DIY Transistors and Homebrew ICs

Thread Starter

Robin Mitchell

Joined Oct 25, 2009
819
Hi all,

I remember some time back about a discussion with DIY transistors and ICs being possible with the general answer being no. Of course, Jeri Ellsworth made DIY transistors but had to use tape to create protected regions on silicon.

I truely believe, just like PCBs, one day people will be able to custom fab chips and transistors (for use or just fun). I am current writing up a project (just about finished) about home photolithography and would like to show just how small you can make something at home!



This design is approximatley 3mm across making the individual wires 64.75um in width. This is dry-film resist on copper exposed with a white 3mm LED in a custom projection system for 1 hour.

With more time and care I cant see a reason why this will not be possible in the future. People could bring back chips that are no longer available (4004 anyone) without using FPGAs or micros. In the future DIP packages could become rare and so people end up making their own. Who knows!

In the next year or two I intend to repeat Jeri's experiment and then apply this to the fabrication so that I can make teeny tiny transistors.

All the best,
Robin
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
Producing small features is obviously possible, but I think the main dificulties in making home-brew transistors/ICs would be getting clean-room conditions and obtaining raw materials of sufficient purity.
 

Thread Starter

Robin Mitchell

Joined Oct 25, 2009
819
Producing small features is obviously possible, but I think the main dificulties in making home-brew transistors/ICs would be getting clean-room conditions and obtaining raw materials of sufficient purity.
Clean room condittions - Not true.
Manufacture during the 1960s did not require extensive clean rooms. Clean rooms are needed when feature sizes become small enough that common dust can cause problems during manufactire. A room with simple dust extraction and with airflow pointing downward would be sufficient to keep the dust count low.

Raw materials of sufficient purity - Alibaba
https://www.alibaba.com/product-det...4788263.html?spm=a2700.7724838.0.0.O78Cj1&s=p

I could make transistors of 200um to 400um in width using the dryfilm technique. If Jeri can make em Im sure they can be made smaller at home!
 

jpanhalt

Joined Jan 18, 2008
11,087
Working to 0.001" (25 um) over several inches with relatively modern machinery is not hard. So, I agree that homemade devices in that range of precision is doable.

What are you using to etch? If you are not using ferric chloride, it might improve your resolution.

John
 

Thread Starter

Robin Mitchell

Joined Oct 25, 2009
819
Working to 0.001" (25 um) over several inches with relatively modern machinery is not hard. So, I agree that homemade devices in that range of precision is doable.

What are you using to etch? If you are not using ferric chloride, it might improve your resolution.

John
Ferric chloride is not good for this size. I was thinking about electroplating the copper off. Do you know of any etchants that may do well for this size?
 

jpanhalt

Joined Jan 18, 2008
11,087
Ferric chloride is said to give the best resolution and least undercutting of all the common enchants. Why won't it work for that size?

Electroetching would be a good alternative. One option is to electroetch about 80%, then finish with chemical etch. That avoids the problem of maintaining electrical contact with everything that needs to be removed.

John
 

jpanhalt

Joined Jan 18, 2008
11,087
The electro-etch process I mentioned was described many years ago by chemelec and posted on ETO or ehemelec's own site, as I recall. I was busy seeing if I could still find it to provide a link, but I cannot. I don't recall specifically what he used as an electrolyte. Maybe dilute sulfuric acid.

John
 

atferrari

Joined Jan 6, 2004
4,764
The electro-etch process I mentioned was described many years ago by chemelec and posted on ETO or ehemelec's own site, as I recall. I was busy seeing if I could still find it to provide a link, but I cannot. I don't recall specifically what he used as an electrolyte. Maybe dilute sulfuric acid.

John
Maybe this .
 

Thread Starter

Robin Mitchell

Joined Oct 25, 2009
819
@atferrari I know this question is aimed at John but undercutting with etchants is when the etchant has a nasty habbit of etching the material below the mask. When the chemical begins to etch the surface copper the copper underneath the mask is left unetched. But eventually the "sides" of the unetched copper come into etchant contact and begin to be etched. This results in traces and features becoming thinner and eventually being totally etched.

Slow etchants have a tendancy to undercut (im sure its slow etchants).
 

atferrari

Joined Jan 6, 2004
4,764
@atferrari I know this question is aimed at John but undercutting with etchants is when the etchant has a nasty habbit of etching the material below the mask. When the chemical begins to etch the surface copper the copper underneath the mask is left unetched. But eventually the "sides" of the unetched copper come into etchant contact and begin to be etched. This results in traces and features becoming thinner and eventually being totally etched.

Slow etchants have a tendancy to undercut (im sure its slow etchants).
Gracias Robin. So now I know the name of what happened to my first PCBs in a distant past...
 
Top