Author Topic: RFM69HW Expected range  (Read 1967 times)

kolaf

  • NewMember
  • *
  • Posts: 30
  • Country: no
RFM69HW Expected range
« on: April 22, 2016, 07:25:14 AM »
I know there are several threads posted on radio range, but many of them seem to be related to configuration issues, and I don't think that is my problem.

I've set up both the node and the Gateway example sketches, configured the correct frequency 868 MHz and defined IS_RFM69HW. I chose to disable the ATC version since I will be moving around testing the range. I use the simple wire monopole 1/4 wave antennas with a length of approximately 86 mm. I have not set the power setting explicitly, but as far as I can see it should be at the maximum 31 by default.

I leave the node on my desk on the first floor in a house constructed using solid timber, with several windows in the room. I plugged the Gateway into my laptop and move around outside to check whether I receive any messages and the associated signal strength.

Inside the house the signal seems quite good, dropping down to around -50 at the other end of the house. Moving outside I get the same signal strength right outside my wall, but as soon as I move a few metres away the signal disappears completely. I then have to move a further 30 m away which more or less brings the line of sight between the node and the gateway through the window as opposed to the timber wall. What are is a site at this point is around -65. I then move into my barn which is around 50 m from the house and made of wood as well. Inside here the connection is intermittent with multiple lost packets and signal strengths down to -78.

Is this the expected behaviour from the HW radio? I seem not to be able to get further away from the house than around 50 to 60 m in most directions, a bit further in one direction. I do not have windows on all sides of the room, which means that in some directions the line of sight assumption from the previous paragraph falls to pieces.

Could this be an issue with the antennas that are quite simplistic, are the wooden walls really blocking the signal that much, have I configured the radio is wrong? Could there be issues with radio interference outside from sources I am unaware of? Is it necessary to do something with the bit rate or other configuration parameters to increase the signal range?

The several reports of the beginning of this forum related to range, but most of them deal with getting a maximum possible range at the cost of reduced  bit rate. How far should I expect to reach without making any changes, and what changes do I need to make to get as far as I need? Also, do have to make these changes on all radios in the network, or can the gateway communicate at a higher data rate with other nodes simultaneously?

As you can see I have a bunch of questions but few answers. Any enlightenment would be very helpful :-)

Thanks.

Felix

  • Administrator
  • Hero Member
  • *****
  • Posts: 6866
  • Country: us
    • LowPowerLab
Re: RFM69HW Expected range
« Reply #1 on: April 22, 2016, 08:49:54 AM »
RF range is not something you can estimate based on some magic formula. What you are getting is not too bad given the default settings the library comes with.
You can look in the Moteino forum for some sticky topics regarding spectacular ranges achieved by fiddling with the settings (bitrates/bandwidth etc).
Besides the major aspects like bitrate, bandwidth, receiver sensitivity, output power and antenna gain, the range depends on so many factors that it's almost ludicrous for anyone to say "such and such board has about this much range". That may be true for a very restricted and well controlled set of conditions but even just rotating the board will start changing the range because of antenna/signal polarization. Usually such a statement has a ton of assumptions and perfect conditions of LOS (line of sight and no obstacles). Not to mention atmospheric conditions, interference, and other factors.

I have heard very many cases where the default settings have given users hundreds of meters in the open. Again, unknown conditions but generally assumed urban LOS with some obstacles.
A RSSI of -80 is still a pretty good signal. When you start approaching -100 that's a rather weak signal approaching the noise floor. The default settings bottom at around -100 so as you get closer you loose more and more packages.

I hope that sheds a little insight but it took me years to understand many of these things I just mentioned. When I started making Moteinos I knew virtually nothing about RF, I just used settings that worked with RFM12B and eventually moved to RFM69 and improved library settings as I started to learn more.

kolaf

  • NewMember
  • *
  • Posts: 30
  • Country: no
Re: RFM69HW Expected range
« Reply #2 on: April 22, 2016, 09:00:06 AM »
Thanks for the input. I found some code in the forum (by you) to reduce the bit rate. I tested this, I was able to get a better signal inside my barn. Well, not a better signal (-85 versus -79), but with the reduced betrayed the messages came through with lower power. I haven't seen anything down to -100, so apparently the drop from  -80 to -100 has a quite severe impact on reception probability.

I tried to locate both antennas vertically so I shouldn't have any polarisation issues.

I will play a bit more and see what I can learn. Guess this post was most motivated by frustration over not knowing what I should expect and whether what I was seeing was normal :-)

Felix

  • Administrator
  • Hero Member
  • *****
  • Posts: 6866
  • Country: us
    • LowPowerLab
Re: RFM69HW Expected range
« Reply #3 on: April 22, 2016, 09:10:00 AM »
If you keep them a few meters away with LOS you should be in the -30 range.
If you move a wall or two away you are already in -40 to 50 which actually feels a dramatic drop, but it's very normal. At the other end of the house it can be -70 or -80 depending how much wall-concrete is in between. I have my mailbox mote across the street about 40m away, the wire monopole sticks out and points towards my side of the street (towards the house) .. I receive about -70..-80 (depending on weather conditions) in my basement which is below ground and has concrete and walls inside too. The output power I think I tuned to level 28-29, otherwise the signal would be even stronger but battery would last significantly less because of the TX spikes. I am perfectly happy with that and I think the RSSI is great at 55K bitrate. This is probably my best example, other motes are inside the house and no dirt or much concrete obstacles, except lots of walls and floors.