> Simple home-made driver for the SX1276 and SX1262 LoRa chip
Beware of what nailed the Meshtastic people: These chips don't have temperature dependent crystal oscillators. Transmitting more than a few milliseconds causing a temperature rise, throwing the clock off, causing transmission warpage, causing timing errors, causing transmission failures.
Temperature Compensated Crystal Oscillators (TCXOs) is what they should be looking for. And to be clear you can get SX1262 variants with such, eg: https://wiki.seeedstudio.com/wio_sx1262/
> In the case of an SX1262 operating at +22 dBm in the US 902 – 928 MHz band, the frequency drift measured during the maximum LoRAWAN™ packet duration stays below the maximum limit, provided thermal insulation is implemented around the crystal during PCB design.
> At extreme temperatures (below -20 °C and above 70 °C), it is recommended to use a TCXO.
> For any other frequency bands corresponding to longer RF packet transmissions at +22 dBm, it is recommended to use a TCXO.
Chat/text over LoRa always sounded intriguing! I'm merely a 1000km away from Sicily (which antirez say they wanna cover with the network at first), I wonder if it would be possible to do what AM radio does, bounce against the ionosphere and get huge ranges for transmission? Maybe it would need way too much transmission power? Maybe hopping all the way P2P style makes more sense, although surely would be slower.
I can't wait for us to somehow figure out how to connect all the various mesh-networks in the world. MeshCore is a whole continent and ocean(s) away from say Freifunk or Guifi, but would be incredible if we could figure out how to interconnect between them, so we could all talk via the already built alt-infra. Maybe the FossaSat mentioned in another comment (https://news.ycombinator.com/item?id=45806794) could finally be one option.
Can you recommend any outdoor node hardware that can provide services for both MeshCore and Meshtastic? Cost is not an issue, optimizing for “plug in and does All The Things.”
You'd need two radios, one antenna stacked on top of each other. Z offset of a minimum height of 1m/3'. In the X,Y offset of 0,0. You want high gain 8-10db antennas in this arrangement; you'd need more space between if you use lower gain antennas.
If you're looking for parts the Diamond BC920 is the best antenna. Station G2 with a $100 3-4mhz cavity filter for each frequency as well.
Shame, but at least now I understand why, thanks :)
> you’re realistically limited to near line of sight paths
Wonder exactly how far one would be able to get up to reach 1,000km line of sight, assuming you put both points equally far up. Guessing it's the curvature of the earth that gets in the way at that distance?
Unlikely since the most ISM bands are UHF (EU is 868 MHz) and skywave propagation (what you’re referring to) is characteristic of HF radiation (3-30MHz). VHF/UHF radiation passes through the ionosphere instead of refracting.
I tried doing something like this on an ESP32 last year, I wanted to understand how to do an actual mesh with devices that can't transmit and receive at the same time, but I didn't find documentation.
I managed to get 2 devices communicating across town at least ! (After having antenna issues)
I think there’s some real value in WANs that are independent of both the internet and existing telco infrastructure but I don’t think I would put a specific transport at the centre of its design. I think you would be much better off with something like bundle protocol v7 and delay tolerant networking as the actual centrepiece and have something like LoRa as one of many transport link options.
Maybe we are talking about a different threat model here but LoRa as the centrepiece falls apart really quickly the moment someone somewhere decides they don’t want your WAN to exist anymore.
> Simple home-made driver for the SX1276 and SX1262 LoRa chip
Beware of what nailed the Meshtastic people: These chips don't have temperature dependent crystal oscillators. Transmitting more than a few milliseconds causing a temperature rise, throwing the clock off, causing transmission warpage, causing timing errors, causing transmission failures.
Temperature Compensated Crystal Oscillators (TCXOs) is what they should be looking for. And to be clear you can get SX1262 variants with such, eg: https://wiki.seeedstudio.com/wio_sx1262/
For the detailed run down, see https://cdn.sparkfun.com/assets/f/f/b/4/2/SX1262_AN-Recommen... page 14
> In the case of an SX1262 operating at +22 dBm in the US 902 – 928 MHz band, the frequency drift measured during the maximum LoRAWAN™ packet duration stays below the maximum limit, provided thermal insulation is implemented around the crystal during PCB design.
