Link budget and mesh topology issues

Support
1

I’m planning and implementing a meshtastic network as pictured here.
Base map is omitted due to privacy reasons.

linkplan
linkplan983×1199 74.1 KB

Numbers: kilometers between nodes
Red marker: router node, fully static location (9 pcs)
Green marker: end user node, mostly static location (8 pcs)
Blue lines: LOS paths between router nodes
Green lines: user radio to router node paths
Orange line : longest link path, about 70 km in total

So far I have tested upto 5 km links and based on numbers, I have lots of link budget to spare and everything should be fine for the longer links.

Questions:

  • The longest path takes 7 hops from end to end, is this unreasonable?
  • I’m assuming that whatever paths the nodes manage to see, are used for shortest routing, is this so?
  • Does the location and distance between nodes affect routing preference somehow?
  • Are RSSI and SNR any kind of metric in mesh routing decisions?
  • Is this network feasible as a whole?
  • Am I abusing the system here?

Thanks for any input.

2
  • I would make sure only the edge nodes set their hop limit to 7. For the ones in the middle this is unnecessary. Note that with 7 hops, 9 nodes are in the route since the origin and destination node do not count (see LoRa Configuration | Meshtastic).
  • Yes, the algorithm tries to use the least amount of hops by letting the node with worst SNR rebroadcast (see Mesh Broadcast Algorithm | Meshtastic).
  • Location/distance between nodes determine the RF quality (SNR), so yes.
  • Yes, see above.
  • With the correct settings (hop limit as above) and limited traffic (turn down telemetry, position updates, etc. if you don’t need them) and if the LOS paths are really LOS (you probably need some elevation for 23 km) and you have good antennas, probably yes.
  • No, I don’t think so. Would be a nice set-up if it works out :slight_smile: