[Coco] Prototype cartridge board?

[gene heskett]

On 8/8/22 02:56, Andrew via Coco wrote:
> > Don Barber <don at dgb3.net>:
> >
> > As such, i'm wondering if there is any source for a prototyping board
> > for the coco 40 pin cartridge port? What do you hardware hackers use
> > when playing with a design for a new cartridge-based peripheral for
> > the coco?
>
> Don, I don't know what other people use off-hand; but not too long 
> back (ok, maybe 6 months ago? Can't remember) I tried my hand at 
> designing an edge connector (and ideally expand it out to a full 
> prototyping board) for the CoCo, using KiCad.
>
> I had never used the software before, though I had played around with 
> both EagleCAD and a few other tools of similar nature. KiCad turned 
> out to be fairly easy and intuitive to work with; I was quite impressed.
>
> I took a file I found for the PC ISA edge connector; since the CoCo 
> has the same 0.1" center per-pin layout and sizing (back in the day, 
> people used to use a cut-down ISA board for prototyping - while there 
> were boards specifically for the CoCo for prototyping, they weren't as 
> cheap as what was available for the PC).
>
> Taking that file, deleting some parts, copy-pasting the "fingers", 
> bringing out some traces to pads...I ended up with something that 
> looked reasonable (in the 3D view) but who knows whether it would've 
> worked.
>
> That aside - there is this board design; I've never had one made or 
> anything to try it out, but it looks reasonable and workable. 
> Everything is there for you to send it off to a cheap board company 
> and have some made.
>
> https://github.com/JayesonLS/TandyCircuitsAndLogic/tree/master/CoCoProtoBoard 
>
>
> If you go this route, and need someone to take some of the extras off 
> your hands (for a price, of course!) - let me know. It's one of the 
> downsides of this kind of thing. You always end up with extras you may 
> or may not use.
>
> But I wouldn't try it using any homebrew methods (that is, something 
> like toner transfer or such); even if you end up with something decent 
> (and can align both sides perfectly) - you still would have to drill 
> all the pads - ugh, no thank you!
>
> Another option - and again, I haven't tried it yet (though I did 
> purchase a few of the boards):
>
> If you go on Amazon or AliExpress, there is a particular double-sided 
> prototyping PCB (0.1" centers on the individual pads) which has 
> "fingers" along the edge that are size about the same as the 
> individual hole pads, but slightly shifted to the hole pads; the 
> length of each "finger" is just about the same as a standard ISA 
> edge-connector "finger".
>
> I bought some of these boards, with the intention of trying to see if 
> I could make one work for prototyping - but that project is still on a 
> back burner, so to speak. My idea was to use some wire-wrapping wire 
> to solder to the edge fingers, and bring them to a set of 20x2 pads on 
> the board, then solder a 20x2 header on and go from there. It's only 
> "downside" seems to be that you have to cut away some of the fingers 
> (it has 27 fingers per side) - but for the price, that didn't seem 
> that terrible!
>
> https://www.amazon.com/gp/product/B081PHJPR7
>
> This is just an example - there might be other sellers on Amazon, or 
> you can find this same board sold by various vendors on AliExpress. 
> Just don't go with any other size or layout. I have only found that 
> this particular size PCB (80 x 120mm, double sided, etc) has the 
> fingers the right length and size. Other boards either have shorter 
> fingers, spacing wrong, etc - which won't work.
>
> Oh - and on github while you're at it, you can look up other CoCo 
> hardware projects, some of which use the expansion slot; maybe one of 
> those board designs can be used as a starting point, if you want to 
> learn whatever software EDA CAD they used, or your own, etc...
>
> I hope this helps!
>
> Andrew L. Ayers
> Glendale, Arizona
> phoenixgarage.org
> github.com/andrew-ayers
>
One of the things often missed by the proto card makers is the grounding 
"ears" present
on the coco stuff. They help control a phenom called ground bounce, and 
with the convoluted
ground paths in the multi-pack adaptor pcb, both versions, its difficult 
to make logic work unless
there is a relative long settling lag between the exciting signal, and 
the sample of the result
caused by that signal since the grounding is so poor, there may be and 
is in the mpi, multiple
state changes lasting up to 80 nanoseconds following a clock transition. 
This is likely the
reason for the 4 phase clocking E & Q clocking in the first place. My 
mpi has solder wick, jumpering
some of those gaps, and additional bypass capacitors which reduced the 
ground bounce enough to
get clean logic as seen on a 100 mhz dual trace scope.  Some of that is 
caused by the r/w signal
being delayed quite a few nanoseconds from the 6809, not quite so 
obvious with the 6309
as its a cmos, and a faster cpu, cutting that lag by 30 or more 
nanoseconds.

The 6309 being cmos could probably run at 8 o 16 times the speed imposed 
by the rest of the
circuitry on the motherboard.

IIRC John Kowalksi designed a circuit that doubled the clock speed when 
it could but it took
some cut and patching of traces on the motherboard that not too many of 
us were willing to do.

There is a possibility that replacing the 74LS ttl logic in much of the 
coco, with faster, far less
power, 74HC stuff, would reduce the logic short circuits that cause the 
ground bounce in the
first place. Untried by me however.  Spot temp measurements of the 
individual chips would
indicate the culprits. Find the coolest chip on the board then the 
warmer ones, replace the
warmer ones with 74HC parts might be a huge help. I quit looking when 
the solder wick and
extra .1 bypass caps cleaned mine up.

Cheers, Gene Heskett.
-- 
"There are four boxes to be used in defense of liberty:
  soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
  - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/>