On Sat, 20 Sep 2008 21:27:20 -0700, John Smith
wrote:

Jeff Liebermann wrote:
...
Others have suggested cut-n-try which is probably good enough. As
long as the coil looks roughly like the original, it should work.
...

Yeah, working as a software engineer and finding I get the most money
from convincing those "with money" that I am worth "some of their money"
has made me aware of the value of reverse engineering--whether software
or hardware--it works!

The very first think I did when starting a new product, was to do a
"survey" of the competitions products. If possible, some were
purchased, dissected, reverse engineered, and on one occasion, cloned.
No need to reinvent the wheel when the competition has done it for
you.

But, if you have built enough "old box tube xmitters", you can almost
guess the dia./turns/and-caps wink Heck, bet I could come up with a
circuit using the same coils, using a fet, and saving the cost of the
filament power! Indeed, if the tube uses an octal socket, in his current
GDO, why not replace the tube with a fet(s)?

Because the typical FET as a much higher transconductance (typically
about 30 millisiemens versus 5 millisiemens) and a much wider
frequency response, that you'll probably spend considerable time
stabilizing the oscillator. It's not easy to build a tunable
oscillator that operates over such a wide frequency range. Also, the
grid and gate currents are quite different. It can be done because
there are FET grid dip oscillators available. However, if you compare
schematics, they are sufficiently different to require a rebuild.

It's also criminal sacrilege to despoil such a vintage boat anchor.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558