Yaw Washout explained and what it means for DJI Avata owners
Вставка
- Опубліковано 22 чер 2024
- All multirotor drones suffer from the effects of yaw washout to some degree but "whoops" are most affected, for the reasons explained in this video.
For examples of yaw washout affecting the DJI Avata, check out these videos:
Mads Tech demonstrates the bigger issue:
• DJI Avata Manual Mode ...
Total loss of Avata drone caused by yaw washout:
• DJI Avata FPV FATAL CR...
Avata falls into the water with this issue:
• DJI Avata fell in to t...
FC Crash initiated by switch from normal to manual mode?
• DJI Avata nose dive issue
Feel free to become one of the heroes who support this channel via Patreon:
/ rcmodelreviews - Наука та технологія
"The Flight Controller's given up and said 'I'm heading right back to China'" This guy 😂
Best quote 😂
Lol yep
You make it all sound so easy... "avoid flying into the ground". The Rutherford quote about clear explanations springs to mind, I'm sure it's way more complex under the surface.
I literally laughed at the diagrams! That's just brilliant. I don't even care if I agree with you or not after that! Hahaha!
My art teacher would be proud (or rolling in her grave) :-D
Best (and only?) explanation of yaw washout on the internet!
I like these sort of videos the mostest. Really this is keeping within the manufacturer limits in flight
As always Bruce - great explanation. 🙂🙂
A Great post with a simple explanation.
Hi Bruce, thanks for the description of yaw wash out, very clear and well presented. I’m only flying the Avata in Manual mode, and it seems to perform well, I guess I’ve been lucky, so far… I have noticed a small amount of wobble in fast yaw turns, both left & right, but it very quickly recovers as you describe. I have upped the RC rates centre 400, max rate 1080 and an expo 0.54, to match my 5”, and I am now very happy with flight performance. David
Wondering if 'perhaps' just an issue with the DJI Default rates. To date, I've not heard the problem replicated from anyone who applied their own customer rates.
I think wind velocity and angle have a lot to do with it. Air is like invisible water just like he was saying about swimming in a waterfall. I’m a complete noob to flying acro and I’d say another key thing is to not go back and forth between acro and the other modes. Every crash I’ve had has been muscle memory going back to using the left stick for elevation and right for speed and direction after studying the video. It’s just a twitch at the worst possible second and I know it’s 100% me. I changed my rates to what murdersFPV runs, got longer sticks, and since my second battery pack ever flying acro I’m amazed. I’m flying just like in the sim and feel like Mr. Steele hahahaha
@@WarriorLife_fpv Bruce description of yaw wash out is very good, and I like the waterfall analogy. And as Bruce describes I have experience a small wobble during yaw turns and luckily nothing more… so far…
this happen because of popular guards ;
people are using it bad watch this ; this explains all :
ua-cam.com/video/u6EH_QGRsFE/v-deo.html
Thanks for the heads up and great info! I’ve never flown my DJI Avata that crazy yet to experience what the others are. So far so good!
Such an amazing source of knowledge, thank you, sir!🙏🏼
Great explanation, Bruce! Thanks! 😃
I hope DJI does something to fix it!
Anyway, stay safe there with your family! 🖖😊
Great information!
I have yet to test this. I’ve been flying DJI Avata just to get more creative with my shots and aerial content has been nothing but much more thrilling to watch thanks to this drone.
When it happened to my Avata, during a filming job, the left yaw rate was very gentle. The wind was calm. Nothing was yawing, pitching or rolling fast. I was at full speed in Sport mode, chasing a side-by-side down a road. I gently yawed to the left enough to slowly move to the left of the road into a field. All of a sudden, it did the fast pitching motion for a moment. During this, I had no control, but did regain control before it fell very far. I continued the chase but it was totally unexpected.
Wooooow
Very high quality content! Thank you so much ❤️
you to me are the true guru man, I watch, listen and LEARN so much from you, one day I'll learn how to make my own diversity receiver from you for fatsharks, the parts are cheap id like to do it myself, but there's no real tutorial out there, not like the way you'd do it anyway(in my friend group I refer to you as the holder of the matrix , Quaddamus Prime !) my crew says "reeespppect !" we love ya dude!
Thanks I learned something great video!!
Thanks Bruce !
Almost lost an Avata at work today to yaw washout, and as a 5-inch pilot the vast majority of the time I had no idea what the hell happened when I tried to pull a high-G yaw/roll turn and the Avata flipped itself over and went straight into normal mode, locking me out of manual until I reset the controls. Thank god I had 30 or so feet of altitude because it very easily could've blown itself to pieces on the asphalt. Never had that happen on any of my quads and I'm trying to avoid having an expensive conversation with my boss.
