> Minimum width of a chip will vary by maker, but figure maybe 0.25mm
> margins, so you should be able to get a reasonable active area on a
> chip 1.5mm wide.

Even smaller would be better.

> Conventional components are a lot more practical with a 1/4" (6mm)
> OD tube.

Indeed - but that's 9 x the area.
MUST fit in 2mm.

> 2mm OD is.. tiny. Good luck with this, and hope there is
> some budget- I think you're going to be needing to use it.

Budget is not liable to be a great problem.
"Selling" the idea is the need.
What I want to do is definitely doable - it's just working out how.
Then there's the "power supply".

Matt said:

> I think the smallest off the shelf package for a PIC10F20x is a 2mmx3mm
> DFN (actually an 8 lead package- 2 leads are N/C)
That was the smallest I found.
PIC was just an example.
Many processor families would do.

> You can buy just the die, but (with Microchip) order qty on die is in # o=
f
> wafers- you buy all the parts that come off the wafer, and the minimum
> order is a non-trivial number of wafers (it ends up being a LOT of parts)=
..

Makes protyping hard :-). Volume probably not an issue long term.


> =A0Assembly of die is non-trivial and not something just any assembly hou=
se
> can do.

> Bare die assembly is sometimes called "blob top."

I've seen it done in China. I was impressed with the relative
simplicity of the process.

--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.