yep… guilty Coos 
at SF 7 it’s very good I think
Too late to update my post 433 (on how to change a 915 Mhz antenna to 868 Mhz)
So posting here.
I thought I’d check the stock antenna provided with the TTN gateway - sure enough it was set for 915 Mhz.
Not only that it looks almost identical to the cheap Chinese one I bought from eby.
So I swapped with one of my doctored antenna (as seen in previous post) and sure enough the RSSI improved for the gateway !
Be careful though, the antenna is fragile and easily broken - I would buy another antenna and doctor that (and keep the supplied TTN antenna spare). Going to buy a few more of these cheap antenna of Eby now !
1 Like
Hi CurlyWurly,
I did what you did (10 posts up from here) with a cheap antenna. Mine was even worse than yours, it originally had a VSWR of 12 and an impedance of 150+j250 ohms on 868MHz. It was tuned for 975 to about 1200 MHz and would have probably performed well on the 1090MHz ADS-B airplane band
Not only enlarged the center conductor but I also enlarged the brass tube at the bottom by soldering a flap of thin copper foil around the tube, making it about 15mm longer. Because of the copper foil it was not easy to slide the plastic top back completely therefor I slided it back partially and did a piece of heat shrink tube around the base. So the result was about a 1cm longer antenna
After the modification it behaved very well on 868MHz. Putting it on my RAK831 gateway it got at least 5dB better RSSI signals compared to the short stock antenna that was delivered with the RAK.
1 Like
With my antenna, the intial VWSR was 2.72 - You’ve done well in getting it work from 12 !!
Any chance of uploading a pic and showing what you did - also showing the VWSR result?
I’ve just ordered some more cheap antenna to try.
Unfortunately I couldn’t get the same ones again - It might be that I have to do what you are doing?
So, which one you advise would be the best read? 
@ CurlyWurly
Soon I expect another cheap 868 antenna which will very likely not work on 868 because it is for 915 or higher frequency. When I remake that antenna to 868MHz I will not forget to make some detail pictures and give some dimensions so others can do the trick too. The one I modified is sealed now,
@ bluejedi
I own several of those pictured books. The ARRL Antenna Handbook is a good book, it covers many antennas from about 100kHz upto about a GHz. Not dry theory but lots of diagrams pictures and practical designs. But mostly for shortwave, not much for VHF/UHF.
There exists not yet a LoRa Antenna Handbook, that would probably be a big hit
1 Like
There is no real need for a ‘LoRa’ antenna book, there is nothing about LoRa that needs different antennas to other UHF systems.
An excellent reference on VHF\UHF systems, and there are sections on propogation and antennas, is the 'VHF\UHF Manual by G.R.Jessop;
https://archive.org/details/Vhf-uhfManual
I have a printed copy that I bought in 1983, its well thumbed …
3 Likes
Can we do moonbouce communication with LoRa?
(illustration copied from https://www.electronics-notes.com/articles/ham_radio/amateur-propagation/moonbounce-propagation-eme.php)
Moonbouncing or EME (Earth-Moon-Earth) is about the highest level of communication sophistication for ham radio amateurs. Can it be done with LoRa? We certainly need BIG antennas.
Here’s an unconstrained exercise to understand what it takes to send LoRa signals to the moon, and back.
The best possible Lora receiver sensitivity is -136 dBm at SF12 and 125 kHz.
Assuming 1000 watt (!!) transmitter power, and using surplus 10 feet diameter c-band satellite antenna dishes for transmit and receive, the link budget could be as follows:
TX 1 kW: 60 dBm
TX Antenna gain 27 dB (10 foot dish)
TX antenna cable loss: -3 dB
Estimated path loss earth-moon-earth at 915 MHz: -260 dB
RX Antenna gain 27 dB (10 foot dish)
RX low noise amplifier: + 24 dB
RX antenna cable loss: -3 dB
Signal at receiver: -128 dBm, this could work…
… they allready do it with laser … there seems to be a reflector on the moon placed by a mission in the last century
Its used to measure the gradual drift of the moon away from the earth…eventually there will be no full eclipses as disk with appear too small in front of the sun! 
Believe it’s some sort of ‘quadrant’ reflector (not sure correct name) - think one corner of inside of a cube with precisely aligned mirrored surfaces that effectively reflect incident light back along the ‘same’ path to sender…pity it doesn’t work for LoRa e.m. wavelengths! 
Forgot to say I think it was Armstrong and co + others - Apollo’s 11, 14 & 15 that left them - think other none manned placements have also happened with instrument packs…
With your numbers guess no need to fry your neighbours nodes or gateways front ends with 1kW…dial back and use available margin - more like 100W?!
Thats the theory in a lab test, but not the practice in the real world.
A LoRa receiver in the real world, outside of Laboratory tests which are probably done in a Faraday shield, will typically see -105dBm of noise. LoRa can operate at 20dB below that noise level so the maximum realised sensitivity will be -125dBm and not -136dBm.
@LoRaTracker Is this noise floor with an omnidirectional antenna? Would expect lower noise when the antenna looks upward to the sky?
And here again an assumption, that does not bear out in practice.
I have tested the affect of a top notch RX LNA on LoRa reception and this LNA cost me around £400. It added 6dBm of useful LoRa signal gain. So if you know of a ‘RX low noise amplifier’ that will provide 24dBm of signal gain, do please tell us which one it is.
There are ‘RX low noise amplifiers’ that will provide high levels of gain, but that is only realistic on FSK type systems that operate at +5dB to +10dB above noise level, LoRa is different in that the signals are some 30dBm lower than what a typical FSK receiver will operate at.
I would not.
Maybe if you were listening up past Pluto it would be different, not tried that, yet.