Xref: utzoo sci.electronics:13137 rec.ham-radio:22490
Path: utzoo!attcan!uunet!aplcen!uakari.primate.wisc.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!apollo!tweed
From: tweed@apollo.HP.COM (David Tweed)
Newsgroups: sci.electronics,rec.ham-radio
Subject: Re: Has anyone made any homemade valves (tubes), semiconductors ...
Summary: it's possible
Message-ID: <4bd35f61.20b6d@apollo.HP.COM>
Date: 26 Jul 90 18:41:00 GMT
Sender: root@apollo.HP.COM
Organization: Hewlett-Packard Apollo Systems Division, Chelmsford, MA
Lines: 81

From: GMoretti@massey.ac.nz (Giovanni Moretti)
>     You can make resistors out of pencil leads, capacitors and
>  inductors are easy, headphones are possible (not easy) and you can
>  use Galena (lead sulphide I think for a diode), but what about things
>  with GAIN.
>  
>  I can't be the only one who has these thoughts ...
You aren't; you're following the footsteps of a long string of hobbyists.

From: cccph@eel.cs.ucla.edu (Charles Hobbs)
>  Transistors, etc. would most likely be too difficult to make at home
>  because of the necessity for "clean rooms" (rooms with no appreciable
>  level of dust floating around. Dust will short-out a transistor,
>  integrated circuit, etc.)
Nah, you're thinking of ICs, where the dust messes up the photomasking.
Remember, the first point-contact transistors were made by hand in the
lab.

A long time ago, I read a story about a person who built his own
vacuum tubes at home. The story was old then, and I don't remember
where I read it (probably Popular Electronics, back when it was
still in the smaller format), but I'll try to summarize.

(Disclaimer: I've never done this myself!)

Back when tubes were first being developed, most of the publicized
effort was going into the Edison/DeForest type of "tube" that was
based on the light *bulb* and required glassblowing skills. However,
there was a grassroots effort (primarily by Canadian experimentors,
I believe) that was based on using a metal *tube*. It is this second
type that is much easier to fabricate at home.

Take a piece of tubing (copper, I think) and find some hard-rubber
stoppers that fit it tightly, one with a hole, the other without.
The one with a hole is fitted with a glass tube that gets connected
to your vacuum pump.

The metal tube itself is your anode; you still need a filament and
a grid. Poke some fairly heavy-guage (#18) solid copper wires through
the stoppers (making their own holes): one in one stopper, and two in
the other. The filament (cannibalized from a light bulb?) gets
stretched down the center of the tube, strung between two of these
wires (crimped, not soldered!), and the grid is the third wire,
spiraling around that.

Remember that the filament is going to get pretty hot, so provide
a fairly-long thermal path before you get to the stopper, otherwise
you won't be able to maintain any sort of vaccuum. (The author of
the article said that he'd forgotten this, and when he powered
up his first tube, "... the vaccuum immediately left with a
piercing shriek.")

OK, put everything together and seal all joints with sealing wax.
("What's sealing wax?" :-). I'll try to do an ASCII picture:


   rubber      copper tube                                rubber
  stopper   ===========================================   stopper
      ##########     \     /-\     /-\     /-\     ##########
      ##########  /\  |   |   |   |   |   |   \---------------- grid connection
fil.-------------/ |  |   |   |   |   |   |        ##########
      ##########   +======|=======|=======|==+     ##########
   ---------------    |   |   |   |   |   |  |   /------------- filament (cathode)
to ---------------    |   |   |   |   |   |  |  /  ##########
vaccuum ########       \-/     \-/     \-/    \/   ##########
pump        ===========================================-------- plate (anode)


The loops in the filament wires help keep them from getting too
hot where they go through the stoppers. Don't worry about all
that stuff having to do with "getters", etc. Most experimentors
just left the vaccuum pump running while they were operating
the tube.

At this point, pump out the air, apply power to the filament
(carefully!) and to the anode (see if you get a current flow).
Then experiment to see what kind of control you get with the grid.
I'm sure it'll take a lot of cut-and-try to get any significant
amount of gain, so have fun!

                                             -- Dave Tweed