Amiga 600 schematics nearly put into Kicad

[Mick]

New version is pretty much identical to a rev 2b board but with U49 added (whatever it is for), from what I've gathered through close inspection 2a was basically unfinished 2b in beta, 2d appears to be mostly elongated landing patterns for IC's with slightly different routing to compensate. I worked out the differences between 2a/2b by looking at via's on back of board etc but I missed 2 ground links under the ram due to a shadow and seeing as nobody will want to build/test slightly botched pcb's I uploaded the files. I've wasted too much money on it already and it's not like I'm rich, I've only been doing this for the community and because I find it strangely enjoyable. If the uploaded files don't work then the only explanation is either the schematics are wrong and there is something different hidden under an IC somewhere, it's something with the cxa2075 mod (the website I looked at did it to A1200 which doesn't use the RGB) or the original boards were 4 layers with internal power planes.

[ex68k]

Originally Posted by Mick

New version is pretty much identical to a rev 2b board but with U49 added

Is this online somewhere? Sorry, if I missed it

[Mick]

I uploaded the KiCad files to the zone a few weeks ago, it's all untested though.

[rvctech]

Originally Posted by Mick

New version is pretty much identical to a rev 2b board but with U49 added (whatever it is for)...

U49 is a reset supervisor, but on a Rev 2D that I looked at, the part isn't even loaded! Looks like it was an option Commodore considered.

The boards are definitely not 4-layer. I have some A600 motherboard "pieces" from many years ago (long story) and I can see that it's a 2-layer.

Thanks for your observations, it's interesting to see that the Rev 2A was indeed a "beta" board. Yes it's fun to develop hardware, but definitely not cheap. Even for companies I've worked for, it has been expensive. Personally it's left a big hole in my pocket a number of times. Can't help it though, must be addictive

[Mick]

It should be alright then it's just a matter of testing which isn't easy when people have lost faith.

I've spent the last 2 weeks doing CD32 but I think I will have to keep it original, it uses c-sync out from pin 11 of the cxa1145 and there isn't one on the newer dac.

[rvctech]

Originally Posted by Mick

It should be alright then it's just a matter of testing which isn't easy when people have lost faith.

I've spent the last 2 weeks doing CD32 but I think I will have to keep it original, it uses c-sync out from pin 11 of the cxa1145 and there isn't one on the newer dac.

I don't think it's a matter of faith. The A600 requires very good soldering and technical skills to assemble and debug. Laying it out is only the first step. It's also quite hard to find certain parts. I think there's perhaps less demand for it compared to the more popular A500/A2000/A1200/A4000. I can think of a few reasons for that.

CD32 boards might not be much in demand either. It takes very advanced soldering skills to assemble and certain parts are really rare. Then you need an optical drive or expansion card to make much use of it. All the CD32s I've come across were repairable, the main issues I see are leaking capacitors or faulty optical drive laser and PSUs. I would personally find it hard to justify the cost and effort to build a CD32 board or to use it.

As long as you're having fun and improving your skills and knowledge that's probably a good thing

In case you're interested, have you seen any of the retro brew computer sites? I'm active on them. There's lots of of opportunities to design cool boards and these are much cheaper and easier to work with: www.retrobrewcomputers.org and www.retrobrewcomputers.org/forum

[CD32Freak]

You could strip all the connectors from a CD32 and build a core board with only the AGA chipset and Akiko. Put some a600 and A1200 connectors around the edge and build separate video/audio/IO boards like the A4000T. The user will have the freedom to build any shape of Amiga they want. I know it sounds like a crazy idea, but there is a lack of modular design with the recent Amiga clones

[rvctech]

Originally Posted by CD32Freak

You could strip all the connectors from a CD32 and build a core board with only the AGA chipset and Akiko. Put some a600 and A1200 connectors around the edge and build separate video/audio/IO boards like the A4000T. The user will have the freedom to build any shape of Amiga they want. I know it sounds like a crazy idea, but there is a lack of modular design with the recent Amiga clones

Someone in Germany already used this approach to make an Amiga in a different form factor.

To be honest I consider any CD32 derivative design to be a waste of the last few remaining Akiko chips in the world. I already repaired a CD32 with a damaged Akiko chip, I hope I don't have to do this again! They are not so hard to damage on a CD32. It's also very hard to solder this part by hand if you are not an expert, and I would hate for people to destroy these precious chips while attempting to "use" them .

