This circuit is nearly identical to the original KK7B R2Pro downconverter design. I used a slightly higher-level mixer (13 dBm rather than 7) which generally improves dynamic range. I used a Mini-Circuits splitter (TSC-2-1) instead of a  hand-wound unit. Transistors are 2N4401 rather than 2N3904 because their noise figure is slightly better; the ‘4401 is an old secret in the audio business. Finally, I used an emitter follower at the output rather than a final gain stage. My next stage has a low-noise instrumentation amplifier at its input, which is provides needed gain while rejecting common-mode noise and also breaking up ground loop hum.

I re-biased the circuit slightly to optimize return loss at the input to the common base amplifier. To do this, I used a resistive return loss bridge that was described in the 1996 ARRL Handbook. That way, the diplexer is accurately loaded at all audio frequencies.

By the way, I did experiment with several other amplifiers. Candidates were: a common emitter amplifier with two transistors in parallel; a common base but with two transistors in parallel; and a couple of low-noise opamps (AD787 and LT1028). Slightly lower noise was achieved with the dual-transistor common emitter circuit but at the cost of much higher complexity, especially a number of large electrolytics. It is, in fact, hard to beat a noise-matched common base in this application, especially if you want to avoid a very high parts count. And besides, it’s a case of diminishing returns. Even a noiseless amplifier would not significantly improve the noise figure of the complete receiver.

An I/Q Downconverter

The diplexer was tweaked for a nearly-perfect 50 ohm match from dc to over 100 MHz. A simple improvement you can make even without a return loss bridge is to compensate for the dc resistance of the last big inductor that goes to ground (~33 mH). Connect an ohm meter in place of the mixer IF output, and adjust the resistor in series with that inductor until you get 50 ohms.

Power supply rejection is enhanced by including a local voltage regulator. The LM317 is a very common part and actually has about as good a noise level as you will find. Remember that this is an audio amplifier, and your direct-conversion receiver has enormous gain. That means any power supply noise gets amplified by something on the order of 100,000 times!

The output goes through a mini-DIN connector. Since the audio feeds a differential amplifier, I used a balanced RFI filter consisting of a common-mode choke and a couple of capacitors.

The enclosure is fabricated from welded 1/8 inch steel plate because the inductors make this circuit susceptible to hum pickup, so ferrous shielding is advised. The box is lined with thin copper foil to provide a really good RF shield, especially where the cover meets the body of the box. SMA jacks are tightened after the box is closed up, so all the shielding and the connector grounds are very well clamped.

Other Little Features





Rcvr HomeReceiver.html

It’s a fact that large inductors in audio systems are susceptible to local magnetic fields, and the very high gain of a receiver system aggravates the problem. I first discovered I had a hum problem when I gathered all the receiver modules together, near the power supply. Even though the downconverter is magnetically shielded, and the power transformer is toroidal and enclosed in a steel box, 60 Hz harmonic buzz was clearly audible. Here is the power spectrum of that noise:

About Hum Pickup

Trying to measure minimum detectable signal sensitivity is almost meaningless with all this buzz. The situation was remedied by moving the downconverter to the far opposite corner of the receiver chassis; distance is your friend when it comes to magnetic (or electrostatic) coupling.

I had a conversation with Rick, KK7B, about this. His answer is simple. The R2-series receivers were primarily designed for battery power operation. Or, you run it on a well-filtered dc supply located several feet away. In either case, you have removed the source of the intense magnetic field and all is well. So be warned! Don’t put this downconverter right on top of a big transformer...