Comments for "Return Loss Bridge"

1st December 2013 00:40

Silva - PY8ELO wrote ...

Hi Alan,

congratulations for your job!

But I cant find the balun reference. You just said it is 1:1 and I would like to know what is that which You has used.

Can You inform it to us please?

Thanks in advance and best regards,



17th December 2011 17:44

Alan Yates wrote...


Yes it should, especially if you build it carefully with SMDs.

Symmetry is a huge help, if everything is equally non-ideal in each leg they tend to cancel each other out. You can also calibrate out any weirdness, but you shouldn't have to if the components and layout are reasonable.



17th November 2011 16:11

Rusty wrote...

Hi Alan. Great work! I have a quick question... will this RLB work at UHF? I am interested in 440~450mHZ range.



9th July 2011 08:22

Arv - K7HKL wrote...

Paul's suggestion is right-on.

Another didn't need to change the SMT resistors for leaded types. You can easily solder a wire onto SMD to make them leaded.

14th June 2009 15:27

Bert wrote ...

Alan, can I use this return loss bridge to measure or calibrate antenna impedance?

1st March 2009 06:45

Alan Yates wrote...


You are absolutely right. You need to terminate the detector port in the Zo, else its poor return loss will make the difference between a short and an open wrong. They should be exactly the same in magnitude, but differ in phase by 180 degrees. Such a test is a basic figure of merit for return loss bridges, right along with their directivity. A good bridge sees less than 1 dB difference between short and open detector power.

Whacking a termination resistance across the detector port will fix the problem, but drastically increase the drive required for good readings using the simple diode detector. Closer to flat the detector return loss becomes insignificant, as there is much less power to re-reflect. I've noted many commercial antenna bridges use the same design - and suffer the ~4 dB difference between open and short. My MFJ-207 does, but like my return loss bride below 2:1 the effect isn't very critical and is unimportant if you are just looking to tune for best return loss.



6th February 2009 18:00

Paul Darlington wrote ...


I write as a newcomer to radio (but an old timer in engineering HI HI).

I was inspired to make a return loss bridge by Eamon Skelton writing in RadCom, January 09, but chose to build your circuit with an integral detector.

I was interested to discover how important it is to load the secondary of the transformer with a "characteristic" resistor - without it, the system shows the 4-and-a-bit dB error you expect (from potential divider sums) between open circuit and short circuit return loss, if driven with a 50 Ohm output RF source.

Is it worth mentioning this on your site for dummies like me? Eamon does mention it tangentially in his RadCom article ("The detector port should be connected to the 50 Ohm input of a sensitive power meter") but doesn't explain why. The ARRL Handbook is equally imprecise in specifying best loading for the detector port, simply telling us to connect it to a power meter "through a step attenuator". Sure, I understand that this attenuator would usually be designed to have characteristic impedance (see below), but it would be useful to spell out the importance of this - especially in those intelligent, practical circuits, like yours, that feature an integral detector.

In playing with my RLB, I made some measurements to characterize the detector and used excel to fit a polynomial to the data to "undo" the non-linearity. This has made a big improvement in high RL measurements.

Let me take this opportunity to tell you what a big fan I am of the site - I have made some step attenuators for audio-frequency use (at 10kOhm impedance) using your circuit and pi pad calculator (I'm an acoustician by trade). I'm playing with the super regen tone dipper now.

Keep up the great work,

Paul Darlington


Leave a comment on this article.