Your idea is good, in which case you might consider to also create a new "Akiko" in a FPGA . The Akiko is just an ASIC that implemented the core functions for the CD32 and added some extra stuff. An updated "Akiko" could dump the unneeded stuff (like the CD-ROM controller) and focus on only the required core parts. Because this part is an ASIC, it is very amenable to being implemented in a modern FPGA. Since you would be making a new PCB design, you could also use whatever suitable part and pin-out arrangement that you prefer. This would also be a sustainable approach that doesn't consume precious parts needed for original Amigas.

[CD32Freak]

There was a seller named Sveta over at AmiBay who sold very cheap NOS Akiko chips, but he got banned. I wonder why, because he was a very kind seller

However, there is hope. The Apollo Vampire team have transfered the Akiko to FPGA: http://www.apollo-core.com/knowledge...28256&z=XJqOl2

[rvctech]

Originally Posted by CD32Freak

There was a seller named Sveta over at AmiBay who sold very cheap NOS Akiko chips, but he got banned. I wonder why, because he was a very kind seller

However, there is hope. The Apollo Vampire team have transfered the Akiko to FPGA: http://www.apollo-core.com/knowledge...28256&z=XJqOl2

Yes I bought 2 from him a few months ago (1 of which I immediately used to repair a CD32). I did not consider them to be very cheap . He had only a limited amount left, and I don't think there are many more left in the world apart from those! (I searched). He was a good seller. I never understood why people get banned on Amibay with no explanation

The Apollo "Akiko" is an implementation of the C2P function of the Akiko to allow CD32 games to work with their core. I guess this means you can run CD32 games with no modification using their products. It's not really a stand-alone physical replacement for the Akiko if you were trying to build a re-implemented Amiga, but this is probably the future!

[PurpleMelbourne]

Has anyone decapped and scanned Akiko?

[rvctech]

Originally Posted by PurpleMelbourne

Has anyone decapped and scanned Akiko?

Unfortunately not. Also the person who designed it does not have the chip schematics anymore (someone asked her), so we can't use that avenue either, although she can probably answer questions and provide some details.

[Shiozaki1]

So how is the project going? I wouldn't mind getting the gerbers and BOM and have a few boards made up.

[Mick]

I've been working on A1200 as a hobby project, I went back to my original files and I'm in the middle of manually routing it which is taking ages.

I've not done anything with A600 in months, the interest in it (or getting help testing it at least) seemed to die down just when it looked most hopeful. I think maybe it's lack of confidence in me or relates to me using CD32Freak's files (for appearance) in much the same way he has taken them down, which is a shame because like I have said to him you can learn a lot from them.

I might go back to my original A600 files later if the A1200 goes well but from what I've been told in PM's there are other A600 projects. To be honest I can't understand why there aren't already boards available when you have about 5 different A500 boards.

[CD32Freak]

Mick, maybe you should not waste your time on the old Commodore designs, but focus more on creating something innovative like the A1200+ by Jeroen Vandezande:
https://bitbucket.org/jvandezande/am...00/src/master/

[Shiozaki1]

Originally Posted by Mick

I've been working on A1200 as a hobby project, I went back to my original files and I'm in the middle of manually routing it which is taking ages.

I've not done anything with A600 in months, the interest in it (or getting help testing it at least) seemed to die down just when it looked most hopeful. I think maybe it's lack of confidence in me or relates to me using CD32Freak's files (for appearance) in much the same way he has taken them down, which is a shame because like I have said to him you can learn a lot from them.

I might go back to my original A600 files later if the A1200 goes well but from what I've been told in PM's there are other A600 projects. To be honest I can't understand why there aren't already boards available when you have about 5 different A500 boards.

so is there a down load for the files?

[DJBase]

Originally Posted by CD32Freak

Mick, maybe you should not waste your time on the old Commodore designs, but focus more on creating something innovative like the A1200+ by Jeroen Vandezande:
https://bitbucket.org/jvandezande/am...00/src/master/

I think there are some people out there who just want a 1:1 copy of the original design without any modifications like me.

[CD32Freak]

Well then my Gerber conversions of the Amiga PCB Explorer boards were as close you could get to a 1:1 copy, but some crybabies in our community said it was wrong to copy a copy of what was illegally obtained in the first place.

[rvctech]

Originally Posted by CD32Freak

Well then my Gerber conversions of the Amiga PCB Explorer boards were as close you could get to a 1:1 copy, but some crybabies in our community said it was wrong to copy a copy of what was illegally obtained in the first place.

Jokes aside, the A600 revision 2A PCB is a preliminary Commodore design that was never released and has a number of issues. It should not be used as-is, it's not very likely to work!

[CD32Freak]

The only way to be sure is with PCB deconstruction techniques:
http://www.grandideastudio.com/pcbdt/

Or Multi-layer printed circuit board analysis system using X-ray equipment:
[ Show youtube player ]