SDR Hardware Setup

Every external SDR experiment in this section shares the same basic signal chain: a feed antenna mounted on the dish, a filtered low-noise amplifier, coax to the SDR, and USB to the computer. This page covers the hardware choices and safety considerations that apply across all of them.

Signal chain

Feed (on dish) → Filtered LNA → Coax → SDR (BladeRF / RTL-SDR) → USB → Computer
      ↑                                        ↑
  Birdcage AZ/EL                         GNU Radio / SDR++ / SatDump

The Birdcage positioner handles pointing. The SDR handles the signal. They connect through separate paths — serial for motor control, USB for RF data.

Critical safety: bypass the LNB bias tee

12-18 VDC on the coax will destroy SDR equipment

The Winegard’s coax path includes a bias tee that injects 12-18 VDC to power the original Ku-band LNB. This voltage is present on the center conductor of the coax going to the dish.

If you connect an SDR or external LNA directly to this coax, the DC voltage will damage or destroy the input stage.

You must either:

1. Use a DC block (inline capacitor) between the coax and your SDR/LNA
2. Disconnect the bias tee at the IDU/power injector (unplug the coax from the Winegard receiver box)
3. Cut the bias tee trace on the PCB (permanent modification)

Option 2 is simplest and reversible. Unplug the coax from the Winegard’s receiver/IDU box and connect it directly to your DC block + LNA chain.

Feed options by frequency

The dish reflector works at any frequency where the surface accuracy is sufficient (roughly wavelengths shorter than the surface error — typically good down to ~500 MHz for a mesh dish, lower for solid). The feed determines what frequency you actually receive.

Frequency	Wavelength	Feed type	Dish gain (est.)	Notes
1090 MHz	27.5 cm	Dipole or patch	~10 dBi	ADS-B — dish gain is modest at this wavelength
1296 MHz	23.1 cm	Helical (5-7 turn) or Yagi	~15 dBi	EME / amateur 23 cm band
1420 MHz	21.1 cm	Helical (5-7 turn) or patch	~15 dBi	Hydrogen line
1525-1559 MHz	~19.5 cm	Helical or patch	~15 dBi	Inmarsat L-band
1575 MHz	19.0 cm	Patch (RHCP)	~12 dBi	GPS L1
1616-1626 MHz	~18.5 cm	Helical	~12 dBi	Iridium downlink
1698-1707 MHz	~17.6 cm	Helical (RHCP)	~18 dBi	NOAA HRPT

Gain estimates assume a 33” × 23” (84 cm × 58 cm) elliptical reflector with ~50% aperture efficiency. Actual gain depends on feed placement, illumination pattern, and surface accuracy. These numbers are starting points — measure your actual pattern with the antenna pattern experiment technique adapted to your feed frequency.

Feed mounting

The Carryout G2’s LNB is mounted on a feed arm that holds the LNB at the reflector’s focal point. For SDR experiments:

1. Remove the stock LNB (it unscrews or unclips from the feed arm bracket)
2. Mount your feed at the same focal point — the position matters more than the exact bracket
3. Secure the coax so it doesn’t snag during AZ/EL moves — route it along the arm and leave a service loop

The focal length of the Carryout G2 reflector hasn’t been precisely measured. Start with the feed at the same distance as the stock LNB and adjust for peak signal on a known source.

LNA selection

A low-noise amplifier at the feed is essential — the SDR’s internal noise figure is typically 3-6 dB, which wastes most of the dish’s advantage. An LNA at the feed drops the system noise temperature dramatically.

Parameter	Recommended	Why
Noise figure	< 1.0 dB	Lower is better — directly sets system sensitivity
Gain	20-30 dB	Enough to overcome coax loss without saturating SDR
Frequency range	Match your experiment	Narrowband filtered LNAs reject out-of-band interference
Filtering	SAW or cavity pre-filter	Critical near cell towers (1700-2100 MHz)
Powering	Bias tee from SDR or separate	Many LNAs accept 3-5 V via coax bias tee

Recommended LNAs by experiment:

Experiment	LNA option	Notes
Hydrogen line	Nooelec SAWbird H1	1420 MHz filtered, 0.7 dB NF
NOAA HRPT	Nooelec SAWbird NOAA	1698 MHz filtered
GPS/GNSS	Nooelec SAWbird GNSS	1575 MHz filtered
Iridium	Nooelec SAWbird Iridium	1626 MHz filtered
EME / ADS-B / Inmarsat	Wideband LNA + bandpass filter	No single-frequency SAWbird available

Powering the LNA

Most SAWbird-style LNAs draw 30-60 mA at 3.3-5V through the coax center conductor (bias tee). Options:

• SDR bias tee — BladeRF and some RTL-SDR dongles have a software-controlled bias tee output
• External bias tee injector — separate powered bias tee between coax and SDR
• Direct power — some LNAs have a separate DC input jack

Don't confuse LNA bias tee with LNB bias tee

The LNA bias tee is 3-5 VDC at milliamps. The Winegard’s LNB bias tee is 12-18 VDC at hundreds of milliamps. They are completely different circuits. The LNB bias tee must be disconnected (see above). The LNA bias tee is what powers your external amplifier.

