I've made few tests to compare 433 MHz antenna performance.
Test setup:
- RFM23BP module hooked to atmega238 (2x) (those are found in OpenLRSng transmitters)
- RF22 library
- OOK modulation / 2.4Kbps / 335KHz RX bandwidth
Uponsuccessfully received message RX DBi was red.
Receiver was only listening (uni-directional communication).
Stock antennas
Hand-built antennas
I've started with a LAB test. Antennas 2 meters apart, TX power 1DBm (1.26mW).
two small spring antennas two stock whip antennas two omni antennas two dipole antennas two omni with spring antennas omni vs v_shape omni vs whip with gnd_plane omni vs stock whip omni vs custom spring MAX / DB -12 -30 -12.5 -4.5 -10.5 -6.5 -3 -12 -4.5 AVERAGE / DB -24.5 -40.0 -21.6 -12.2 -18.6 -16.5 -10.6 -22.2 -10.1
*Average reading was achieved by trying to get max RSSI. Antenna was rotated in all directions possible. Consider it as real-life indication of how would such antenna perform on the flying aircraft.
CONCLUSIONS:
Omni antenna (TX) and Whip ground-plane antenna(RX) is the best combo.
Two Dipoles perform nearly identical to Omni (TX) and Custom Spring (RX).
V-Shape antenna is also a good candidate, however, average reading is the worst from the 4 top performers.
Stock antennas are good for nothing.
Then, the FIELD tests. Antennas 300 meters apart, 1 meter above ground. TX power 17DBm (50mW).
omni vs omni omni vs stock whip omni vs stock spring omni vs custom spring omni vs wire with spring omni vs dipole omni vs wire with gnd plane omni vs v_shape dipole vs dipole dipole vs v_shape dipole vs wire with gnd_plane dipole vs custom spring v_shape vs wire with gnd plane MAX / DB -68 -70.5 -72.5 -69 -70 -73.5 -67 -66.5 -67 -62.5 -64.5 -66.5 -77 AVERAGE / DB -71.04 -72.55 -74.94 -72.10 -75.56 -77.66 -73.54 -72.59 -75.15 -68.35 -72.04 -71.47 -108.42
CONCLUSIONS:
dipole(TX) and V-shape(RX) is best combo
dipole(TX) and wire with ground-plane(RX) is 2nd best combo
omni(TX) and V-shape(RX) or custom-spring(RX) is 3rd best combo
reverse combinations are bad performers (i.e. placing V-Shape on TX side)
Real ground-air performance should be few DBs higher, because my antennas were tested relatively close to the ground.
stock vs stock combinations were not tested due to conclusions made in LAB tests.