hello all!

recently i have been flying my AOS5 bnf for endurance purposes as a test bed for my future lightweight 5in LR build. here is the setup:

stock iflight AOS5 V2 HD bnf running 6S, 2207 1855kv motors, spinning biblades ethix P3B 5130. empty weight 400g

lipo 1300mah tattu R-line. 200g

ELRS settings:
transmitter power set to dynamic, packet rate set to 50Hz on both RX and TX and BF4.4.0 presets.

all LR test flights done in urban environment suspectedly RF noisy but also done at high altitude around 130m at the farthest point 1.5kms from home. i have not yet ventured any farther for safety reasons since my vtx antenna can get obstructed by the lipo pack.

gps rescue with BF4.4.0 actually works like a charm (with M10 chip module) despite the bucking bull flight behavior during the return flight to home point. otherwise it works and was tested up to 1.4km out under some heavy crosswind.

FYI the T-antenna is mounted stock on the left motor arm as recommended by chris rosser. 

now here’s the quirk if it were a quirk at all: during the return phase of flight beyond 800 meters away, there are times when i get RXLOSS flashing on the OSD along with LQ value decreasing to 50-70% momentarily BUT handling is not affected at all, i relinquish control immediately once i move the sticks. 

there was also a couple of times when i actually experienced a “micro failsafe” if it were indeed what i thought. i get RXLOSS and the quad immediately enters angle mode (as i have set in failsafe tab) but i also relinquish immediate control after moving the sticks.

just today, i also heard an audible warning “SENSOR LOST” for the very first time mid-flight at about 800m away from home LOS unobstructed. oddly enough the LQ and RSSI values were stable as i recall the entire time i was hearing this advisory for at least 3x.

on all these RXLOSS and low LQ events, despite the audible and OSD warnings the quad was actually always in control and never really degraded into a total loss of RC link or went deep into an unrecoverable failsafe. i suspect the RF noisy location is not doing me any favors and perhaps these acute events can be attributed to momentary interference.

SO anyone here who can explain my actual observations? or better yet had the same experience with ELRS?

It's possible that you're either changing your antenna polarization, or you're pointing a null back at yourself. These are standard characteristics of linearly polarized antennas.

I don't know for the RXLOSS, but I know wrt telemetry it happens a lot on the radio side and you can improve it by changing the packet rate ratios, but I still use 1:64 or 1:32 most of the time without issue. I get telemetry lost occasionally on 1:32 with 250mW / dynamic, but it's mainly when i'm say flying in a different quarry to the one I took off from. if you can keep your VTX and Rx antennas away from each other, do so.

One thing you might want to do when you come to do the real build is setup two antennas using a diversity ELRS board. Mount one horizontally and the other vertically. That would (hopefully) give you the best signal pickup as possible. For me personally i would probably use a diversity setup, but in my case mount a dipole between the front legs and one between the back legs, so as long as you are moving in a straight line, you should have LoS to one of those dipoles always. On the Tx side, I would run at 1W, but keep dynamic enabled. Technically my radio will take me out to 5km since it does support 1W Tx, but whether I would want to is a different thing. But based on the way antennas work, I think if you mounted one front, one back and kept them both horizontal, it should work. But I only know stuff from watching other LR people doing flights and tutorials, so take it with a pinch of salt. Test it for yourself and find something that works for you.

(19-Jan-2024, 09:37 PM)Suros Wrote: It's possible that you're either changing your antenna polarization, or you're pointing a null back at yourself. These are standard characteristics of linearly polarized antennas.

my present T antenna install is horizontal on both the quad and boxer transmitter so no cross polarization per se. BUT when i do make the 180 turn going back this motion plus any bit of interference on where i make that return could trigger the acute LQ/RSSI degradation events. this could be why i get this event worse on some places, NIL in others. but i suspect the arm mounted rx T antenna greatly aggravates the issue as the turning motion obviously points the null at my location especially at greater distances. i suppose the slow and shallow 180 turn aggravates this even more. maybe i should just make a quick 180 pivot to mitigate this???

i shall make an effort to reposition the antenna in a vertical orientation as the video recommends and see how it works! see below my present reluctant LR test bed:


if the lipo pack were only bottom mounted then the same motor arm mounted installation done vertically would have instantly fixed the issue.?

(19-Jan-2024, 09:45 PM)Pathfinder075 Wrote: I don't know for the RXLOSS, but I know wrt telemetry it happens a lot on the radio side and you can improve it by changing the packet rate ratios, but I still use 1:64 or 1:32 most of the time without issue.  I get telemetry lost occasionally on 1:32 with 250mW / dynamic, but it's mainly when i'm say flying in a different quarry to the one I took off from.  if you can keep your VTX and Rx antennas away from each other, do so.

