Amiga 600 schematics nearly put into Kicad

[Acill]

Boards and parts set have been mailed off to Ian.

[Mick]

Thanks Acill, if I can get it working I'll be sure to send you a couple of PCB's as a thank you.

As an update here's where I'm at atm, I've gone back to basics so it's just A600 with the newer DAC/s-video, better to walk before I can run.



It looked really nice and orderly until I had to switch to 0 degree traces due to importing/exporting weirdness so now it looks like a website done in Comic Sans but it's better than autorouting I guess.

I'm still unsure how best to do power/decoupling, I think that will be the problem more than anything if it's something I did and not a manufacturing issue.

[rvctech]

Originally Posted by Acill

The parts included are from my working 600 I decided to strip for this project to be sure I was working with good parts.

You killed a perfectly working A600 for this?

[Stedy]

Originally Posted by Mick

I'm still unsure how best to do power/decoupling, I think that will be the problem more than anything if it's something I did and not a manufacturing issue.

For decoupling, a few primers:
https://www.allaboutcircuits.com/tec...ss-capacitors/

https://www.allaboutcircuits.com/tec...ng-inductance/

and
https://www.analog.com/media/en/trai...als/MT-101.pdf

PCB design is an intriguing topic, it can also drive you slightly mad if you're not careful.

Changing topic, I have KiCAd and a few other tools setup in my workshop now, ready for the A600 prototype to arrive. Also managed to locate my RS232 to USB adaptor for DiagROM, none of my PCs have real serial ports.

[Acill]

Originally Posted by rvctech

You killed a perfectly working A600 for this?

Got to do something to make new ones. How else do you think its done?

[rvctech]

Originally Posted by Acill

Got to do something to make new ones. How else do you think its done?

Take parts off of a dead machine? Use loose parts that were spares to start with?

What's the point of making a new one, if you killed a working one?

<rant>
This is a nasty side effect of clone Amiga boards.

One often hears that clone boards help replace "unrepairable" boards, but they often provide an excuse to strip and destroy perfectly good working hardware in order to have the "latest thing". "Want" rather than "need".

From the German forums I have heard of parts being damaged due to incompetent removal attempts from Amiga boards. Great way to preserve the Amiga!
</rant>

The above rant is not personal! It's a general commentary based on the worrying trend that I see amongst Amiga "owners" across the internet

[torsti76]

Originally Posted by rvctech

Take parts off of a dead machine? Use loose parts that were spares to start with?

What's the point of making a new one, if you killed a working one?

<rant>
This is a great example of the nasty side of clone Amiga boards.

People often state that it's to help replace "unrepairable" boards, but it often seems to act as an excuse to strip and destroy perfectly good working hardware in order to have the "latest thing".

From the German forums I have heard of parts being damaged due to incompetent removal attempts from Amiga boards. Great way to preserve the Amiga!
</rant>

If you've read the thread carefully, you will have noticed that it's about getting a replacement PCB running. There seems to be something wrong with the PCB, though.
Thus, it's perfectly reasonable to make sure all other parts are working to ensure you are debugging the real problem.

Although I'm an opponent of slaughtering working Amigas, too, I think in this case Acill's decision made perfect sense and in turn may lead to numerous dead machines being resurrected.

Also, I am pretty sure that Acill has enough soldering experience not to damage parts while removing them.

[rvctech]

Originally Posted by torsti76

If you've read the thread carefully, you will have noticed that it's about getting a replacement PCB running. There seems to be something wrong with the PCB, though.
Thus, it's perfectly reasonable to make sure all other parts are working to ensure you are debugging the real problem.

I've read the thread carefully indeed, and you may have noticed that I provided some (basic) support in prior posts?

Due to the age and rarity of any Amiga, there is very little excuse for destroying a working one. In my opinion none.

Also, I am pretty sure that Acill has enough soldering experience not to damage parts while removing them.

I did not suggest or infer such a thing.

I wrote very clearly that destruction of parts has been discussed on the German forums. I wrote this to highlight the poor outcomes that are occurring from the proliferation of clone boards. Another poor outcome is the rapid disappearance of spare parts that would otherwise be used to repair an Amiga. I was making a general commentary, but I have gone off-topic so I won't write about this any further.

[torsti76]

@rvctech First of all, let's keep this calm. I didn't want to offend you personally, as I'm sure you didn't intend towards anyone either.

I see that you amended your post and it's now much clearer to me that you just wanted to state your opinion, which I, in principle, completely share.

