SX1280 2.4GHz LoRa

Hi Guys,

I am looking for help to develop code for SX1280 Nodes and Gateways for TTN. The benefits of 2.4GHz is faster data rater and no fair usage limits on broadcasting but the range seems to be slighly lower but still very acceptable.

I have designed a board along the lines of the TT-GO LoRa boards which has an ESP32, header pins for OLED, GPS and BME280 and is being used on high altitude weather balloons. The boards take an RFM95 868MHz module but I have also had a few more manufactured which will take a NiceRF LoRa128 module which has an SX1280 chip. We are using one of the boards as a gateway with the single channel gateway code from Things4U for testing but in Africa where the balloons are being launched by university students, they have a commercial gateway connected by 4G to TTN

Anyhow, Id be interested in developing an ESP32 breakout board with the LoRa128 module on board and make the Eagle design files and gerbers freely available so people can get their own boards made and maybe offer a few on fully assembled on Tindie at a reasonable cost but what I need is some help in terms of coding.

I have found a SX1280 library written by Stuart Robinson but I dont really know where to start incorporating this into either the node or gateway code and am hoping someone will be interested in helping with the coding. I will send you a pair of the boards with ESP32 and LoRa128 modules onboard in return for your assistance and to enable you to test the code etc. The code needs to be open source and freely available along with the PCB design files and hopefully, this exciting development from Semtech can be put to good use.

Please get in touch if this interests you. Thanks, Steve.

Paging @LoRaTracker

Semtech do produce a reference design for a 2.4Ghz LoRaWAN Gateway;

It would not make a lot of sense to move away from that design.

Whilst the use of 2.4Ghz allows you to legally run at 100% duty cycle from both nodes and the Gateway it would seem very unlikly that any TTN support would fund the movement of potentially very large amounts of data between thier servers for free.

I do sell some basic boards which are easy to use for experimenting with easy to build ‘nodes’ that can take a SX1280 (or SX126x or SX127X) plug in board;


And slightly off topic but if there was a suitable network provider for the backend, 2.4Ghz LoRa would allow for fairly significant ammouts of data to be moved around, I currently have a setup running that will move a 64K file (a JPG image) error free between two Arduinos in around 8.5 seconds.

@LoRa Tracker, whilst I think it is likely that TTN may impose data limits, they did announce support for 2.4GHz at The Things Conference 2021 a few weeks ago.

2.4Ghz isn’t for everyone. A LoRa128 module is about three times the cost of an RFM95 and the range is lower so for many use cases, it is not worth it. There are of course use cases which would benefit from fast data rate without sending very large amounts of data. Our own use case doesn’t need fast data rates or send large packets but technically, it breaks the rules sending a packet every five minutes for a couple of hours, then nothing for a week so its an ideal candidate for 2.4GHz

The use on ships is for a specific reason, 2.4Ghz is a global band, an important attribute for use on ships.

However, in most places in the World, the legal duty cycle limit allows for more than the fair use limits that TTN impose as part of the deal to allow the community to use their servers for free. For 2.4Ghz to be advantageous over UHF in most fixed installations, TTN would have to increase the fair access (air time) limit over UHF considerably, have they said they would ?

For private network use then indeed 2.4Ghz would allow in some circumstances, say large factories, for significant increases in the amount of data that can be moved around, due mainly to the lack of duty cycle limits.

Given the shorter range of 2.4Ghz LoRa at ground level, the use for some form of World wide data collection system for things that move internationally may be limited.

I can see there are applications for 2.4Ghz LoRaWAN in private setups, ships, factories etc, not so sure about general applications in the community …

@CyBaart 2.4Ghz is more a TTI thing than TTN as, as far as I am aware, no TTN implementation and support as yet. TTN has a fair use policy which applies to both the on air utilisation and by extension the impact of the amount node of traffic on the free back end services, as opposed to legal duty cycle limits implied from

The example use case for 2.4Ghz LoRa/LoRaWAN is international shipping and logistics or similar global roaming use cases where a device may end up in any one of a number of different regulatory regimes and associated band plans, and won’t necessarily have knowledge or capability to change band plan and avoid becoming illegal! Indeed in extremis think of an actual ship itself where you might be monitoring not only cargo but also the ships infrastructure - engines, pumps, fuel/water levels, air con, motors, galley’s, crew quarters, etc. Where a device designed for use in say US suddenly becomes illegal when it sails into EU waters and docks in say Rotterdam, or heads to Tokyo Japan, or Souel S.Korea, or Haifa Israel… having LoRaWAN on 2.4 GHz suddenly becomes very attractive! Faster data rates are the cherry on the cake not the reason d’etre as such.


