SoftRF version mb202

[Moshe Braner]

This has been in a development branch for a while.  Now merged into "master".  Versions mb200+ switched from the old radio-chip libraries to using RadioLib.  This allows use of the new LR1110 radio chip.

Now my version of SoftRF also supports the Seeed Sensecap T1000-E and the Elecrow Thinknode M3.  The T1000-E can record flight logs, the M3 cannot since it does not have SPI flash storage.  Thus the T1000-E is "better" - but be aware that on a new T1000-E you must downgrade the bootlogger before installing SoftRF.  See the documentation.  If you do want to use the M3, be sure to get the Meshtastic version, the LoraWAN version will not work!

Theoretically this new version also supports the T-Echo Plus and the Thinknode M1, but I don't have them on hand to test.  Reportedly there are still some issues on the M1, involving the buzzer and LEDs.

And of course this version of SoftRF still supports the T-Beam (not the T-Beam Supreme!) and the T-Echo, although there are few (intentional) changes in functionality from version mb179/mb180.  Please report any issues.

The RSSIs reported by RadioLib differ from those reported by the old libraries.  It is best to delete the range.txt file and start collecting new range data.

Version mb200+ also abandoned the old PilotAware radio protocol in favor of the new ADS-L-based PilotAware radio protocol.  Commercial PilotAware devices still automatically accomodate old-protocol devices, but this version of SoftRF does not.

[Alan Hall]

I did a hack of 1.80 to receive both old and new Pilotaware (transmit on the old protocol only as this will be received on old or new units). I have been using it to observe passing aircaft at my home and the vast majority are still on the old protocol so I assume they are not yet upgraded. Pilotaware are, as usual, silent on their plans. For that reason I would not want to go to Pilotaware ADS-L only at this stage.

Alan

[Moshe Braner]

OK. I thought they are currently transitioning, so not worth implementing "dynamic" protocol selection.  I could easily add reception of the old protocol, since the outer packet structure is the same.  Is that worth doing?  Or, can use my version 1.80 for a while yet - or your hacked version.

[Alan Hall]

Well they are certainly transitioning but somewhat secretly and with no published timescale! I'd wait a bit and see what PAW do, they will have to make an update announcement eventually. I would not bother with full dual-mode functionality at this stage and ideally I'd probably revert 2-02 to the old protocol for the time being, but your suggestion of sticking with 1-80 is a sensible compromise. The main thing is people should realise that adopting 2-02 right now for no other good reason will significantly reduce mutual visibility. 

[FrankT]

https://www.pilotaware.com/post/pilotaware-adopts-adsl-iconspicuity

Op maandag 18 mei 2026 om 19:50:37 UTC+2 schreef Alan Hall:

[Moshe Braner]

Working on MB203, which will restore reception (but not transmission) of the old PilotAware protocol.  But the main change is "messages", in both FANET and "Latest" protocol.  Transmits pilot name etc, and outputs $PFLAM sentences when messages arrive.  Also found that MB202 fails to work in FANET protocol, fixed in the upcoming MB203.  If anybody wants to "beta test" MB203 let me know.

My T-Echo died, and so did the Thinknode M3.  Why would the bootloader get corrupted, if that's what it is?  If anybody can test on either of those (or the M1) let me know.  I now only have T-Beams and the T1000-E.

On Monday, May 18, 2026 at 10:52:58 AM UTC-4 Moshe Braner wrote:

This has been in a development branch for a while.  Now merged into "master".  Versions mb200+ switched from the old radio-chip libraries to using RadioLib.  This allows use of the new LR1110 radio chip.

Now my version of SoftRF also supports the Seeed Sensecap T1000-E and the Elecrow Thinknode M3.  The T1000-E can record flight logs, the M3 cannot since it does not have SPI flash storage.  Thus the T1000-E is "better" - but be aware that on a new T1000-E you must downgrade the bootlogger before installing SoftRF.  See the documentation.  If you do want to use the M3, be sure to get the Meshtastic version, the LoraWAN version will not work!

...

[Rens Peeters]

Hi Moshe,

I have received my T-echo Plus and was testing it yesterday with different software versions. Also I learned a lot from your guides. Thank you!

With your firmware MB180 and 202, most seem to work. But I can not test the main functionality well, because I only own just this FLARM device. While using it I encountered some issues and remarks.

