[Amps] Directional coupler design software

Dr. David Kirkby drkirkby@ntlworld.com
Mon, 16 Dec 2002 01:04:52 +0000


Some may be interested that I've expanded my Arbitrary Transmission
Line Calculator atlc
http://atlc.sourceforge.net/

to include a program 'design_coupler' with the ability to
automatically generate a set of dimensions for a directional coupler
with a cross section like that shown below at:

http://atlc.sourceforge.net/couplers.html

if given the frequency range, coupling factor and optionally the
length. The software gives the coupling vs. frequency in addition to
the dimensions. The equations used are exact analytical equations -
not approximations. 

If you want to go the other way and have a transmission line (2 or 3
conductors) and want to know its impedance (or impedances in the case
of couplers), you can do this using 'atlc'. There are some examples
(and quite nice looking pictures) at 
http://atlc.sourceforge.net/examples.html
Those are generated using a computationally intensive numerical
technique which is not exact, has very small errors. Comparisons
against coaxial cable, which has an exact solution, show a *maximum
error* of only 0.186 % and an RMS error of under 0.14%. With some
other structures the errors are a bit larger, but always under 1% for
any structure (coax, off-centre coax, symmetrical strip transmission
line) considered to date:
http://atlc.sourceforge.net/accuracy.html
This excludes open-wire transmission lines, for which the software
works quite poorly, due to the lack of any well defined boundary
conditions. 

PS, The program only runs on Unix systems. There are some old
command-line drive Windoze binaries, but these do not have the
functionality to design couplers and I don't intend building any that
do. If someone else wants to take the C source and produce some
Windoze binaries they can do. 

***CAN YOU HELP ??***

If anyone with access to a Unix workstation with 2 or more CPUs (other
than a Sun) and would let me know how 'atlc' compiles if configured
for multi-processor operation, I would be interested to know. 

-- 
Dr. David Kirkby,
Senior Research Fellow,
Department of Medical Physics,
University College London,
11-20 Capper St, London, WC1E 6JA.
Tel: 020 7679 6408 Fax: 020 7679 6269
Internal telephone: ext 46408
e-mail davek@medphys.ucl.ac.uk