Checking my telemetry is valid for sending over LORA

[James Bell]

Hi everyone,

I'm going to try and get my project over the line this summer - I just wanted to check that my telemetry is correct - this is the message I send out over LORA:

$$SHARK1,143,12:12:12,51.46278,-0.00298,63,0,0,03*CE5A

$$SHARK1,146,12:26:16,51.46287,-0.00283,62,0,0,03*E5CD

The 143, 146 etc are just my message ids - they increase sequentially every 5 seconds. They restart from 1 if the pi restarts but I'm not expecting that to occur / be an issue

I can decode it with my own receiver, but I'd rather not solely rely on my own receiver on launch day!

Thanks,
James

[Nick McCloud]

On Monday, 15 June 2026 at 13:28:27 UTC+1 James Bell wrote:

I'm going to try and get my project over the line this summer - I just wanted to check that my telemetry is correct - this is the message I send out over LORA:

What receivers are you expecting people to use as there are many different software on different platforms.

 

$$SHARK1,143,12:12:12,51.46278,-0.00298,63,0,0,03*CE5A

$$SHARK1,146,12:26:16,51.46287,-0.00283,62,0,0,03*E5CD

Looks rather short without the Field String - not sure what the 03 may be. What format are you using and what software are you using to test it with.

 

I can decode it with my own receiver, but I'd rather not solely rely on my own receiver on launch day!

There are very few people that receive LoRa in the general area of London / South-east - to be precise, ONE. Plus a few on the north coast of France & the Netherlands. See the launch data from the a couple of Saturday's back.

Which firmware did you start from, or did you roll your own?

Where are you thinking of launching from? Launching from your location above will put it in contention with a whole bunch of airplanes ....

There's so much more detail that will help us steer you if you care to share. Step back and look at it through the eyes of someone that knows a bit but nothing about what is on your bench.

[James Bell]

Thanks Nick - that's really helpful, apologies for not giving more to work with!

Format and the mystery fields: I'm sending a standard UKHAS-style ASCII sentence:

$$SHARK1,sentence_id,time,lat,lon,altitude,temperature,battery,sats*CRC

 - SHARK1 = payload callsign
 - sentence_id = packet sequence number (increments every 5s; resets to 1 only on a Pi restart, which I don't expect)
 - time = GPS UTC, HH:MM:SS
 - lat / lon = decimal degrees
 - altitude = metres
 - temperature = degrees C (not yet wired in — currently transmits 0)
 - battery = voltage (not yet wired in — currently transmits 0)
 - sats = GPS satellite count
 - CRC = CRC16-CCITT (poly 0x1021, init 0xFFFF, no reflection), computed over everything between $$ and *

So in the example I first sent, the two 0s were those temperature/battery placeholders and the trailing 03 was the satellite count. I'll either populate them properly or drop them before flight.

Firmware: I've rolled my own rather than starting from PITS. Node.js on a Pi Zero 2 W reads the u-blox M8Q over I2C and builds the sentence, piped locally to a small Python service that drives the Uputronics (SX1276) board via adafruit_rfm9x. Happy to share the relevant section if it would help you spot anything.

Radio: the board is the Uputronics 434 MHz LoRa hat, so this is a 434 MHz flight, and need to choose exact frequency later. The plan is Mode 2 (explicit, coding 4:8, bandwidth 62.5 kHz, SF8, LDRO off) - if you think this is not correct please let me know!

Receiving, coverage and launch site: to be clear, the coordinates in those test sentences are just my bench location, not a launch site (I understand the big airspace restrictions around here!). I haven't actually decided on a launch location as yet - it looks like Cambridge area might be a popular location from looking through the community posts? This would be a fair trek but given I've been working on this some time I don't mind making a trip of it. Would appreciate if there are any recommendations for launch sites in the south or south east but sounds like you would recommend against it given mentioning only one station?

Thanks again,
James

[Nick McCloud]

On Tuesday, 16 June 2026 at 15:12:34 UTC+1 James Bell wrote:

Thanks Nick - that's really helpful, apologies for not giving more to work with!

Cool, I'll chug through the details late evening when I get some quiet to concentrate.

