Jeroen's Project Journals

Introduction

2026-01-19

Some time ago I became aware of Meshtastic, which is a LoRa-based mesh network for text-messaging. I had looked into LoRa and LoRaWan for work a few years ago, so I knew a little, and this seemed like a cool application. This was also at a time when I became interested in preparedness for my family. I’m not necessarily a doom-and-gloom kind of person, and I really don’t want to call myself a “prepper”, but we do seem to line in a timeline where it is not inconceivable that large quantities of fecal matter will start colliding with rapidly spinning ventilation devices in a serious way at some point. So I started thinking about what I would want to have in place should something serious happen. One thing that is important to have in an emergency is a means of communication. We take it for granted that we can whip out a device and send someone a message that gets delivered almost instantly, but the infrastructure required to make that work is large, complex and vulnerable. My house communicates with the rest of the world through 2 channels: a gigabit fiber internet connection, and the cellular network. If both of these channels stop working for some reason then I no longer have internet, TV, radio or phone service anymore.

If this happens, there are 2 things I would like to have/do:

  1. Have a means of mobile communication so someone on the move in the neighbourhood can stay in touch with home.
  2. Check up on my parents, who live on the other side of our (small) town, about 2 km away.

For 1), I got a pair of Baofeng UV21R radio’s. These are cheap, well-built, and in my (admittedly limited) testing have excellent range. But they don’t make it to my parents’ place. For 2), the plan is to setup a LoRa-based link. I initially tried Meshtastic, but I was disappointed by the poor delivery reliability. Sometimes a message would get delivered, sometimes not. Arguably worse was that the delivery confirmation was also unreliable. Sometimes a message would get delivered, but I would get a “failed” on my end, or vice-verse. I suspected all this was due to the “all clients are repeaters” approach used by Meshtastic.

That was when I came across MeshCore, which was at that time in very early development. MC works with dedicated repeaters, which is an approach that I like better. So I am now in the process of building a couple of “solar nodes”: Solar-battery powered repeaters installed in a fixed location that provide mesh-communication.

Ideally I would install one repeater on my house and one on my parents’ house, and these 2 would be able to communicate directly. This can work, if the nodes have an almost unobstructed line-of-sight. When I experimented with Meshtastic I placed a node on my roof and it would receive chirps from a node almost 12 km away, which was placed on top of a 60m high building. Unfortunately, I strongly suspect this will not work between my house and my parents, if for no other reason than the node I could hear was transmitting at a highly illegal 5W of power.