Amigas and their spare parts are getting rare at an alarming rate.
So, you are right in stating that there's no excuse for destroying one of those rare pieces.

However, I also understand why Acill did this. In my opinion, if this sacrifice hadn't been made there would be little to no chance this new board will ever come to life. With a dead board, as you suggested earlier, you never know which parts may be damaged.

In the end, I don't know what kills more Amigas: Stupid people with bad soldering skills or capacitor and battery leakage. For the latter ones this work is vital.
I've recapped quite a few Amigas myself and always got lucky that the boards were still usable. But there are also numerous reports where people had to do all sorts of pad and trace repairs and often ended up with instable or non-booting systems.

Let's give those poor Amigas a chance, too. As Spock said: "The lives of the many outweigh the lives of the few."

[rvctech]

@torsti76 No offence taken or intended at all

In the case of the A600, we can confidently guess that damage is almost always caused by unskilled recap attempts. In my experience leaking capacitors rarely damage copper (this process is extremely slow). Fortunately pad and trace repairs are straightforward because the A600 is a very simple 2 layer board and runs at a very low speed by the standards of today. It does take a lot of time though

[CD32Freak]

Thanks a lot, Stedy, for sharing your (Amiga) knowledge over the years and now

Share your knowledge.
It’s a way
to achieve immortality.

--Dalai Lama

[Mick]

Originally Posted by Stedy

For decoupling, a few primers:
https://www.allaboutcircuits.com/tec...ss-capacitors/

https://www.allaboutcircuits.com/tec...ng-inductance/

and
https://www.analog.com/media/en/trai...als/MT-101.pdf

PCB design is an intriguing topic, it can also drive you slightly mad if you're not careful.

The autorouter did it for the most part "decent" according to this:
https://www.allaboutcircuits.com/upl...A_DeCap2_3.JPG

What I'm wondering though is if it's better to route with traces and separate the decoupling caps more? so that you won't get multiple IC's nearby fighting over a single capacitor if that makes sense? I mean is the A600 layout optimised (or not) based on it being a trace design and not power planes? a lot of it is over my head to be honest all I can tell you is that everything is connected where it should be minus a couple of minor errors on the board I sent Acill. I've manually routed most of the board now and not found a single discrepancy compared with AmigaPCB.

I've not finished manually routing the power and some other small connections but I took some time to edit footprints to make it more Commodore and less generic KiCad. I really enjoy the design side of things but the electrical side is something I could use help with.

[Mick]

Could it also be an issue with via's being too small? I read that via's carry 3 times more than the equivalent sized trace due to pi but is inductance or something an issue?

[Acill]

Killing my spare A600 wasnt a big deal to me. I have enough spare parts and dead A600 boards here to revive more if this boards gets to a working state. i can assure you that the parts are good and that in the end more A600 machines will be around as a result in this small sacrifice to a machine that was used for a better purpose than the many other sitting in some guys attic or basement collecting dust doing nothing.

[Mick]

I'm going to increase via drill sizes to .4mm under recommendation from stedy and I might as well re-route it and try to keep with 45degree angles, it seems I was at the limit of what the manufacturer could do with 0.3mm. stedy still hasn't got the board but if I get more boards made I'd rather play it safe. Not sure if I'll be able to convince anyone to build another one at this point though.

[Stedy]

Hi,

Board arrived this morning. Have some assembly to do but it will keep me out of mischief.

I checked the drill sizes of my A600, the signal tracks have 0.45mm drill and the power was 0.70mm. I used my mini drill set (0.3 to 1.5mm) to size the drills.

In my day job, I have seen PCB failures due to vias snapping many times. Microvias, not used here are the weakest, designs at the limits of fabricators capabilities are next to go. This may not be the cause.

Will do some basic continuity tests before assembling.

[Mick]

Send me a few pics if you get the chance please stedy.

I just wanted to ask is it intentional when you have a thick (power) trace going to an IC and then it narrows just before entry other than physical fit?

[Mick]

I manually routed the signals just power left, leaving it for now though I'm a bit of a dead end. I'm hoping stedy can find out what was wrong with the first board or someone can look at the files and tell me. It looks seriously good though if you copper/ground fill the empty space.

Does anybody want some nice looking PCB's that don't work?

[Acill]

It does look good, and I want a few more boards once this is workked out!

[alenppc]

That board looks gorgeous. One thing that could potentially be useful is to replace the SMD 68k with an A500 dip one so we can solder accelerators directly instead of the 68k, but not sure if there is space for that.