Composed by Naman // Claude
GS. — Observations from 26.85° N, 80.95° E
Reference Three chains

The instruments. Three chains, three targets.

The rig is not one antenna. It’s three operating configurations of the same rooftop — a V-dipole on a pole for the polar pass, a one-metre offset dish aimed at GK-2A through the morning haze, and the same dish repointed south-southwest when Elektro-L5 decides to broadcast. Each configuration is its own instrument. Each instrument writes its own log.

I
Instrument I VHF · 137.900 MHz

137 MHz V-dipole

Target METEOR M2-4 (and M2-3)

A pair of crossed half-wave elements on a rooftop pole, listening for the LRPT downlink as METEOR sweeps overhead. The simplest instrument on the roof — and the one that has produced the only polar-orbit imagery in the register so far.

Diagram The 137 MHz V-dipole chain, antenna to decoder.
Rooftop · Lucknow NAS · /satellite I 01 V-dipole Linear, ~1 m elements Pole, ~3.6 m, RHCP loss ~3 dB II 02 Receiver RTL-SDR Blog V4 cu8 · 1.024 Msps · gain 25 dB III 03 Decoder SatDump v2.0.0-α meteor_m2-x_lrpt · --dc_block=true
Antenna
Crossed V-dipole, linear polarisation
Mount
Rooftop pole, ~3.6 m, no rotator
Frequency
137.900 MHz (LRPT primary)
Sample rate
1.024 Msps (Nyquist for 1024 ksymbol/s QPSK)
Gain
RTL gain 25 dB (no physical LNA in chain)
Decode
SatDump on NAS, --dc_block=true mandatory for cu8

Operative note The V-dipole is linear and LRPT is circular, so there's an unavoidable ~3 dB cross-polarisation loss before anything else fails. The instrument's real failure mode isn't the polarisation though — it's the tuner gain. Pushing the R828D above 30 dB compresses the front end and raises the noise floor by 20+ dB; at 25 dB the rig is the quietest it has ever been measured.

View captures from this instrument

II
Instrument II L · 1692.140 MHz

L-band chain · GK-2A

Target GEO-KOMPSAT-2A Geostationary, 128.2°E

A one-metre offset dish on the rooftop, repurposed from a Ku-band TV setup, with a hand-wound RHCP helix at the focus and a SAWbird+ GOES doing the front-end work. Aimed thirty degrees up and 112° east of true. Decodes a full-disk Earth every ten minutes.

Diagram The L-band chain · GK-2A chain, antenna to composite.
Rooftop · Lucknow NAS · /satellite I 01 Antenna 1 m parabolic az 112.6° · el 29.6° II 02 Feed 3-turn helix, RHCP 1692.14 MHz centre III 03 Front-end SAWbird+ GOES LNA + SAW filter, ext. PSU IV 04 Receiver RTL-SDR Blog V4 cu8 · 2.4 Msps V 05 Decoder SatDump v2.0.0-α pipeline · gk2a_lrit VI 06 Composite Sanchez v1.0.26.1 blue underlay · haze 0.6
Antenna
1 m offset parabolic (ex-Ku TV)
Feed
3-turn RHCP helix, hand-wound
Pointing
az 112.6° true · el 29.6°
Front-end
Nooelec SAWbird+ GOES (LNA + SAW, ext. PSU)
Frequency
1692.140 MHz (LRIT, BPSK 128 ksymb/s)
Sample rate
2.4 Msps cu8
Post
Sanchez · geostationary projection · blue underlay

Operative note The chain is shared between GK-2A and Elektro-L5 — only the pointing changes. SAWbird+ GOES is externally powered; the SDR's bias-tee must remain off. Beamwidth at L-band with the 1 m dish is roughly 12°, so a re-aim for any non-GK-2A target is a deliberate manual operation, not a click.

View captures from this instrument

III
Instrument III L · 1691.000 MHz

L-band chain · Elektro-L5

Target Elektro-L5 Geostationary, 76°E · commissioning

The same one-metre dish and helix-and-SAWbird front-end as Instrument II, repointed south-southwest at the 76°E geostationary slot. Listening for Elektro-L5's sporadic LRIT broadcasts during its commissioning window — currently silent more often than not.

Diagram The L-band chain · Elektro-L5 chain, antenna to composite.
Rooftop · Lucknow NAS · /satellite I 01 Antenna 1 m parabolic az 190.9° · el 58.0° II 02 Feed 3-turn helix, RHCP 1691.000 MHz centre III 03 Front-end SAWbird+ GOES LNA + SAW filter, ext. PSU IV 04 Receiver RTL-SDR Blog V4 cu8 · 2.4 Msps V 05 Decoder SatDump v2.0.0-α pipeline · elektro_lrit VI 06 Composite Sanchez (planned) geostationary · 76°E centred
Antenna
1 m offset parabolic (shared with II)
Feed
3-turn RHCP helix
Pointing
az 190.9° true · el 58.0° (target 76°E equator)
Front-end
Nooelec SAWbird+ GOES (LNA + SAW, ext. PSU)
Frequency
1691.000 MHz (LRIT)
Schedule
Sporadic — slots at :12/:27/:42/:57 UTC, mostly silent

Operative note L5 has been in commissioning since 2026-02-12, with broadcasts clustered around 08:00 UTC and rarely lit outside that window. The first-light attempt on 2026-05-17 produced zero CADU — Viterbi never synced, FFT confirmed no signal in band — leaving the unanswered question of whether the dish was on the bird, the polarisation sense was wrong, or L5 simply wasn't transmitting during that slot.

View captures (commissioning, no images yet)

Standing notes

Across instruments
Switching
Only one chain is wired to the RTL-SDR at a time. The SPDT switch the rig was designed around is not yet installed, so moving between instruments is a manual coax swap and, for II–III, a physical re-aim.
Cadence
GK-2A transmits continuously; capture slots are scheduled on the hour and the half hour. METEOR is captured per-pass when the orbit is favourable. Elektro-L5 captures track its sporadic broadcast cadence, mostly within an 05:30–09:30 UTC window.
Calibration
The 137 MHz chain uses the always-on ORBCOMM constellation as a chain canary — if frames decode, the rig is alive. The L-band chains have no equivalent free signal; GGAK at 1693 MHz is a candidate but unverified on this rig.
Retention
Raw IQ from a successful decode is purged within an hour. Failed-decode IQ is held seven days in /satellite/_raw-iq/needs-review/, then auto-deleted.
Provenance
Every IQ pulled and processed is logged one JSON line at a time in /satellite/.processed-iq-log.ndjson — keyed on filename and first-megabyte SHA-256, then projected to a canonical per-IQ state file the website reads from.