[sdiy] Designing Analog ICs

Tom Corbitt tom.corbitt at gmail.com
Wed Sep 7 02:18:03 CEST 2011


Hit tab and enter too soon..

$500 to $1000 a part soon becomes a pipedream. You also are not likely
to get it right the first go round, so you better factor in at least
two spins. Add in the fact that current process sizes aren't conducive
to analog chip and you really only have a few choices left. Your best
bet would be to try to find a university that accepted some old fab
equipment a decade ago and is still running grad school level training
on it (even a decade is iffy, you really want 15-20 year old stuff)

I've been down the road you're on and failed (my dream was to remake
the lm3909) and I worked for a company that had a fab branch (and gave
me some margin die space to play with) I got some chips almost kinda
worked, but there is no way I'd ever be able to afford to make the
number of chips required to sell them at a hobbyist or boutique price
level.

The positive is that most of the VLSI books on the processes and
designs you would be working on are now thrift store prices. The
design blocks you need aren't cutting edge, you can pretty much cookie
cutter them down and they should work (maybe not well)

Who knows, Jeri (Ellsworth) fabs transistors in her basement, you just
need to follow in her footsteps and prove me wrong.

 Tom


On Tue, Sep 6, 2011 at 7:55 PM, Tom Corbitt <tom.corbitt at gmail.com> wrote:
> For a real custom IC, you're off by an order of magnitude, 100's of
> thousands would be required Even If you're going with a place like
> xfab (meaning you're willing to roll your design with their prefab
> parts) you're talking min 50k plus a wafer. Say maybe 200 to 500 parts
> yield per wafer just for the die, then add on the wirebonding and
> packaging, and $500 a part
>
> On Tue, Sep 6, 2011 at 4:36 PM, Joel B <onephatcat at earthlink.net> wrote:
>>
>>> Presumably you already have the tens of thousands of dollars and you now just need to brush up on the knowledge? ;-)
>>
>>
>> That is what KickStarter is for!
>>
>> given the success of the Rockit 8bit Synth kickstarter project, hopefully someone on the list will start a Kickstarter project to re-release all those Curtis chips in quantity, or at least some really good clones, and get the Analog Synth industry out of the "scavenging for chips" mode.
>>
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>



More information about the Synth-diy mailing list