Home-made SSI chips

Hi all!
Please don't laugh.. I'm serious.

After having made PCB's with the photographic-like technique, and
being very pleased with the resolution I've been able to achieve,
I was wondering what would it take to make my own SSI chips?

Of course I'm not talking about 90 nM resolution.. but something
like 0.1 millimeters (at best). I have a 2400²dpi laser printer.

How does one make a transistor? Is this technology reachable from
the (enthusiast) hobbyst? Where could I find the necessary materials?

Thanks!
John

JJ

Making transistors typically needs much smaller geomtries to do
anything beyond a laboratory curiousity, though to a first
approximation transistors are scalable upwards to 100um geometries.

However, the chemicals and gasses you'll need are not something
you want to mess with. You'll likely not be able to get them
anyway. Many will kill on contact or inhalation. A few will
combust spontaneously on contact with air.

Let us know where you live so we can all be sure to be far away
when you try.



JJ wrote:
> Hi all!
> Please don't laugh.. I'm serious.
>
> After having made PCB's with the photographic-like technique, and
> being very pleased with the resolution I've been able to achieve,
> I was wondering what would it take to make my own SSI chips?
>
> Of course I'm not talking about 90 nM resolution.. but something
> like 0.1 millimeters (at best). I have a 2400²dpi laser printer.
>
> How does one make a transistor? Is this technology reachable from
> the (enthusiast) hobbyst? Where could I find the necessary materials?
>
> Thanks!
> John
>

In article <00tbb.539288$YN5.363846@sccrnsc01>, Paul <> wrote:
>Making transistors typically needs much smaller geomtries to do
>anything beyond a laboratory curiousity, though to a first
>approximation transistors are scalable upwards to 100um geometries.
>
>However, the chemicals and gasses you'll need are not something
>you want to mess with. You'll likely not be able to get them
>anyway. Many will kill on contact or inhalation. A few will
>combust spontaneously on contact with air.
>

but what if he didn't use silicon or germanium based transistors? There's
a lot of buzz lately about organic semiconductors and there have been some
demonstrations where transistors are 'printed' using inkjet printers.

Unless they can afford an FPGA dev kit, hobbyists are pretty much locked
out these days - there's also the issue of surface mount components that
are pretty much impossible to solder using hobbyist equipment. So
something like this where you could draw a schematic (or write in HDL),
synthesize and print a circuit onto a transparency - well, that would be
very cool for the hobbyist. Actually, it could be pretty nice for
prototyping in industry as well, I suppose.

>Let us know where you live so we can all be sure to be far away
>when you try.

Kind of like a meph lab...


phil

> Unless they can afford an FPGA dev kit, hobbyists are pretty much locked
> out these days

There are FPGA starter kits around the price of a good ink-jet printer. If a
hobbist can afford one he/she can afford the other.

> - there's also the issue of surface mount components that
> are pretty much impossible to solder using hobbyist equipment.

That's not true. I and many many other hobbist soldering SMD parts
routinely. It seems harder than it actually is. I would even say it's faster
to work with SMD parts than with traditional THD ones.

> So
> something like this where you could draw a schematic (or write in HDL),
> synthesize and print a circuit onto a transparency - well, that would be
> very cool for the hobbyist.

Cool indeed. This technology is so new however I would afraid the price of
the materials are quite high for the hobbiest. There's couple of more
problem to solve too. One is contacts: how would you put leads on such a
cirtuit. You almost certainly can't solder anything on them. One solution
might be to use metal clips on the sides. Not sure though how many pins can
you apply that way. The other is packaging. I'm almost sure that such a
device would be most sensitive to mechanical stress. how would you protect
it?

No nit-picking, I'm truly interested.

Regards,
Andras Tantos

In article <3f6f8736$>,
Andras Tantos <> wrote:
>> Unless they can afford an FPGA dev kit, hobbyists are pretty much locked
>> out these days
>
>There are FPGA starter kits around the price of a good ink-jet printer. If a
>hobbist can afford one he/she can afford the other.
>
>> - there's also the issue of surface mount components that
>> are pretty much impossible to solder using hobbyist equipment.
>
>That's not true. I and many many other hobbist soldering SMD parts
>routinely. It seems harder than it actually is. I would even say it's faster
>to work with SMD parts than with traditional THD ones.

Perhaps. I haven't played with that sort of thing for years.

>
>> So
>> something like this where you could draw a schematic (or write in HDL),
>> synthesize and print a circuit onto a transparency - well, that would be
>> very cool for the hobbyist.
>
>Cool indeed. This technology is so new however I would afraid the price of
>the materials are quite high for the hobbiest. There's couple of more
>problem to solve too. One is contacts: how would you put leads on such a
>cirtuit. You almost certainly can't solder anything on them. One solution
>might be to use metal clips on the sides. Not sure though how many pins can
>you apply that way. The other is packaging. I'm almost sure that such a
>device would be most sensitive to mechanical stress. how would you protect
>it?
>
>No nit-picking, I'm truly interested.
>

I'm engaging in pure speculation here... I belive these organic
semiconductors are made to be flexible - there was even talk about
printing them on on clothes so you could put sensors directly into
clothing.

