively tell me I chose the wrong license, please do.  I am always interested in using the correct license to make my intentions clear.

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/ef60d3ab-985f-47ca-936e-2d58907f6adfn%40googlegroups.com.

[huesito_williams]

As far as I know GPL is only used to protect software code. TAPR would be more appropiate to electronics. However you can use the GPL license to show your intentions. In this case with OPP it's clearly an open hardware and software project with the intention of SHARE the knowledge and passion for this hobby. 

I cannot imagine to sell a product based on OPP without sharing the project, to me it's the base of this whole thing. 

I have enjoyed making projects with OPP and that's thanks for people that has invested many hours of their free time to SHARE this. 

Anyway, the board from Cobra it's an excelent idea and can help a lot to people to start a new project, but I think everything must stay whithin open hardware and software. 

[cobra18t]

@huesito_williams, CobraPin files will be released on Github after I am done fulfilling the Kickstarter rewards. I need to focus on making the final board changes, testing, verification, and delivering boards in the short term.

[Gabriel M.]

I am in on the kick starter, any way for an unpopulated version?

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/041a0af8-e80d-4d08-967b-7d8c67e38f23n%40googlegroups.com.

[cobra18t]

Thanks for the support! If you want bare boards, the gerber files will be released after the Kickstarter fulfillment is done, so you can have them fabricated. I am not supplying any bare boards. There are over 160 surface mount components in addition to all the through-hole stuff, so gauge your appetite for tackling those.

[Gabriel M.]

...I need a covid project LOL

But seriously... This board will be a game changer

To view this discussion on the web visit https://groups.google.com/d/msgid/mpf-users/76d12f2f-92f5-4a55-9901-45963596bf19n%40googlegroups.com.