1st..love the drawings! Spot On! 😂😂. Great overview of the issue people are reporting. I have the AVATA and can confirm it’s definitely limited to “manual more”. I’ve seen none of these issues in GPS modes (Normal & Sports Mode). Your the “calm voice” of reason on the crazy UA-cam world! I’ve totally customized my AVATA by hard wiring a GoPro Bones to it so not going to speculate on issue. Hopefully, it all gets figured out. Always enjoy your take on a subject. 👍👍👍
Interesting insight about only happening on LEFT Yaw turns only (not right). I didn’t pickup on that until now! 👍👍
@@RGMGFitness I think that's just because people mostly turn left with quads. To me it only happened twice in hard right turns, once 120-150cm off the ground and recovered both times.
@@FilipRadelic Yes, thought about that too (about more left turning folks). Interesting, that it happened twice to you (on RT Turns). I did see one person on UA-cam (OriginaldoBo) try to re-create the issue but said he couldn't. However, he had changed his rates from the DJI defaults. Hope it gets figured out....sounds like quite a few folks are having issues. Perhaps only an issue with 'default rates'...but don't know until some more testing gets done.
@@RGMGFitness there might be something about the rates. I lowered my rates after it happened (not because of it, but because the original values are ridiculous even for freestyle, let alone cinematic 😄) and I haven't seen it again.
@@FilipRadelic I hear ya. The Avata is my cinematic quad. Not planning on doing anything crazy with it. See a lot of folks flying that thing way to aggressively.
Thanks Bruce.
It's not really a duct, doesn't have the effect of one (would have to be deeper to be a duct, as with the first drone you showed). It's simply a prop guard.
I love your diagrams!
Thank you so much,
wow this video is so good, thank you so much, please continue with this avata topic. I have two avatas, I don't know that to do know. My avatas directly start to make 180 degree tumble when I make a fast flat yaw turn
Don't do that...? Lol
What a great explanation.
DJI are not taking responsibility for this.
Are you Aussie? Where? So good to meet a smart one. Not common.
Taking responsibility for what?
I think the software may be contributing...in some videos the home point moves during flight maybe an indication of more issues...just a thought......great explanation!
thanks for the educational video, i will go to bed smarter tonight 👏👍👍
Wow! Luckily I’m just happy flying my DJI Avata on normal mode. I guess others need to chill a bit and not fly it so aggressively. I think DJI Avata is something you fly to get cinematic shots while also having a more adventurous moves but nothing too crazy. It’s a fantastic tool if used correctly. But thanks for all the info. I’ve got so much to learn first. Just my humble opinion.
17:19 I only have around 30-35 flights on mine and had this happen once about 120-150cm off ground while I was doing a hard *right* yaw pushing it chasing a dirt bike. And it didn't crash, stabilised a few cm off the ground. I think the reason it seems to mostly happen in left turns is people generally prefer turning left with quads 😄
Anyway, a bit early to say, but I'm almost sure it's overblown way out of proportion based on my experience, and I wouldn't normally fly a whoop like I did when it happened to me, but Avata kept up with the dirt bikes so well that for a second I forgot I was flying an underpowered brick. Still, DJI should definitely look into it and do whatever can be done to make it less likely to happen, or at least to make it more likely to recover if it's physically impossible to prevent.
I 100% agree with you on the left hand turns hahahaha! I literally had the same thought today. I think it also has to do with wind direction and speed while executing the movement. I’m a complete noob but soon realized that there was a reason that birds performed their movements gently into and out of strong currents instead of balls to the wall. Gravity. Lol
I hope dji can make it better with some updates. I dont fly manual with the avata that often, so I have never experienced this. Feels really stable in Normal and Sports mode. So over water I just fly careful without changing direction that quick.
Great explanation Bruce. So you chase fast moving vehicles at your own risk if you want to keep them in shot?
The short answer, its not a race quad, its not a freestyle quad its a cinewhoop. Its for a very specific type of flying/filming and if you fly outside of that you could possibly get a washout flip. if you are flying faster than a slow jog or in strong winds get racer or a freestyle not a whoop. This is DJI first cinewhoop I am sure they have a lot to learn.
RobHK (I think his name on UA-cam is) said in a post within the last few days that DJI had "told him" that they've fixed "the problem" in a firmware update. He was the guy who lost his flying over rocks near the ocean. Here's the quote from a post of his "I was confirmed by DJI, that the crash problems is fixed with the latest firmware update."... So if that's to be considered true and correct, DJI have acknowledged an issue, at least.
no they did not, they say that they fixed it but it still happens on my Avata
Really bad because this might also happen if you fly manually in turbulent strong wind conditions which normally isn't really a problem for most drones.
Yay, an other whiteboard class! 😊
Teach teacher.