As far as mounting the plastic goes: I suspect you could have soome sort
of spring loaded frame that clamps onto the piece of plastic and makes
electrical contact with pads printed at the edge of the plastic.

Anyway, it makes for interesting speculation. These devices are still
apparently very slow, but for most hobby projects that would be fine.

Phil

Phil Tomson wrote:
> In article <00tbb.539288$YN5.363846@sccrnsc01>, Paul <> wrote:
>
>>
>
> Unless they can afford an FPGA dev kit, hobbyists are pretty much locked
> out these days - there's also the issue of surface mount components that

Not quite true. While it's not exactly cheap, you can get
processed, packaged die from Mosis for $1k or for a "Tiny Chip".
I know lots of hobbies which can set you back more than that.

Paul

JJ wrote:
> Hi all!
> Please don't laugh.. I'm serious.
>
> After having made PCB's with the photographic-like technique, and
> being very pleased with the resolution I've been able to achieve,
> I was wondering what would it take to make my own SSI chips?
>
> Of course I'm not talking about 90 nM resolution.. but something
> like 0.1 millimeters (at best). I have a 2400²dpi laser printer.
>
> How does one make a transistor? Is this technology reachable from
> the (enthusiast) hobbyst? Where could I find the necessary materials?

I have no afordable answer to your question but I like your idea very
much and beg you to keep us informed of your progress.

Best regards,
--
Renaud Pacalet, GET/ENST/COMELEC/SoC
Institut Eurecom BP 193, 2229 route des Cretes
F-06904 Sophia-Antipolis Cedex
Tel : +33 (0) 4 9300 2770
Fax : +33 (0) 4 9300 2627
Fight Spam! Join EuroCAUCE: http://www.euro.cauce.org/

"JJ" <> wrote in message news:Mcibb.334808$. ..
> Hi all!
> Please don't laugh.. I'm serious.
>
> After having made PCB's with the photographic-like technique, and
> being very pleased with the resolution I've been able to achieve,
> I was wondering what would it take to make my own SSI chips?
>
> Of course I'm not talking about 90 nM resolution.. but something
> like 0.1 millimeters (at best). I have a 2400²dpi laser printer.
>
> How does one make a transistor? Is this technology reachable from
> the (enthusiast) hobbyst? Where could I find the necessary materials?
>
> Thanks!
> John

I believe a number of uni's had projects that made transistors by printing them with a special ink
jet cartridge.

Read something about being able to buy the cartridges.
Also about prototyping a small cpu core by pinning
sheets of paper together, with circuit running at under 20MHz.
Was large though.

google is your friend

http://www.google.com/search?sourcei...rs+with+inkjet

http://www.spectrum.ieee.org/WEBONLY...01/nmicro.html
http://www.mstnews.de/pdf_aktuell/mst_so03_p3.pdf
http://content.aip.org/APPLAB/v83/i10/2070_1.html
http://www.physicstoday.com/pt/vol-54/iss-2/p20.html
http://www.trnmag.com/Stories/2003/0...al_060403.html
http://www.nivi.com/papers/PrintedTransistor.pdf fets


http://web.media.mit.edu/~ewilhelm/p...g_electronics/ see the list of references
epecially print your next pc

more links here
http://www.bfprinting.com/inkjetprinting/

also have a look at e-ink

On 22 Sep 2003 20:24:16 GMT, (Phil Tomson) wrote:

>In article <00tbb.539288$YN5.363846@sccrnsc01>, Paul <> wrote:
>>Making transistors typically needs much smaller geomtries to do
>>anything beyond a laboratory curiousity, though to a first
>>approximation transistors are scalable upwards to 100um geometries.
>>
>>However, the chemicals and gasses you'll need are not something
>>you want to mess with. You'll likely not be able to get them
>>anyway. Many will kill on contact or inhalation. A few will
>>combust spontaneously on contact with air.
>>
>
>but what if he didn't use silicon or germanium based transistors? There's
>a lot of buzz lately about organic semiconductors and there have been some
>demonstrations where transistors are 'printed' using inkjet printers.
>
>Unless they can afford an FPGA dev kit, hobbyists are pretty much locked
>out these days - there's also the issue of surface mount components that
>are pretty much impossible to solder using hobbyist equipment. So
>something like this where you could draw a schematic (or write in HDL),
>synthesize and print a circuit onto a transparency - well, that would be
>very cool for the hobbyist. Actually, it could be pretty nice for
>prototyping in industry as well, I suppose.
>
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

SMD technology for hobbiest is not that difficult in small size boards
and small quantities. The heat source for soldering can be as simple
as a 4"x4" hot plate. At about 270 C you can have the PCB(FR4) on the
hot plate for as long as two minutes. I have done up to 100 pin BGA
parts using 20mil pads 30 mil spacing as well as 44 pin TSOP devices
by this method. The hardest part is to slide teh board off the hot
plate without disturbing the parts. For this I have some 4 in square
by 10 mil thick ceramic plates. Slide the board onto the ceramic
plate.

There are processes to make simple single sided boards. What I used
for my senior project did 10 mil runners at 10 mil spacing with no
trouble.

To make an IC, now that is different.

james

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>Let us know where you live so we can all be sure to be far away
>>when you try.
>
>Kind of like a meph lab...
>
>
>phil