The Hybrid Ring Diplexer (duplexer) is a strange device compared to the standard Band Pass or Reject type.  What makes it different is the circuit that creates the pass.  In common duplexer types, the pass is either created by the band pass effect of a dual coupling loop cavity or the pseudo pass effect of a single loop cavity in BpBr, or the flat skirt leading up to the beginning of the notch in a reject (notch) type cavity or duplexer.   The hybrid ring creates the pass response from the primary characteristics of a large notch cavity, but you are probably wondering how a notch cavity can create a pass response.

The hybrid ring consists of several sections of coaxial cable, usually of a double shielded nature like RG214.  The sections of the cable are electrically multiples of a 1/4 wavelength (in coaxial cable - considering the velocity factor).  RF energy on the design frequency is injected into a Tee connector and split into two (unequal) paths then recombined back into another single connection point with another Tee.  One path is electrically longer than the other, by 1/2 wavelength.  Since one path is 1/2 wavelength longer, power split at the originating Tee is recombined but totally canceled out at the terminating tee. Why? Because one path is 180 degrees out of phase with the other, all power close to the design frequency is canceled out at the combining termination.  Still with me?  Good.  Now lets insert a notch cavity, tuned to the design pass frequency, into one of the two paths.  By doing so, power being notched out of one path eliminates the canceling effect; which results in most of the originating power being transferred to the termination by the other path.  The neat part of this whole scenario is a Band Pass circuit is created that has the inverted characteristics of a notch cavity; sharp pass peak and steep skirts.  As in any cavity, the amount of coupling determines the bandpass width and insertion loss.  Due to the circuit, a pass response that has very defined width and loss results.  Variable length stubs are used to cancel out reactances which allow the notch to be optimized on a particular frequency to be rejected.

The Hybrid Ring Duplexer was a creation of the late Gilbert L. Boelke, W2EUP.  Gil was a very gifted and talented RF design engineer, and had become a friend of mine just before his passing.  Gil sold the concept of the Hybrid Ring Duplexer to Sinclair Technologies many years ago, but he was the only man I knew who could explain it in a language I could understand.  Presented below are articles that help explain in further detail the concept and engineering behind the Hybrid Ring Duplexer.  I hope this information allows you to see how neat this circuit really is.

Hybrid Ring Diplexer (Duplexer) Explanation (in PDF) by Gilbert L. Boelke  W2EUP - SK
Hybrid Ring Duplexer Tuning Instructions (Manual CM-106) by Sinclair Technologies

The following was offered by Allan Crites - WA9ZZU by request of myself, and is included below by permission.

After re-examining the information I have from Sinclair pertaining to the response curves of a Hybrid Ring Duplexer and the tuning instructions they provide, and other foregoing information previously provided, I have concluded the following:

1) The Hybrid Ring Duplexer is primarily a band stop or band notch device which affords little in the way of band pass resonator benefits.

2) The Hybrid Ring duplexer provides some 120 dB of isolation with only 4 resonators (not including the Xc stub which does some clean-up of any Z mismatch).

3) A BP/BS duplexer needs 3 resonators in each leg to achieve the 120 dB isolation as compared to a Hybrid Ring duplexer.

4) For a 4 resonator Hybrid Ring duplexer to equal a 6 resonator BP/BS duplexer BP response there would need to be maybe 2 additional BP resonators (one in each leg).

5) Granted the construction of the coaxial "rings" in a Hybrid Ring duplexer is critical and is paramount to the achievement of excellent isolation performance (as others have already noted in being able to arrive at the necessary 1/4 wave and 3/4 wave lengths).

6) Did the other commentators here have 1/4 wave resonators or were they loaded helical resonators in their Hybrid Ring duplexers, (which would make the coupling more critical)?

7) Motorola in 1966 once had a (450-470) Hybrid Tee product p/n SP 4000 MRAB which was touted as for use in a duplexer configuration with 2 BP cavity resonators in each leg. The product never went anywhere that I was aware of but used 75 ohm coax. Perhaps others here can comment on whether their experiences in Hybrid Rings used other than 50 ohm coax.

This site, its contents, and look & feel are Copyrighted©
2001 Kevin Custer W3KKC
All Rights Reserved.

HTML  04, May 2001  W3KKC
UPDATED 12, October 2011 - W3KKC