One thing you might want to do when you come to do the real build is setup two antennas using a diversity ELRS board.  Mount one horizontally and the other vertically.  That would (hopefully) give you the best signal pickup as possible.  For me personally i would probably use a diversity setup, but in my case mount a dipole between the front legs and one between the back legs, so as long as you are moving in a straight line, you should have LoS to one of those dipoles always.  On the Tx side, I would run at 1W, but keep dynamic enabled.  Technically my radio will take me out to 5km since it does support 1W Tx, but whether I would want to is a different thing.  But based on the way antennas work, I think if you mounted one front, one back and kept them both horizontal, it should work.  But I only know stuff from watching other LR people doing flights and tutorials, so take it with a pinch of salt.  Test it for yourself and find something that works for you.

i did read something about the packet ratios which may fix the audible “telemetry/sensor” issues. good thing it’s limited to the RX capability to send back packets to the TX and not something that will voodoo spell the TX into telling the quad to failsafe!?

i am also seriously considering the dual antenna set up which affords redundancy. another option is actual dual diversity receiver (RP4TD) which i see is already available in some online stores. would dual antennas only be adequate for the task? my only nit for the dual antennas is why they provide an oddly short lead wire for both T antennas! why not make the other one long enough to straddle the entire quad length?????

If range is a big deal for you, definitely grab a diversity receiver. I'm planning on swapping my diversity setup from an L configuration to a V configuration when I can 3D print again. It's not like I'm doing any rolls, so it will keep one antenna visible at all times rather than getting blocked by the battery.

1) What radio do you have? What EdgeTX version are you running? What telemetry ratio are you using?
Some sensor/telemetry lost messages can be due to a timing difference between ELRS and OpenTX/EdgeTX. This has been resolved in recent EdgeTX versions. Also, running a 50Hz packet rate means you need a high telemetry ratio (1:16 or greater, or select Std) to have enough telemetry bandwidth to send both link stats (~2 packets per second) and telemetry from the FC.

2) Have you done a bench test to confirm TX and RX are working properly? Follow the procedure here, and report the observed RSSI dBm:
https://www.expresslrs.org/quick-start/p...bench-test

3) What ELRS version and regulatory domain are you using? If in the EU, LBT can significantly lower LQ. I would avoid using it if at all possible.

4) A horizontal antenna mounting is not ideal due to nulls, as discussed in the other posts above. Ideally you would have the antenna mounted vertically and as high as possible so it is not shielded by the battery or carbon. Using something like a TrueRC BARDpole can help with this, and Prodrone makes excellent, inexpensive antennas with longer coax for ease of mounting: https://shop.prodrone.pl/en_US/c/2.4Ghz-Antennas/22

5) Diversity and especially true diversity receivers (eg BetaFPV SuperD, HappyModel EP1 Dual, RadioMaster RP4TD) are a good option as well, as you can position the two antennas to cover the other's nulls. Even an EP1 should work extremely well at 50Hz and less than a few km range, however, so make sure you do the benchtest above to confirm everything is working properly.

(19-Jan-2024, 11:39 PM)V-22 Wrote: 1) What radio do you have? What EdgeTX version are you running? What telemetry ratio are you using?
Some sensor/telemetry lost messages can be due to a timing difference between ELRS and OpenTX/EdgeTX. This has been resolved in recent EdgeTX versions. Also, running a 50Hz packet rate means you need a high telemetry ratio (1:16 or greater, or select Std) to have enough telemetry bandwidth to send both link stats (~2 packets per second) and telemetry from the FC.

RADIOMASTER BOXER, EDGE TX VER.2.8.0 stock factory, i selected 1:8 telemetry ratio, will see what happens!

2) Have you done a bench test to confirm TX and RX are working properly? Follow the procedure here, and report the observed RSSI dBm:
https://www.expresslrs.org/quick-start/p...bench-test

JUST PERFORMED BENCH TEST. i had a good reading of -27dbm rssi via osd.

3) What ELRS version and regulatory domain are you using? If in the EU, LBT can significantly lower LQ. I would avoid using it if at all possible.

DEFINITELY NOT LBT as i recall from the last check.

4) A horizontal antenna mounting is not ideal due to nulls, as discussed in the other posts above. Ideally you would have the antenna mounted vertically and as high as possible so it is not shielded by the battery or carbon. Using something like a TrueRC BARDpole can help with this, and Prodrone makes excellent, inexpensive antennas with longer coax for ease of mounting: https://shop.prodrone.pl/en_US/c/2.4Ghz-Antennas/22

YES! aside from the other nitty details i think the motor arm antenna install is calling it a stretch to perform well flying LR. i am already considering a vertical orientation with some type of mast mounting. sadly the T antenna lead may not be long enough for this… yet!

5) Diversity and especially true diversity receivers (eg BetaFPV SuperD, HappyModel EP1 Dual, RadioMaster RP4TD) are a good option as well, as you can position the two antennas to cover the other's nulls. Even an EP1 should work extremely well at 50Hz and less than a few km range, however, so make sure you do the benchtest above to confirm everything is working properly.