• MB180 and MB202; Responsiveness buttons Bad. Touch and mode button must be pressed multiple times before it starts working. Ones it reacts its behaves better following presses. Found out that the press must be very short then to work. Especially the touch button functions are difficult to get going. Like changing the radar range and setting settings on the t-echo it self. Linar's mainstream version of softRF does not have this issue.
• MB202 only; Switch off the device is not possible. Long press mode button, "power off screen saver" screen shows. But device reboots after that, to be live again.
• MB180 and MB202; Screen switches off or turns black after a while (screen saver?) This is not needed/wanted for eink. antighost setting is set to 1=auto. according to your "under the hood" document this should be only happening when this setting is off, right?
• MB180 and MB202; Settings status screen is the Protocol setting status wrong? P:T+A A:LAT for when I set Protocol to Latest and Alternative protocol to eg. ADS-L. Or maybe you can explain the different abbreviations in jour guide, so I and others can understand.
• MB180 and MB202; IMU not found at startup test. In mainstream version ok. But maybe not used as also there no g-load screen?
• MB180 and MB202; altitude/pressure/temp screen is missing. Is baro altitude still used for climb rate or something? I have the version with BME280 sensor.
• MB180 only; RX packages number adding up slowly over time when the t-echo is laying around for a while. No aircrafts or FLARM devices in the proximity. This does not seem to happen yet in MB202 or mainstream SoftRF (might be under different settings/ conditions). Must valid FLARM information be received to have this RX value adding up? Or can it be disturbing signals, where the MB180 firmware is more sensitive too, in my area? 
• MB202 only; has a constant blue LED. Other versions blue over green flashing, till bluetooth conection is made than only constant blue. Not sure if this was a deliberate change? Can you elaborate on what different LED colours mean?
• In the settings.txt file there are some settings.

units,0            # 0=metric 1=imperial 2=mixed

zoom,2

rotate,0

orientation,0

adb,0

epdidpref,4        # 0=reg 1=tail 2=model 3=type, 4=hex

viewmode,1         # 0=status 1=radar 2=text ...

antighost,1        # 0=off 1=auto 2=2min 3=5min

In your "under the hood" document these are called differently like al with epd_  in front. I assume the upper labels are correct as a change has effect.

I hope jou can use my experiences so far. I'm also happy tot test your MB303 and future versions on my T-echo plus!

Cheers,

Rens

 

Op vrijdag 29 mei 2026 om 21:56:11 UTC+2 schreef Moshe Braner:

[Moshe Braner]

Thank you Rens for the detailed feedback.  Versions MB20x are still in beta test.  I can try and fix some of those things "blind", but alas my T-Echo died (I suspect the bootloader got corrupted, but I don't know how).  More to say after I work on these issues.  After that I will be happy if you (or others) can test it again.

The regular T-Echo does not have an IMU?  Do you have a T-Echo Plus?  Maybe that would explain some of the differences in behavior of your T-Echo Plus and my plain T-Echo when it still worked.

Regarding the settings displayed in the summary screen, P:T+A means laTest as main protocol, Ads-l as secondary/alternative.  The A:LAT means the "latest" collision Alarm Algorithm.  This is in the User Guide.  Being able to change a few of the most important settings with the buttons, without any app or other device, is nice.  But you get more complete control, and less button-pushing-and-touching, via either editing the settings.txt file, or using Vlad's Web-BT HTML page.  It is still in flux, but I'll release MB203 when it and that page work together well.  Until then, simply connect to a PC via USB and open settings.txt.  Also I'll update the documentation.

Yes, now that I realize that the T-Echo has an RGB LED (next to the EPD screen, separate from the other LED) I did change what it does somewhat.  Red while booting, green after that, blue when it has a GPS fix, IIRC.  It does not show BT status.  That too will be documented.  If you prefer a different color scheme, let me know.

Similar RGB color scheme for the M3.  I still don't know how to use the LEDs on the M1 (Oliver?) because I have not had one in my hands.  The T1000E only has red and green LED colors so flashing green until GPS fix.  The M3 & T1000E don't have any display at all, thus the Web-BT HTML page - and, on the T1000E, editing settings.txt - are the only ways to configure them.

[Alan Hall]

Has anyone else tried V2-02 on the original TBeam hardware version? I am having problems and before investigating further it would be great to know if it’s just me. Tests were done using “Latest” and “Pilotaware” protocols with identical results.

I have two units with PCBs marked “T22 V0.7” and which identify in the diagnostics as “Tbeam rev 5”. They both work perfectly on V1-80. They have SX127x radios which may be the significant factor given the change of radio libraries. 

 On V2-02, one of them works but the receive sensitivity is 30dB lower than with the V1-80 firmware, i.e. it’s deaf as a post! Transmission is fine.

 The other does not transmit or receive at all on V2-02 although otherwise it seems heathy. The serial debug constantly prints “tx timeout”.

[Moshe Braner]

Hi Alan.  Thanks for reporting.  There must be issues with the Radiolib interface to the sx1276 radio module.  I have been concentrating on developing the next version on the nRF52 models (T-Echo, T1000-E, etc).  I will get back to working on the T-Beam version soon.  Meanwhile use MB180 on the T-Beam.  - Moshe