My Support Pages

Thursday, 4 January 2024

Audio Frequency Return Loss Bridge — 50 Ω —



 


Above — +/- 15 VDC input and ground ports on die cast chassis.

 
Above — Side view showing all the input and output ports.


Above — Schematic of 50 Ω differential bridge assembly. I employed a split DC supply to boost headroom and simplify op-amp biasing.I use the moderate power BD139/140 for the filter transistors: a sturdy part with low flicker noise --  no apologies.


Above — Input ports. Left: DC input (direct with a wire) using an SMA connector. Middle: AC coupled port with RCA jack. Built in 220 µF coupling cap allows testing of 50 ohm input Z audio amplifiers with no worries about the bridge causing a DC disturbance of the biasing or current.
Right: 50 Ω audio signal generator input with a BNC connector.

 

Above — The output of the instrumentation amp U1 gets buffered by the U2a follower. Low impedance output to use a 50 Ω terminated DSO as the detector.

 

Above — In analog output direct conversion or superhet receivers that use a diode ring product detector, we often employ a simple post product detector network that some refer to as a diplexer. It's not quite a diplexer, although, it does provide a 50 Ω termination to a narrow band of RF frequencies.
You might sweep this network at AF and RF with return loss bridges to study the input match versus frequency.

Above — My current post product detector network with part values chosen to try and match from 200 Hz to 200 MHz. This proved very difficult with such a simple network because the bandwidth is huge and really this calls for 2-3 networks to get it done. However, in simple receivers, this basic network works OK. The impedance match looks terrible from ~ 1 to 4 MHz, however, trying to fix this worsened the match elsewhere.

I performed the above AF measurements with my old audio return loss bridge built in 2010. It failed recently -- and that failure prompted me to design and build this new AF return loss bridge.

Compromise is a key term in simpler RF design. The network components shown gave me the best overall input Z match from 200 Hz to 200 MHz. This network also provided decent low-pass filtration of the RF lurking in the product detector's audio output. A 220 µF (or higher value) audio coupling capacitor helps keep the input noise down in the AF preamp.


Above — A 50 MHz wide sweep of the post product detector network in a tracking generator-spectrum analyzer. The 220 µF capacitor was removed for this RF measure.

 

Above — Testing gear used in the video: a 50 Ω Mini Circuits SMA terminator + barrel connector to 50 Ω coax -- and an RCA jack with a 2K potentiometer.


 Above — It's always fun to acquire more test gear.