This article documents the implemention and the testing of a MAR6/MSA06-based amplifier which can be supplied from the coax feeder (phantom power). Its intent to address the concerns highlighted above.
The MAR/MSA MMIC series is based on a Darlington pair with resistive feedback. The input and output mismatch are relatively stable over frequency because they are set primarily by resistive divider networks. A combination of series (RE) and shunt (RF) feedback desensitizes against variations in active device parameters.
To realize a phantom-powered masthead pre-amplifier, the circuit below was assembled on the PCB shown below. A 390R resistor is used to drop the 12V phantom supply voltage to the device's specified 3.5V operating voltage. The DC blocking capacitors are dimensioned to support operation down to 3MHz (80m). The PCB is 0.8mm thick FR4. The microstrip width is sized to provide 50R Zo. The ground pads have three viaholes each. To house the assembly and to provide some measure of screening, a 40x30x23mm enclosure is fabricated from 1.6mm thick single sided PCB scraps - this type of enclosure is popularized by the ARRL Handbook for the step attenuator project ("Low power step attenuator," ARRL Handbook 1992, ch. 25). The side panels are soldered together at the edges, while the top and bottom panels are held by 25x3mm machine screws. RF connections are made through the BNC connectors.
The amplifier's output is connected to a Minicircuits ZFBT-6G biastee for inserting the 12V supply. The current drawn is ~21mA. The biastee's influence on the test results is removed by calibration. We did not perform nonlinear tests, such as OIP3 and P1dB, because they are not expected to differ from datasheet's specifications.
Input and output return losses are better than -21dB and -13dB, respectively in the 3dB gain bandwidth (3-780MHz). The gain is greater than 15dB up to 1GHz.
The evaluated prototype is unconditionally stable over 10MHz to 6GHz range. The MAR6/MSA-06 has a 10GHz fT and so, ideally, its stability should be evaluated to to this frequency. However, this was not done due to equipment limitation. Although the manufacturer's data indicates potential instability at the device-level, i.e. k<1 over 1.0-1.5GHz, the prototype satisfies the stability criteria k>1 and B>0 over 10-6000MHz. The reason behind this difference could be different substrate materials; the manufacturer characterizes the device using low loss ceramic PCB (ICM fixture) whereas the actual prototype is constructed with lossy FR4 PCB which may have fortuitously stabilized it. Another possible reason for the better than expected stability is that the 390R bias resistor acts as parallel resistive stabilization .
 K. Wagner, "Bredbåndsforstærker: En MAR, en Monotolithic Amplifier," [Online] Available: http://techdoc.kvindesland.no/radio/audiforradio/bredfandsforsterker.pdf
 J. J. Carr, "Using the Mar-X series of very wideband monolithic microwave integrated circuits (MMIC)," [Online] Available: http://techdoc.kvindesland.no/radio/audiforradio/marex_mmic.pdf
ON6MU, "RE-HFA1MAR6 Wideband VHF/UHF/SHF monolithic PreAmp based on MARx-series" [Online] Available: http://users.belgacom.net/hamradio/schemas/preamp_HF_VHF_UHF_SHF_wideband_MAR6.htm#HF_VHFPreAmp
 Reddit discussion list, "Wideband Preamp - LNA for the rtlsdr. Schematics inside. RTLSDR" [Online] Available:
 Reddit discussion list, "We should get serious about making the direct conversion receiver conversions work great RTLSDR" [Online] Available: http://www.reddit.com/r/RTLSDR/comments/vq5q3/we_should_get_serious_about_making_the_direct/
 L. Besser, "Avoiding RF oscillation," Applied Microwave & RF, Spring 1995.
(c) July 2012. This work is protected by copyright and may not be reproduced in any form without the expressed consent of the owner.