Re: NEC-LIST: circular to rectangular waveguide adapter

From: Dan Bathker <dab_at_email.domain.hidden>
Date: Mon, 24 Jul 2000 10:41:58 -0700

For any reasonable diameter of the circular guide I have frequently
assumed Z1 (rectangular, TE-10 mode) = 1.0 and Z2 (circular, TE-11
mode) = 2.0 and then a single quarterwave (in *guide* wavelengths)
transformer requires a Z' = sqrt(Z1Z2).

The quarterwave section has a crossection that provides the Z': It is
called a "Pyle" waveguide, one having a rectangular outline with the
inner 4 corners truncated by a circle. Not difficult to machine.
Article by Pyle (Australian) in old MTT gives the cutoff frequency as
a function of height/width. Knowing cutoff frequency gets you to Z'
and guide wavelength. Sorry but I don't have the Pyle reference but
look at IEEE MTT Vol15, No 2, pp128-129 Feb 1967. No trouble with 15%
BW.

The Pyle crossection can simultaneously be thought of as the right
impedance transformer in a circuit-sense and a mode transformer in the
EM sense (albeit going from TE-10 rectangular to TE-11 circular is not
requiring much spacial field transformations).

For more exact design, when you can freely select the round waveguide
diameter it is best to make cutoff frequency of the round TE-11 guide
equal to cutoff frequency of the rectangular TE-10 guide. Then the
relationship Z1 = 1.0 and Z2 =2.0 is exact, in some sense, (in the
waveguide impedance voltage-power definition sense) AND the impedances
"track" with frequency (same dispersion). This way you can use
several quarterwave sections and get octave bandwidths. (See IEEE MTT
above).

Regards

dan bathker

At 03:38 PM 7/23/2000 +0200, Zvi Frank wrote:
>Does anyone have a design for a circular to rectangular waveguide
>adapter in WR28 with about 10% to 15 % bandwidth? I would like it to
>be as thin as possible.
Received on Tue Jul 25 2000 - 17:16:54 EDT

This archive was generated by hypermail 2.2.0 : Sat Oct 02 2010 - 00:10:40 EDT