4 ms·
If you're willing to experiment, you'll find out that you can tune pretty much anything from a random spool of wire to an iron frame bed, and in this case custa
by anilakar 8y ago
If you're willing to experiment, you'll find out that you can tune pretty much anything from a random spool of wire to an iron frame bed, and in this case custard. I've found that an inductively coupled sausage is a nice match on the VHF air band with a proper number of turns.
However, getting a nice 1:1 voltage standing wave ratio only means that the transmitter sees a proper 50 Ω load that does not reflect anything back. Most of the radiation probably happens in the feedline and the rest is absorbed as resistive losses. The whole setup only works because at lower "high" frequencies signal levels are relatively strong and receiver sensitivity is rarely a problem. Most HF transceivers actually have front panel knobs for attenuation and gain reduction.
Also, FT8 is a relatively new digital modulation that works at low SNR and is not feasible for transmitting any information apart from the callsigns of the communicating parties.
- philpem 8y agoExactly this. A dummy load (a 50-ohm resistor) will give you a perfect 1:1 SWR, but makes a rubbish antenna... an SWR match just means it's less likely to fry the transceiver's power stage. You could transmit data over FT8, but you're right -- with identification requirements you'd be limited to maybe half a dozen characters per 15-second cycle. Slow as molasses but it'll be receivable even if the signal is down in the noise floor.
- nominatronic 8y ago> Slow as molasses Slow as custard, surely.