DONT DO IT. Single channel packet forwarders, not Gateways, are not and can never be LoRaWAN compliant/capable and worse still become disruptive to other users on the same LoRaWAN networks. 5+ years later TTN is still battling to correct the mistakes inherent in allowing SCPF’s on the TTN network in the early days, and as with this post still trying to correct potential users impressions and re-educate wrt their use … Start 2.4Ghz ‘clean’ and don’t fall into that trap.

The amount of traffic is not the only consideration for the fair use policy. The policy is also in place to force users to consider playing nice on the shared resource of available airtime. With the legal limits a couple of hundred devices can saturate all channels making LoRaWAN unusable at that location and its surroundings.


Except that in looking up what a “real” 2.4 GHz gateway would cost and its carrying capacity, I just discovered that Semtech themselves are promoting using node radios on the network end, and as far as I can tell don’t even have a reference design for a concentrator-type 2.4 GHz gateway. One would think the SX130x could handle the baseband with a different RF chip; but maybe there are more differences. Or maybe they’re thinking non-LoRaWan schemes more than LoRaWan ones.

Anyway, what is currently available for 2.4 GHz is a box with three single-frequency single-SF node radios for uplink, and one for downlink (which sounds a lot like one of the more notorious not-a-gateway boxes we get questions about here…)

If TTN starts to support a bandplan, the real “mistake not to repeat” would be to not have a way to define the capabilities of a gateway when registering it - especially when starting from a position where nobody’s box is really a proper gateway.

1 Like

One potentially interesting application, if TTN were ever to support 2.4Ghz LoRa, is that the SX1280 devices have the internal firmware to measure distances from ‘node’ to ‘gateway’ by measuring the time of flight of a packet exchange.

Yes, that was used for early demos of LoRa capabilities for potential indoor location/asset tracking or placement. Great fun was had doing customer intro’s/demos where one ST Eiger eval unit (such as later used for the SMTC LoRa EVK’s) with proto location enabled Si was hidden in customers office or factory then another handed to them and told go play ‘hot/hotter, cold/colder’ (using signal strength and ToF) to find the hidden unit. Placing in a higher or floor level position often foxed many as people don’t always look/think in 3 dimensions (think Star Trek - Wrath of Khan, nebula chase scene :wink: )

Glider pilots do it in 3 dimensions - and silently too

And I gather often do it solo too! :wink:


Are you specifically refering to the ranging capabilty of the SX1280 ?

Most all of your post makes no sense to me at all.

Early proto of what was finally to be baked into1280 wrt ‘ranging’ was demoed on early LoRa si proto/demo systems - on a P2P basis each speially configured Eiger was basically a node radio - long before we had GW’s :wink: … not to worry…your coment just triggered some old fond memories :slight_smile: …moving on…

Moving on maybe.

How ranging works with the SX1280s I understand, having written code for it, your posts on the subject I cannot translate at all.

@Jeff-UK didn’t you read the part of my post where I specifically mentioned that we were using commercial gateways, the single channel gateway is ONLY used for testing, making sure the code works.

Even testing usage of a deficient device pretending to be a “gateway” would not be acceptable on TTN; if you want to do that, you must use your own isolated server infrastructure.

However, at present there’s nothing for 2.4 GHz which approaches the capabilities of a conventional gateway.

If/when TTN supports 2.4 GHz, it’s going to be absolutely critical that the support establish minimum standards for what is allowed to be registered as a gateway and/or provide multiple categories of registration.

Even the boxes with 3 node class radios basically preclude variable-SF ADR, since they can’t receive multiple SF’s. Variable power is still an option though…

So everyone else’s devices know not to send electromagnetic radiation in your direction?

The only way to ameliorate this is to run your own stack.

Try looking at Geroskipou, Cyprus on the TTN gateway map. Not another device for miles.

Why don’t you all just stand me against the wall and shoot me, you would obviously rather do that than offer helpful advice. Sarcasm doesn’t help. Should be called TNN, the nasty network. Thanks a bunch.