Fipy LoRa range

[imasd]

I am testing my Fipy coverage with Lora and I do not have good results. Connecting two fipy, I reach 700 meters without obstacles and 300 meters with obstacles. The configurations are:

• frequency 868000000
• tx_power = 14
• bandwidth = 125 KHz
• sf = 12
• coding_rate = CODING_4_8
  The antennas are 3 dBi.
  I've also tried connecting a fipy to the Multitech Gateway Conduit, but the results are similar or even worse.
  The documentation says that it reaches 40 km then, what am I doing wrong?
  Thank you.

[robert-hh]

@imasd It depends a lot on the antennas. I made a trial run with a Lopy as a gateway, lambda/4 rubber duck antenna indoor, an IC880a gateway, Aurel GP868 antenna outdoor (not optimally placed) and a Fipy as the mote (what word!) with a simple lambda/4 antenna. The best distances achieved were about ~1400m with only a few building in between. If larger buildings were in between the Fipy and the gateway, the distance went down to about 800 m. Setting SF7, BW125, Coding 4_5. But: at some points the signal from the Fipy was picked up by a gateway 6000 m apart. That one sits on a hill with a free line of sight to the places, where it received the FiPy.
I will repeat that with SF12 and a better antenna for the FiPy mote.
The 40km figure may be possible without obstacles in between, optimal antennas and weather conditions. But that's just marketing.
Some of my figures below

[robert-hh]

@imasd I made another walk, or better to say drive, through the neighborhood, this time the mote set to SF12. The mote was set for the test on the top of the car, and I trid not to be in between the mote at my gateways. Figures below. The area is a mix of industrial buildings, shops, and smaller and larger residential houses, small city stuff. In addition of my gateway, the signals were picked up by five other gateways. The notation nnn@dddd means: rssi nnn at a distance of dddd meters. Interestingly, the IC880a never showed rssi levels below -124db, which is the leve at which it picks up noise packets. The Lopy shows levels down to -138 db, which is the LoPy packet noise level. Either the rssi calibration of both devices is different, or the noise figure of the LoPy is that much better. If I should guess, it's the first. Looking at the boards, it could be a combination of both.

[imasd]

@robert-hh Thank you very much for your answer.
In the Lopy-Lopy communication, why do you have communication in Swimming Bath (2360m) but do not have in Fire Brigade (1520m)? Did you have direct vision, without buildings?
I see that the HLK Gateway reaches more than 8 km, how is that possible? Do they have a better gateway, different antennas or what is the difference?
I am working on a project for which I need to have more than 2 km of range indoors, but with the Fipy modules I can not have more than 300 m. Is there a difference in coverage in using a Lopy or a Fipy? Could you give me the reference of your antennas to test if I get better results? Thank you.

[robert-hh]

@imasd It is all related to the position of nearby buildings, reflections, etc. Both the Nanogateway and IC880a a run by me. And yes, at all test there were building in between. And at no point a clear line of sight to the gateway.
I am a little bit sceptic about the HLK node myself. I understand the results for Winepa and HBRS. Their antennas are on the roof of high building, and Winepa sits on the top of a hill. But the place of HLK is low, in a living area. So it may be registered at the wrong place. I just took the position from the TTN map. I could drive to that place & look. With these results, there should be a visible antenna on the roof top.
I have my serious doubt that you will achieve 2 km indoors. At least you should forsee in the middle of the area you want to cover a gateway with a good, high positioned outdoor antenna.

[imasd]

@robert-hh Today we have done tests again with much better results. Putting the Fipy Gateway outside, we get connection up to 2 km in urban environment, discontinuously. We believe that the problem was because of the window glass, which has a sun protection filter, which is why we have read it, it reduces the distance a lot. Thank you very much for your help.

[robert-hh]

@imasd I was noisy an so I drove over to the HLK place. And indeed, at that spot there is a good sensitive antenna on the top of a 5 story building. The antenna looked like the TaoGlas Barracuda, pretty long. And other short wave antennas. So the person seems to know what she/he is doing.