The BIG and SMALL ANTENNA topic part 2

Hi Remko,
Thanks for your feedback, unfortunately I only have one gateway so I cannot swap it out. I must say the behaviour is odd.

Chris Dann

Hi LoRaTracker,
I am observing the reducing RSSI on the recieve side at my gateway. I have two known devices that have not moved so I am guessing that my problem is in the recieve transmission line or I guess it is possible that as Remko suggested there is an intermittent reciever issue with my gateway.

Chris Dann

Yes, but if there is a loss or other problem with the receiving of the antenna, you would expect it to affect the transmitter also.

Or are you saying your gateway has seperate recive and transmit paths, from the gateway electronics to the antenna(s) ?

Hi Chris,
The process you may have to follow is that of elimination so you can exclude possible causes.

Hi All Great forum. I have yet to find an antenna I really like for a gateway and have tried a number over the last few years. I was keen to see what the Chinese ones are like. Like many of their statement there are often over stated, but I though I give a 10dbi colinear one ago 900mm long and also wanted to see the “vibrator” construction. I ordered this one from AliExpress.

I have tested it and the antennas VSWR was 1.38-1.45 so not great but ok. I range tested the gain with both up and down tilt of 5° and 10° and horizontal using test signal and calibrated antenna and spectrum analyser. The max gain was around 5.6 dBi max.

As it was so much less than the spec, I then decided to remove the radiating element (Just unscrew the base after heat treatment) to expose the antenna. I enclose the pictures below, for those interested. Not a vibrator construction but a modified CoCo antenna. A matching feed element (like a shorted balun dipole and λ/4 matching section), 2 sections of λ/2 and a nominal shorted λ/4.

The antenna was not as expected and explains why lacking gain. Its sad that they don’t make antenna to spec, as all the construction was very good practically. The have a good fibre glass radome, well-sealed, Brass CoCo and feed section.

For those interested the λ/2 sections are 107.2mm and λ/4 39.7mm. It uses solid dielectric (Vf of 0.66) of 9mm diameter and a 10mm o/d and i/d of 9.1mm brass tube. This would give the designed resonate frequency of around 920-930 MHz, but with radome and the very short λ/4 sections give a frequency of 860- 915MHz for less than 2:1 VSWR. Useable but not 10dBi.

IMG_0829 IMG_0830 IMG_0831 IMG_0832 IMG_0834


Hi Remko,
I think water must be getting in somewhere. It might be the gateway or it could be the antenna. I have replaced all of the cables and connectors again and the performace remained poor. We have had 2 days without rain and the weather has been sunny. All of a sudden I see a 20db gain.

Chris Dann

Today we started our long-term testing of the RAK 5,9 dBi beta-antenna on the newest gateway of our community (TTN-Apeldoorn)

First coverage measurement shows expected coverage!

There is a minor issue with TTNMapper not showing the gateway location.