lament geode
Hey all, we have 2 NewBeeDrone hummingbird v3s, tx16s binds to both without issues, but boxer only binds to one of them. Many many firmware flashes later, configuration changes boxer still fails to bind to one of the drones, I'm out of ideas what could be wrong. Does anyone have any sugestions? Thanks
mossy girder
are the settings in the lua script the same on both radios?
lament geode
Yeah, i believe every single combination of settings was tested on the boxer. Stangest thing is that boxer binds with no issues with one of the drones, but not other. Even though both are the same (same hardware revision and betaflight version). Betaflight configuration profile were coppied over from the red drone to blue one to see if its any help in getting the boxer to bind, but did not help.
Issue seems 'deeper' than what we could change with settings that are present on user interface
mossy girder
Yeah, i believe every single combination of settings was tested on the boxer. St...
on the non-binding drone, could you go into the CLI tab and execute version please
lament geode
Blue one
Red one
Betaflight / HUMMINGBIRD_F4_V3 (HummingBird F4 V3) 4.4.0 Sep 18 2023 / 18:08:31 (norevision) MSP API: 1.45
post screenshot/photo of the expressLRS lua script on the tx16s then from the boxer
also get the output of get expresslrs from both whoops and compare
lament geode
Boxer is now on EU LBT in hopes that that will change anything, did not help
Outputs from get expresslrs are identical, except for line containing expresslrs_uid
# get expresslrs
expresslrs_uid = 0,0,149,209,176,210
Array length: 6
Default value: 0,0,0,0,0,0
expresslrs_domain = ISM2400
Allowed values: ISM2400, CE2400
expresslrs_rate_index = 2
Allowed range: 0 - 3
Default value: 0
expresslrs_switch_mode = WIDE
Allowed values: WIDE, HYBRID
expresslrs_model_id = 255
Allowed range: 0 - 255
but this is the important bit
the UID is the equivalent of the binding phrase
that last four numbers should at least match
lament geode
When profile was coppied over from red to blue, this setting was synced up. However boxer still did not bind
mossy girder
put tx16s into wifi mode, load the webui, take note of the UID..
exit wifi mode on the tx16s
put the boxer into wifi mode, load the webui, take note of the UID
exit wifi mode on the boxer
are those UID the exact same numbers?
do they match with the drones' UID
again, the UID is the translated binding phrase
lament geode
Ok, give me a moment, will try that.
mossy girder
if the UID on the tx16s and boxer are the same, and they match with both drones' UID, then they should all work interchangeably
(this is why the binding phrase is a much easier binding procedure)
lament geode
Both drones had UID from tx16s. Although first 2 bytes were different, remaining 4 identical.
Tx16s firmware: overriden 205,124,149,209,176,210
Red: 0,0,149,209,176,210
Blue: 205,124,149,209,176,210
Boxer: flashed 148,112,120,112,69,238
mossy girder
Boxer: flashed 148,112,120,112,69,238
so the boxer shouldn't sync with either
update the binding phrase on the boxer so it matches with the drones and the tx16s
0,0 as the first two UID numbers meant the the drone was bound traditionally
lament geode
I attemted to manually set boxers uid on the red one, there now seems to be stange behavior in the receiver page
mossy girder
why?
it's the boxer that you should've set the binding phrase/UID on since that's the only one that's different
mossy girder
switch mode is different on the boxer
the boxer is also on LBT while the tx16s is on ISM
lament geode
Yeah, but LBT boxer works on the red drone 😦
mossy girder
are you from the EU?
lament geode
Ye
mossy girder
so flash LBT on the TX16s
their regions should be the same
their UID should be the same
lament geode
Agreed, but that's that's seems to be completely unrelated to problem I'm having. It would make sense if one of the controllers work with both drones, and another with none.
But thing is, that drones seem identical, however boxer only works with one of the drones, and not another
mossy girder
the boxer has a very different UID than the rest.. and it has a different region as well.. it WILL not sync with any of the drones
lament geode
But it does, just fine with the red one after going through standard binding
mossy girder
that just doesn't make sense
lament geode
But it does not bind with blue one, even though tx16s can bind with both just fine
lament geode
that just doesn't make sense
Aye, thats why im here, because it's a mystery to me
mossy girder
and hence why I am advising setting these devices into the same region and binding phrases/UIDs
take it or leave it
lament geode
Already did that before flashing EU LBT on boxer, thats why they different now.
Could something like imperfections in crystal oscillators have enough impact that underlying lora fails to decode on boxer?
lament geode
Added one more test case to the list. Pocket binds just fine with the blu, boxer binds just fine with yellow. Just click button on drone and bind in ELRS lua script and it all works.
Binding between blu and boxer continues to fail...
mossy girder
have you reflashed the boxer to what it should have?
lament geode
Boxer remains EU LBT as that is the correct domain
mossy girder
since you're not too keen on following my advice, I don't see the point of posting here anymore..
lament geode
Will try that again, but i'm grasping straws as the 3rd sample does make situation any more clear. Just a moment
mossy girder
reflash the boxer with the same domain as the working radios, with the same binding phrase as the others
lament geode
Reflashed boxed with ISM, 3.4.3 with same uid.
Drone has same uid configured.
Tx16s when powered on auto-connects to drone, boxer does not; no signs of life on the betaflight receiver tab.
Packet rate, telem ratio, switch mode are identical between radios
mossy girder
and boxer still syncs with the other drones, but not this particular one?
lament geode
Yup
mossy girder
so now we can easily form a difinitive conclusion since everything is in the same parameters
lament geode
And this particular drone with every other radio we have here (except the boxer)
mossy girder
forming a conclusion with different parameters that's proven will not work anyway is pointless
and yes, a slight variation on the receiver's crystal freq, especially if they didn't used a good tcxo can be a factor why this drone syncs with the other radios and not with the others..
lament geode
I understand, and thats fair enough. It's just before comign here many hours were spent and frustration is strong
mossy girder
then again, this particular drone and nbd in general doesn't seem to be validated, with both the betaflight project and the expressLRS project..
they're even using their own firmwares here and no official betaflight target is available
lament geode
I've got all the tools, steady hands, micro probes, ~decent~ oscilloscope and patience. What could I try to narrow things down?
mossy girder
I've got all the tools, steady hands, micro probes, ~decent~ oscilloscope and pa...
since it's the receiver on a particular drone that isn't working, that's pretty narrow now..
and there's just really one xo for the sx128x rf chip on the drone..
so if you can replace that xo, it might fix the issue
not sure if a tcxo is a drop-in replacement, but I'd imagine there'd be a couple of passives required that's different for a regular xo
lament geode
All sx128x designs use 25mhz xo or they may vary?