AW: [sdiy] Having a hard time finding the chips you want - roll y our own!
Czech Martin
Martin.Czech at Micronas.com
Thu Aug 30 12:20:49 CEST 2001
> it with a CNC mill.You can figure out exactly what the
> content of the chip
Well. this may work for the matal layers, and perhaps the poly layer, too.
But Well, N-implant and P-Implant etc. look exactly the same after milling
it down. Special chemical preparation will be needed to make these
observeable.
We "reverse" engineer only our own chips (ie. we look into rejected
or otherwise malfunctioning parts in order to find out what went
wrong and to improove).
The procedure is something like:
Plastic encapsulation is jet etched.
Metall Layers are chemical etched.
Then substrate is etched in various ways.
AFAIK no mechanical milling steps are involved here,
but could turn out usefull.
Additional methods are x-ray and ion-beam milling,
the later could be regarded as mechanical milling in nano world.
All I wanted to say is that you must be skilled, need good recipies
a lot of talent and patience. Deep burried layers as in
bipolar processes can cause special grievance.
It is a lot of work. If you pay a third party doing it,
it will therefore cost 10000s of $.
And if it comes to measurement: it is already a hard work
to measure something meaningfull on your own designs.
If you finally have all the layers (assuming you know the process),
you'll have to draw a schematic and understand the circuit.
This can be non trivial as well.
Historical note:
It is possible to successfully reverse engineer, some people
in the former GDR spend a good deal of their working life
reverse engineering western technology. Later on these
chips were produced in the east and got strange names
like L110 etc.
(buy this I won't say that all of their electronic
stuff was "stolen")
Even today they are about the best people you can get
for this job.
m.c.
More information about the Synth-diy
mailing list