Interesting idea but no, the difference between the Avata and your other Cinewhoops is that it is top-heavy which initiates FC freakout and tumble.
I think the issue is there rate controller for there drones is quite weak and it seems to not be able to deal with a situation of excessive error very well.
They should put a form of anti crash into there firmware like the feature in betalight where when it recognises its lost control it switches quickly into a mode using the accelerometer to help it regain control of the drone.
It looks to be props in. I wonder if it would still happen with props out. What do u think?
Ring vortex used to happen on my small BRUSHED whoops :-)
You have to take into consideration that DJI makes film making tools not race quads or freestyle quads. The average film maker is going to most likely fly at much slower speeds and smoother motion. If the film maker wants to shoot more fast pace aggressive filming they will use the tool that DJI created for that. The DJI Fpv, was made for aggressive fast pace flying. The avata was designed for a cinematic fpv filming experience in close proximity to people and structures with smooth flowing movement.
They did limit the yaw rate by releasing it with the motion controller. The dji fpv was for fast filming for chasing cars and stuff like that, the Avata was not. If you use these tools for what they were designed for, you’ll likely have no problems. The real problem is trying to race a donkey like a thoroughbred
@@ChulleyChupster predictable
Sounds logical. As a newbie with only the Motion Controller, I've yet to encounter any problem. But I'll keep an eye on those left hand turns. Have you seen the video of the guy flying over a river with the motion controller and his AVATA kept losing altitude? It then started flying in an unstable manner and went backward which isn't supposed to be possible. He attempted to get it to shore, but it smacked into the river and drowned. He killed comments for some reason, but the footage is still up for viewing. Flight Controller error? Love to hear your take. It was posted Oct 4th on Gary Bell's channel. The title is DJI AVATA Malfunction.
mine is crashing too? can they fix this with software??
I wonder if it is suffering from Gyro Overflow. 🤔
I think problem is connected with safety features inside software, betaflight fights to keep quad till is disarmed. Problem is exagerated by worse power to weight ratio due to big battery and efficient powertrain not optimised for raw power.
Props out vs props in issue?
Every DJI drone has tilted motors to aid yaw. But this one doesn't because of duct I guess!
Pushers suck, imho. Whoops handle far better, and are much quieter. The top-mounted battery is less than optimal, compared to a transverse, under-slung mount. My daily driver's a 100mm, 2S beastie, 125g AUW. I run 1104 7500 Kv motors with Gemfam D51 ducted props. Absolutely zero yaw washout. I've tried it several times as a pusher: it's twitchy, less efficient, and sounds like I'm making a smoothie with live kittens. Could be the duct's shape, but it felt weird in the handling department. That broomstick feeling sucks. . .
You cannae change the laws of physics Jim lol 😂
Hi Bruce, Great video, leave it to you for a 20 minute video on why a brick won't fly. I have been saying from the beginning that I didn't think that the avata ducts were not going to work. They designed it with 1/4 of the ducts covered and the motors are to small, and of course it's a pig. Thanks again for the information on why the AVATA is going to fail.
I have been saying the same thing. The battery and the frame is covering too much of the duct. It is probably unstable flying normally. I now know why my ET125 doesn't have this problem. The battery is on the bottom(not covering a duct) and the canopy is very smooth on top, offering little resistance to air coming from the side.
Flys great… Dont know why or if there are so much big problems…
ua-cam.com/video/sFU1C53pT0I/v-deo.html
Wow im seeing Avata issues all over UA-cam now, so many people posting how it's just crashing especially in manual. There is something seriously wrong with this drone. Im loving my Dji FPV i was thinking about getting the avata to film in tighter spaces and also for my real estate business but idk man i think we need to wait a few more months for this drone to get some firmware because people having them go haywire with simple turns mid air and just falling to their death either on land or water, not good at all if they cant fix this with a firmware then sadly a recall
Funny, it reminds me of how the dji Naza in the early days would not handle high 'G"s... remember it would start to barrel roll in the air as we got to be better pilots and make sharp moves... so may be it is more fc limitations than aerodynamics.... but still... I love that Avata...
I think it's a return to home issue. It's returning to the ground instead.
I really feel like DJI has themselves in a real pickle. Their customer base just wants to fly things without needing any knowledge of how things work, and then apparently manual mode really isn't full manual mode as the Avata appears to try some sort of self recovery. IMO it might just be better if manual mode was really acro mode light Betaflight
Each time you attempt a silly yaw spin, Goose spins in his grave.
I prefer the Dsynch explanation of Justin Davis, and the physic efect of mass top in a long extensión of frame body and post tumble with the inmediate trust lost of Joshua BARDWELL....
Desync is a possibility -- but that doesn't explain the post-flip loss of control that some people are experiencing.