THIS ELRS TO-DO LIST IS MOST APPRECIATED SIR! dang i just unboxed the TX and quad and flew. it was excellent as it is in stock form and has impressively good range at 1km+ despite the less than ideal antenna location and orientation. now the objective is to optimize everything for reliable LR performance. cheers!

hello again to all of you who responded to this thread!

first off let me eat some humble pie. on the TX settings i have set transmitter output power to dynamic, 250mw only, thinking that i may have read in the past that setting dynamic to ON will output 1W when needed even when the selected output power was lower eg. 250mw in my case. BUT i may have also understood this incorrectly as manifested by my recent experiences with various events of degraded rc link in the form of audible warnings, flashing osd advisories, diminishing LQ and RSSI, etc! 

SO as suggested i set output power into the 1W full monty! i also set TLM ratio to 1:8 and still haven’t repositioned the T antenna… yet! i just came in from a test flight up to 1.5kms outbound and the whole flight was blissfully silent and uneventful. no audible warnings, no osd alarm at all! 

i did notice during the outbound leg TX output power was a steady swing from 100-250mw only like before up to 1.5kms. BUT! when i returned there were episodes of the TX switching the full 1W for a few seconds only, then dropping to a low of 100mw, then 50mw even at a distance of 800m from home. during the entire time i had rock solid LQ that barely hit the 90s region and RSSI was a modest -80ish only. a far cry from the previous outings where i had LQ dropping momentarily to mid 50s which can be very unsettling at best!

i can only guess that if my antenna installation were fully optimized for LR then i don’t think the system may even need to switch the full 1W? is this a correct assumption??

i was very fortunate that my misinterpretation of the dynamic transmitter output settings did not ever mature into a full blown failsafe event during my initial LR endurance attempts. i suppose this is testament to ELRS’s stellar performance! after setting the full 1W output on demand i anticipate this thing can go reeeally far especially when i optimize the antenna position despite my RF noisy urban environment.

cheers!

It's definitely worth noting that with a proper antenna configuration, vast range is possible on next to nothing. The ELRS range contest has a couple of people doing 10km on 10mW. Improving your antenna configuration would certainly offer you much higher reliability without blasting a full watt. My own dynamic power doesn't go past 100mW until about 4km.

(21-Jan-2024, 01:12 PM)Suros Wrote: It's definitely worth noting that with a proper antenna configuration, vast range is possible on next to nothing. The ELRS range contest has a couple of people doing 10km on 10mW. Improving your antenna configuration would certainly offer you much higher reliability without blasting a full watt. My own dynamic power doesn't go past 100mW until about 4km.

ah yes the antenna orientation alone did quick wonders for the rc link! i repositioned the T-antenna in this ghetto inspired design for a quick test flight…

placing it on the very lowest, unobstructed location on the quad offered excellent LQ which barely budged out of :100% on a 2.5km, 130m altitude outbound flight. output power was a steady 25mw rising to 100mw till 2.5km. while turning back it momentarily spiked a full 1W during the null which immediately dropped back to 50mw, then 25mw on the return leg. the entire flight never had any audible or flashing advisory at all. BUT i am still going for a vertical installation which may have the best perfomance during the nulls on the turn. 

i saw a 15.5cm T-antenna from foxeer which will afford a lengthy run to some kind of tpu print mast mounting against one of the frame standoffs. if i get the dual diversity RX i can do twice on either side.

(23-Jan-2024, 10:46 AM)hawk01 Wrote: ah yes the antenna orientation alone did quick wonders for the rc link! i repositioned the T-antenna in this ghetto inspired design for a quick test flight…

placing it on the very lowest, unobstructed location on the quad offered excellent LQ which barely budged out of :100% on a 2.5km, 130m altitude outbound flight. output power was a steady 25mw rising to 100mw till 2.5km. while turning back it momentarily spiked a full 1W during the null which immediately dropped back to 50mw, then 25mw on the return leg. the entire flight never had any audible or flashing advisory at all. BUT i am still going for a vertical installation which may have the best perfomance during the nulls on the turn. 

i saw a 15.5cm T-antenna from foxeer which will afford a lengthy run to some kind of tpu print mast mounting against one of the frame standoffs. if i get the dual diversity RX i can do twice on either side.

thought i’d extend this thread to monitor the progress (or lack thereof) of my reluctant AOS5 bnf as a test bed for endurance and range testing for an upcoming lightweight LR build. here is the latest antenna tweak i have done in the quest for the best possible RC link with a single T-antenna. 

notice the 100% organic, non conductive, RF inert bamboo skewer material carefully wrapped in electrical tape. this afforded the farthest location of the active element from any frame surface:



two flight tests out to 1.5km going to different locations in my RF noisy urban environment yielded better results than the previous install. this one consistently had the most stable LQ and RSSI values, again no RXLOSS warnings, most apparent change being the (very) momentary 1W dynamic output power during the turnaround.

i did not fly all the way out to 2.5kms as i noticed video feed mildly stuttered at high altitude on the return flight possibly due to the vtx antenna obstructed by the lipo pack.

this poses the question, if i install dual antennas are they both supposed to be horizontally oriented like the TX or can one of them be oriented vertically? 

if both were to be horizontally oriented, i plan to have them perpendicular to one another in order to address the nulls. if both were to be vertically oriented, i have to find a way to route one of them aft most of the quad in order to address possible blind spots.

inputs are highly appreciated!