Forgive my technical ignorance, but does this info resolve node board-gate?

There's still lots of unknowns like what's going on in the boot-up code, is it proprietary, is it encoded.

This just makes it more likely people can service these after problems occur. But there's no guarantee problems can be fixed: there's still dependencies on those unknowns, such as bricking scenarios.

And the schematics they did post are NOT of every board or every incarnation of every board.

:S but they did post Something. Which honestly; was surprising to me.
Glad they may some kind of effort.
Honestly; I think this thread should continue until the post schematics for every node board... and we as a community should hold the ACCOUNTABLE.

It's interesting to look at the dates on the pages on the schematics. Some of them they were drafting this past week (assuming the dates are when they were originally drawn or last modified and not when they were digitally printed - which is the standard assumption in my field of engineering)

Someone else was probably right when they said about Stern needed to clean up the schematics for general users to parse through.

Quite possible someone was rushed and did not tidy things up when originally doing the schematics.

There is still some problems if you want to nitpick. Like they left the part designation number the default application generated value... ie multiple U1's with the G$1, G$2 after it on the CPU schematic.

Anyways great start to see Stern follow through and post at least some of the schematics.

I looked up them over, not bad, but I couldn’t find more then half of the boards for Game of Thrones. Maybe the rest will show up today?

What’s the difference between a SD card image, and the game code? Could the SD card image contain an auto flasher to program the micro controllers onthe node boards? Both have the same version number of code.

2BB12625-9862-45A6-81D5-227562932518 (resized).jpeg

I believe the image is there just in case something goes wrong with the SD card in your game.

It also makes it so you can upgrade by swapping SD cards instead of plugging in a USB if you so desire.

Lots of people here saying Stern wouldn't meet their time commitment and general Debbie downers. While it may not be 100% every Spike board ever and error free (my High Speed schematic still has errors BTW) it shows effort and they did a large amount of work to release this by the time they said they would. I give them credit for that.

And did you notice on the Spike 1 cpu schematics, all of the pages were out of sequence (page x of x) from what the box on the schematic says. Plus when I print double sided, the pages are way out of alignment.

 
The schematics seem made with two different tools, one looks like Eagle and the other Orcad. That makes me think that Stern does not design its electronic boards, but other companies do it on request, I doubt Stern has electronic engineers on staff, when the design of new boards is something timely, it is not a usual job as is Game design, playfield, graphic, mechanical parts, etc ...

As for Stern having to do something to free the schematics, that does not make sense, because to make the boards, first the schematics are made (with computer software, like Eagle, Altium, Orcad, etc...), and then the boards comes out. So if the boards were manufactured, without errors, the schematics were from the beginning available and also without errors, it was just a matter of making them public, nothing more.

If the schematics have not been released before, it is a purely commercial matter, probably to have the boards manufacturing and repair business controlled. Surely by the received complaints from clients, they have decided to make public the schematics, which also does not allow a 100% of repairs, if the programs of the microcontrollers are not accessible, if any one is damaged.

According to the last Stern of the Union announcement, they are also supposed to release a spike system manual:

"We will also be releasing a manual and schematics of the SPIKE system by the end of the month."

It can be harder to properly document a system than it is to design and build it

I doubt Stern omitted schematics out of malice, more likely a lack of time.

looks like eagle to me as well which is strange as you would usually contract out a big job like this to someone who would use more professional software like altium designer etc... eagle is mostly used by hobbiest as its cheap! not big companies.

Honestly, electronics in pinball machines are really not really complex enough to warrant a necessity for more professional software. Eagle is more than capable enough. Now if you were making a modern computer motherboard with 8 layers and tons of high frequency signaling, that would be another thing.

I would not discredit any of Stern's schematics or designs simply because they were made in Eagle.

I'm going to also guess they probably didn't contract these out. They may have just made a switch internally to use Eagle instead of OrCAD. And i wouldn't blame them.

The cpu board will have tons of high speed signals for video / memory.

like others have said i don't think these are the original schematics but just made for support.

The CPU has integrated memory. The video signals have maybe 4-8 differential pairs max.

I was actually looking through a motherboard schematic the other day. It was 212 pages. (see more here if you are interested https://www.opencompute.org/wiki/Server/ProjectOlympus)

Also see page 7 of this document and how it outlines 100Gbps signaling https://www.intel.com/content/dam/www/programmable/us/en/pdfs/literature/an/an684.pdf

There's a lot of additional notes on the schematics that make it seem to me like they are mostly original. It doesn't make sense to add notes on resistor size selection if they were just trying to remake a schematic to show basic layout.

Takes a look...
spike.gif
Quietly turns attention back to SAM games.

Sweet, thank you! Downloaded in case the pinpocalypse comes

This system seems very straightforward. Mostly just MCUs and shift registers to expand the IO.

I can't keypost my own posts (even if they are "key") because it has upset people previously, so I have to wait for someone else to do it. Sorry.

147 is just a link to Stern's page, people can find that easily enough on their own. I key-posted the valuable one though.

Thanks, I think one more HAS to be keyposted for the index to appear (pretty sure 2 is the minimum key posts for the index to show).

I did an initial review of what Stern has provided so far. Based on the schematics listed, I saw errors in labeled IC components by comparison of several most common node and control boards, but it was not serious. The lead diagrams were legible and non-obtuse in terms of coded electronic standards. Based on the required number of schematics, this was a start, but there is a large number of unique node boards MISSING including SPIKE II system. Both CPUs were listed.