It does seem you may be fixated on the UKHAS wiki which as a 'standard' was well extended beyond such a small set of fields - it is pretty ancient tech now - more reading here & Googling may be worth while.

 

Would appreciate if there are any recommendations for launch sites in the south or south east but sounds like you would recommend against it given mentioning only one station?

Two things:

Permission to launch from the government is a legal requirement. For which they will want to know where you want to launch from.

And if there is/was only one LoRa station in London, going somewhere else may reveal more UK listeners, or not, it's hard to say, you'll find out once the balloon is >5km and has good line of sight of the UK. Most of the listeners are on HorusBinary on the north coast of the Netherlands.

 

[Nick McCloud]

On Tuesday, 16 June 2026 at 15:12:34 UTC+1 James Bell wrote:

Format and the mystery fields: I'm sending a standard UKHAS-style ASCII sentence:

As I mentioned above, is a bit truncated to what people send and I'd have to wonder if most current software which would tend towards being legacy would expect a field string so it knows what it is getting. 

So in the example I first sent, the two 0s were those temperature/battery placeholders and the trailing 03 was the satellite count. I'll either populate them properly or drop them before flight.

As above, not sure how the eco-system will cope with this.

 

Firmware: I've rolled my own rather than starting from PITS. Node.js on a Pi Zero 2 W reads the u-blox M8Q over I2C and builds the sentence, piped locally to a small Python service that drives the Uputronics (SX1276) board via adafruit_rfm9x. Happy to share the relevant section if it would help you spot anything.

Each to their own but that's quite a combo you have going on there. I suspect that without any Zig, Rust, Erlang or Haskell in the mix it may not qualify for full-stack status. Real men would do it in BrainFeck, just saying. 😉

Don't forget to turn on flight mode so it can work above 18,000'.

 

Radio: the board is the Uputronics 434 MHz LoRa hat, so this is a 434 MHz flight, and need to choose exact frequency later. The plan is Mode 2 (explicit, coding 4:8, bandwidth 62.5 kHz, SF8, LDRO off) - if you think this is not correct please let me know!

Uputronics board, check, a gold standard. Mode 2 is the most prevalent for telemetry, good. 

 

Receiving, coverage and launch site: to be clear, the coordinates in those test sentences are just my bench location, not a launch site

OK, we've seen a lot over the years, a good percentage generated by my own mis-fires, so we don't assume anything.

 

Would appreciate if there are any recommendations for launch sites in the south or south east

I think I fully thrashed this one out a couple of times over the last few months in terms of finding a launch site, so please read those posts. TL;DR: find the corner of a big park, don't look dangerous.

No where inside the M25 is going to be clear of airspace without requiring permission from ATC to launch and without aircharts & flying experience you won't be talking their language. You can download the KLM's from NATS for airport CTR & ATZs for Google Earth and can import the prediction KLM to see how it may intersect with active flights.

Going South radically limits the amount of space you have for your flight without finding salt water. Look at predicitons every day for a fortnight to see what sort of patterns you get. Today Hever Castle would get wet. Go West but NOT Windsor, like Cheltenham, there's a nice site there with a donut shaped building that would love you to rock up with a big gas cyclinder.

 

but sounds like you would recommend against it given mentioning only one station?

As above, it's not clear what the HAB receiver network looks like in the South. There are peeps that like this sort of things up North that will help out but they will only hear you when you get high. Plus I can run a base station at home as well as mobile. You can do the same, use a decent antenna on a high point.

Plus almost all licensed Radio Amateurs can receive HorusBinary with the kit they have, albeit maybe needing to install HorusGUI. Whereas LoRa needs LoRa, which is a thing for some, but on 868MHz due to Meshtossertic / Meshmeh, not on 434Mhz. Modules are available on Amazon but then it needs an MCU on the side plus something like HABBase.

For receiving using LoRa, look at: https://github.com/RoelKroes/TBTracker-RX

For HorusBinary, look at: https://github.com/projecthorus/chasemapper

Your transmitter & receiver combo needs to be rock solid and at a minium, you need at least one person near by that can wrangle receiving during the flight as well as you being within a couple of km's of the predicted landing area - from the static prediction before launch and then as it is updated by the tracking site.