SDR hardware

BladeRF

The BladeRF 2.0 micro (xA4 or xA9) is a good fit for these experiments:

Spec	BladeRF 2.0 micro xA4
Frequency range	47 MHz - 6 GHz
Bandwidth	Up to 56 MHz
ADC/DAC	12-bit
Interface	USB 3.0
FPGA	Altera Cyclone V
Bias tee	Software-controlled, 4.5 V
Price	~$480

The xA9 variant adds a larger FPGA for more simultaneous channels but isn’t necessary for single-feed experiments.

RTL-SDR

For receive-only experiments (all of these), an RTL-SDR Blog V4 works and costs ~$30:

Spec	RTL-SDR Blog V4
Frequency range	24 MHz - 1.766 GHz
Bandwidth	Up to 3.2 MHz (2.56 MHz stable)
ADC	8-bit
Interface	USB 2.0
Bias tee	Software-controlled, 4.5 V
Price	~$30

The 8-bit ADC limits dynamic range compared to the BladeRF, but for narrowband signals (hydrogen line, HRPT, GPS) it’s perfectly adequate. The 1.766 GHz upper limit covers all L-band experiments listed here.

RTL-SDR frequency limit

The RTL-SDR V4 maxes out at 1.766 GHz. This covers NOAA HRPT (1698-1707 MHz), Iridium (1616-1626 MHz), GPS L1 (1575 MHz), and all lower frequencies. For Ku-band experiments, you need the BladeRF or a dedicated downconverter.

Software stack

Software	Purpose	Experiments
GNU Radio	Signal processing flowgraphs	All (general purpose)
SDR++	Spectrum visualization and recording	All (quick-look)
SatDump	Weather satellite decoding	NOAA HRPT
gnss-sdr	GPS/GNSS signal processing	GPS/GNSS
gr-iridium	Iridium burst detection	Iridium
dump1090	ADS-B decoding	ADS-B
Jaero	Inmarsat AERO decoding	Inmarsat

Installation (Arch Linux)

# Core SDR tools
sudo pacman -S gnuradio soapysdr soapy-sdr-module-bladerf

# SDR++ (AUR)
yay -S sdrpp-git

# SatDump
yay -S satdump-git

# GNU Radio companion (GUI)
sudo pacman -S gnuradio-companion

Tracking modes

Different experiments need different tracking strategies. The Birdcage positioner supports all of these through its rotctld interface or direct motor commands.

Tracking mode	Description	Experiments
Fixed pointing	Point at a known AZ/EL and hold	Inmarsat, geostationary targets
Step-and-integrate	Move to position, dwell for N seconds, move to next	Hydrogen line, ADS-B survey
Continuous LEO pass	Track moving satellite in real-time via Gpredict/rotctld	NOAA HRPT, Iridium
Lunar rate	Track the Moon’s ~0.5°/min drift	EME monitoring
Drift scan	Hold EL fixed, let sky drift through beam	Solar monitoring (alternative)

For LEO tracking, Gpredict drives the positioner through the rotctld TCP interface at 127.0.0.1:4533. For step-and-integrate, script the motor commands directly through the Birdcage CLI or use the TUI’s preset system to define a grid of positions.

Recording while tracking

For LEO pass experiments (HRPT, Iridium), start the SDR recording before the pass begins and stop after it ends. The tracking doesn’t need to be synchronized with the recording — the SDR captures continuously while the positioner follows the satellite. Decode the recording offline.

Next steps

Pick an experiment and follow its dedicated guide. Each page lists the specific feed, LNA, SDR settings, and software configuration needed.

Hydrogen Line The most accessible radio astronomy experiment — detect galactic hydrogen at 1420 MHz.

NOAA HRPT Weather satellite imagery from LEO passes at 1.7 GHz.

EME Monitoring Listen for moonbounce signals on the 23 cm amateur band.