Basically, Stern needs to keep churning them out, and reorganize them by game title (14 SPIKE titles to date) for easiest use. Added test diagnostics, node operation architecture (for the general end user), and training is still required to be developed to consider this a true success. This does not negate the difficulty for some boards to be repaired even with a digital hot air rework station and available components. Also, double check components based your boards and if there were revisions.

If people don't keep up the pressure on this situation, developments will quickly stagnate. This is why I made so much noise over the past months which helped lead to a response by Stern, but certainly not overlooking the important inquiries by owners, distributors, and operators of the entire SPIKE operating system as a whole.

Good luck Stern fans!

And they *still* have not provided any schematics for "The Pin" machines. Nor have they provided game code for them. Those are Spike system machines.

What are the part numbers labeled on the boards in one of 'the pins"? Seems at least the CPU board would match the part number used in Spike 1 games list GoT and WWE.

They are different. I have Spike "The Pin" CPU board as 520-5318-01. Spike I CPU boards are 520-5936-00 and 520-5936-01. They're not interchangeable from what I hear.

Would be nice if someone could verify that as it is secondhand information to me.

I'm happy to see that Stern is releasing these schematics. There is a lot of speculation as to why the long wait and some of it may be alluded as its very likely that Stern Pinball does not engineer nor manufacture these boards these boards but contracts it out to a 3rd party. If they are going to a 3rd party for the board design and manufacture, the 3rd party is under no obligation to provide any schematic diagrams to the purchaser unless this is a specific requirement of the manufacturing contract (which costs more money). Its entirely possible that Stern had to pay additional monies to obtain these schematics from the 3rd party and also receive permission to release them to the public. This is of course also speculation, but something to consider.

520-5318-01 Avengers The Pin cpu board

520-5318-02 Transformers The Pin cpu board

520-xxxx-xx Spiderman The Pin cpu board (I think it should be 520-6936-00)

Stern made these machines to introduce the Spike system and has NOT provided ANY support (manuals, schematics, parts lists, game code, etc) whatsoever for them. Stern had to have had schematics, bill of materials, parts lists, game code, etc to build "The Pin" machines. These machines are out there and sometime they will need repairs. Us technicians need the documentation to do repairs.

Here’s the parts list for Avengers The Pin

4D915D31-346C-48DB-803A-9042462DC92C (resized).png

5AED9CE6-15FE-4117-B53F-C0BA8EC5175B (resized).png

Did stern ever fix the IC labeling errors in the initial schematic release?

Has anyone successfully resurrected a duff node board 8 ( 520-7017-72 ) by replacing D9 diode? My Deadpool node 8 failed ,usual node 8 not found message. Poking around with my test meter and comparing readings with a good board I found D9 is dead short ( D9 is part of the 6v circuit so explains the lack of working 6v on this board ) It's the same diode that stern issued a bulletin for with a half assed repair with a cable tie. In that case the repair is to prevent a broken leg due to vibration ( so open circuit ) but on this occasion D9 is dead short. I've a replacement for it on the way, I'm wondering though if something else on the board could have gone causing this diode to fail. Does anyone have any experience with this repair? The machine itself is fine, I've installed a board off another machine to prove that.

To answer my own question above, just replacing D9 wasn't going to do it. I just removed D9 after obtaining a replacement but on closer inspection there's further damage to the board beyond my capabilities to repair, a little square thing U9 looks blown. The board has to be replaced at my cost, it's barely more than a year old. The spike system is shit, no more Sterns for me!.

If you want to send the dead board to me after you get a new one, I'm still searching for a blown one i can write up repair instructions for.

I can't pay you other than attempting to repair your board and potentially having the community benefit from that. International shipping might be a killer though.

You know there is going to be a parade in your honor if you've figured out how to fix these things.

Try to send Borygard a PM. He may be able to fix it.

I've offered to work out the repair procedure, but I need someone to send me a broken one to work it out with. So far nobody has.

when that catch diode in the SWMP breaks the output voltage will raise up 50% or so and kill other devices. I found this out by forgetting to stuff the catch diode in lm2679t circuit. Luckily i had the other boards unplugged. The 5v LED was really bright and when I checked voltage it was like 8vdc instead of 5vdc.

My best guess is that is what happens to these node boards when d9 breaks. Good chance many logic chips would be killed.

All the more reason Stern should have had a seperate +5 and +12 volt power supply. By sending all boards +48 volts and having those boards regulate the +48 down to +5 volts is just asking for major trouble.

Spooky and American use seperate +5 and +12 volt power supplies.

Can we over-spec that diode with one less likely to fail?

Sounds like vibration is the issue. Is this board near the flippers?

The root cause is a harmonic vibration that certain playfield layouts make worse.

There is one really close to the bottom left flipper in at least iron maiden and Jurassic park. Flippers very likely causing the vibrations that will be a problem at some point. I'd speculate shaker motors and sling shots, etc contribute too.

Mounting the boards on some kind of anti vibration grommets would probably help.

I don't think they are failing from over load, just getting vibrated to the point that the, leads, case or solder joints cracks. They are using a diode in a to220 transistor looking type package and zippy tie it to another mosfet coil driving looking transistors with a nylon space between them to reduce vibration. Maybe an R-6 type package diode mounted like example "A" below would be a better choice. R6 diodes have quite substantial leads on them. A to220 seems like it would be a vibrate a lot worse with the heavy part at the top hanging off of two transistor like flexible leads. Stern would have to change the PCB layout tho. An R-6 package will probably not fit.
Untitled (resized).png

Have you contacted Stern or your Stern distributor? It might be under warranty.