> At extreme temperatures (below -20 °C and above 70 °C), it is recommended to use a TCXO.
> For any other frequency bands corresponding to longer RF packet transmissions at +22 dBm, it is recommended to use a TCXO.
Theory and reality are different here.
As used in the meshtastic devices this chip does actually fail doing normal Lora transmission under reasonable conditions.
I know because I've seen the exact failure.
You've seen the failures in variants with a TCXO?
You mean they don't have temperature compensated? What you described is temperature dependent
too late to edit now :)
This is a radio module issue, not a chip issue.
Cheap modules have cheap crystals, better ones have a TCXO.
The chip itself supports using a TCXO instead of a regular crystal
Chat/text over LoRa always sounded intriguing! I'm merely a 1000km away from Sicily (which antirez say they wanna cover with the network at first), I wonder if it would be possible to do what AM radio does, bounce against the ionosphere and get huge ranges for transmission? Maybe it would need way too much transmission power? Maybe hopping all the way P2P style makes more sense, although surely would be slower.
Checkout MeshCore; we're doing 400 miles using 12 hops going from north Vancouver to Eugene Oregon https://analyzer.letsme.sh/map?lat=47.36113&long=-122.20419&...
> 400 miles using 12 hops
What's the E2E latency on that, for curiosities sake :)
6 seconds or so
Ah, that's not too bad honestly, nicely done.
I can't wait for us to somehow figure out how to connect all the various mesh-networks in the world. MeshCore is a whole continent and ocean(s) away from say Freifunk or Guifi, but would be incredible if we could figure out how to interconnect between them, so we could all talk via the already built alt-infra. Maybe the FossaSat mentioned in another comment (https://news.ycombinator.com/item?id=45806794) could finally be one option.
Can you recommend any outdoor node hardware that can provide services for both MeshCore and Meshtastic? Cost is not an issue, optimizing for “plug in and does All The Things.”
You'd need two radios, one antenna stacked on top of each other. Z offset of a minimum height of 1m/3'. In the X,Y offset of 0,0. You want high gain 8-10db antennas in this arrangement; you'd need more space between if you use lower gain antennas.
If you're looking for parts the Diamond BC920 is the best antenna. Station G2 with a $100 3-4mhz cavity filter for each frequency as well.
Not at 900MHz. Skywave propagation mode is only available under about 30MHz. Above that, the signal reliably continues on into space.
At frequencies this high, you’re realistically limited to near line of sight paths.
> Not at 900MHz
Shame, but at least now I understand why, thanks :)
> you’re realistically limited to near line of sight paths
Wonder exactly how far one would be able to get up to reach 1,000km line of sight, assuming you put both points equally far up. Guessing it's the curvature of the earth that gets in the way at that distance?
Yep. There are cubesats running LoRa payloads in low earth orbit[0], and between mountaintops, terrestrial paths of hundreds of km are possible[1].
[0] https://www.hackster.io/news/fossasat-1-an-open-source-satel...
[1] https://meshtastic.org/docs/overview/range-tests/
> Using metal tape measures to build simple quarter-wave monopole antennas is a hack to that’s actually pretty common in the small satellite community
This is just too crazy, thanks a lot for sharing these links! :)
Unlikely since the most ISM bands are UHF (EU is 868 MHz) and skywave propagation (what you’re referring to) is characteristic of HF radiation (3-30MHz). VHF/UHF radiation passes through the ionosphere instead of refracting.
I tried doing something like this on an ESP32 last year, I wanted to understand how to do an actual mesh with devices that can't transmit and receive at the same time, but I didn't find documentation.
I managed to get 2 devices communicating across town at least ! (After having antenna issues)
I think there’s some real value in WANs that are independent of both the internet and existing telco infrastructure but I don’t think I would put a specific transport at the centre of its design. I think you would be much better off with something like bundle protocol v7 and delay tolerant networking as the actual centrepiece and have something like LoRa as one of many transport link options.
Maybe we are talking about a different threat model here but LoRa as the centrepiece falls apart really quickly the moment someone somewhere decides they don’t want your WAN to exist anymore.
Previously [2023]: https://news.ycombinator.com/item?id=35882967