Summary of "Should I Share More of My Reticulum + Learning LoRa Process?"
TechRapper video summary (Reticulum + first LoRa/RNode implementation)
- Ad-hoc “learning the ropes” project: The speaker shares their personal process for learning how to build networks with Reticulum, focusing today on adding entry-level LoRa devices they just received.
- Reticulum background + prior steps (context):
- They previously built a Reticulum network stack to connect a single device type using UHF radios.
- They later added a second medium: local Wi‑Fi (without internet), enabling phone communication inside the house while using handheld radios out in the field.
- New hardware + goal:
- Today they introduce two inexpensive LoRa devices operating at 915 MHz (US band).
- Goal: compare how LoRa performs vs their earlier UHF/VHF/HF experience, especially for range vs data throughput.
Firmware choices and what they used
The speaker describes three major firmware options in the LoRa community:
- Meshtastic
- Mesh Central
- RNode (for Reticulum)
They install RNode by using the Meshtastic app’s “RNode flasher”:
- Plug the device in via USB
- Identify the device
- Select the product model (they mention LilyGo LoRa32 v2.1 and the 915 MHz variant)
- Download the firmware from the flasher UI
- Flash it locally
Reticulum provisioning steps emphasized
After flashing, the Meshtastic interface guides several setup steps:
- Select device type
- Flash firmware
- Provision EEPROM
- Set identity
- Each device has a unique cryptographic identity needed for the network
Next hurdle: adding the LoRa interface settings inside their setup (details to be covered later).
LoRa configuration focus (range-oriented)
Their target is very long range, willing to sacrifice bandwidth:
- They don’t care about streaming voice/video
- They want simple text messages and robustness
Connection approach (USB first)
They choose to use USB rather than Bluetooth initially:
- Bluetooth pairing with a phone is possible
- Their goal is off-grid compute / plug-and-play with zero configuration
- They mention they bought bare boards, so power is handled by USB for now
Key LoRa settings they use as a baseline
They list parameters under the LoRa settings and note these are starting values (not universal recommendations):
- Frequency: 915 MHz
- Bandwidth: 125 kHz (starting point)
- Spread factor: 9
- Coding rate: 5
- Output power: 17 dBm
Important rule: the radio parameters must match across devices for communication to work.
Standardizing documentation (“field cards”)
They plan to standardize setup and tracking via “field cards,” including:
- Device/radio identities and identifiers such as:
- identity hash
- LXMF address
- QR code (printed on field cards)
- Operational procedures and example workflows
Demo and verification
After connecting/configuring the devices:
- The UI shows a visual indicator (they describe a “waterfall” display and other indicators).
Message + file transfer test
They:
- Send a message
- Attach a 718-byte file
- Observe transmit cycles (green light flashes)
- Confirm peers exchange data
- Note that the received backup can be downloaded after transfer
They explicitly mention they don’t yet know how speed will change with range (range impact is still an open question).
Overall takeaway / learning approach
They frame the project as general problem-solving:
- They describe limited LoRa experience (about 2 hours at the time)
- They rely on:
- research (white papers, tradeoffs like spread factor vs coding rate vs bandwidth)
- extensive notebook-based documentation
- field-standardization via printed cards
They also emphasize the value of civilian-owned, independent, end-to-end encrypted networks that enable anonymous interconnection.
Main speakers and sources
- Speaker: TechRapper / The Tech Prepper (single main presenter)
- Referenced software/firmware sources:
- Meshtastic app (RNode flasher tool)
- RNode (Reticulum-focused LoRa firmware)
- Mentioned as other options: Meshtastic and Mesh Central
Category
Technology
Share this summary
Is the summary off?
If you think the summary is inaccurate, you can reprocess it with the latest model.