@@xjet mmm....post flip or tumble because lost totally trust and the weight its all on top...its something usual on cinewhoops...
@@humbertocortes68 Also, if this was just a motor desync then I would not expect it to pitch inverted without any rolling movement. In the instances where the yaw washout triggers this flip the Avata almost always pitches nose down. That would require both the front motors to desync simultaneously -- an unlikely situation.
@@xjet may be....
.....the time will tell us! The problem exist and dji need fix it!
@@humbertocortes68 Yet Ken (OriginalDoBo) believes there is *no* problem. On his community tab he just stated _"Drone manufacturers aren’t responsible for your ignorance"_ . I guess part of that reasoning could be all the DJI affiliate links in the descriptions of his videos 🙂
I have a couple good examples of this happening to my Avata..
Including a right yaw example!!
Yeah mine had the exact same issue. Only lasted 16 days before it killed itself. Really wish I would have seen this before buying an Avata. Big regret. ua-cam.com/video/J-fM4vqba4E/v-deo.html
Bad pid loop, going to invalid values . Not enought clamping for intégral term. Big software bug.
I think it's the shape of the ducks themselves, I don't have one so I don't know but maybe you room remove the ducks and test it
Justin Davies has a video explaining why the AVATA has the issue, he stated that it’s from the big ducts and flight controller having a bit of a spaz under high loads at speed. Not sure if you’ve seen it yet.
Now they are saying it's a software issue.
Couldn't have said it better myself... No really I couldn't.
Hopefully dji will come up with a software update that should fix the problem with it, because otherwise it is a ripper of a quad.
Here in Brazil there is a popular saying, which is more or less that and says "killed the snake and showed the dick". And Bruce, that's exactly what you did with this Avata problem, you identified the cause and showed the solution.
I think you are all just too harsh on Avata. I was flying it yesterday in manual and flies really great. I normaly fly my ImpulseRc APEX HD 5” on Kiss. It probably has some issue but its showing by missuse. Its cinematic cinewhoop. Its for nice flow videos, not for racing turns. I have seen most of avata crashes and those pilots just dont fly it like that. They turn quickly not tilting inside curve, just ugly fast flat turn or tilt it to outside turn.
Flight controller crash in flight is unacceptable, flight controller crash that leads to motors continuing to spin is a reason to ground all airframes until the bugs are fixed AND the reason why such failure passed to the production code is identified.
Yeah if you just fly it slow and smooth as you should you are fine. That's why I rarely fly it in manual. That's also maybe a reason they go with the motion controller. If it is a fc crash then I hope they fix it. This drone is just not a acro drone. People seem to not understand.
@@HannesDev Flight Controller must not crash in a way that leaves vehicle unsaved under any circumstances. This includes any abnormal situations that might theoretically happen.
I don't think your yaw turbulence explanation is correct. The linear yaw rotation speed is less than a regular forward flight linear speed. The turbulences are not that signifiant and are less than in forward flight (accounting also for forward flight angle). The drop on big yaw movement is due the the FC software (inherent limitations + bug) and how PID loops and I integration works + the fact that same motors must correct both yaw, pitch and roll angles + the fact that motor thrust is limited, not infinite.
Plus, the fact that Avata does not recover on such instabilities and ends up going inverted, is also purely a FC software issue/bug. Any decent FC software (ArduPilot, BF) can recover from impossible positions (hint: check out ArduPilot throw mode).
It's a DJI product. That means I won't waste my money on one.
I think that sounds like a software bug maybe they are just tracking 180 degrees and lazy coder goofed
There only been one vortex that I have been affected by, since flying FPV. That's the money-sucking one.
Yeah not easy to filter that out. But ironically whoops are the solution for that haha.
@@slowentropy4531 I foolishly set the 'I-gain" to a high value, hoping that it meant "income gain", but it didn't work...
@@clifffield1 ah yes well it was worth a try mate, as long as the wallet didn't have a fly away.
@@slowentropy4531 Sadly it was more like a failure to launch. Haha.
@@clifffield1 hahaha, it's all worth it if you're crazy enough to pursue our crazy lil hobby haha.
Lol "...I'm headed back to China."
AVATA IS TRAAAAAAASH
Lucky it doesn't have folding arms. Beta stick to Beta-flight and stuff that has been developed by enthusiasts not by Bachelors of Com Sci.
Pilot Error........... Most of these guys are trying to fly the Avata like a Freestyle god in a manner the quad was never designed to fly. The Avata is made for slow close proximity flying mainly indoors and around actors. It is not made for high speed dynamic chasing or doing tricks. What is happening is the exterior forces are exceeding the FC capability to react to them. Thus the quad falls into an unrecoverable death roll...Custom builds used to have the same problem when we were overloading the Processor on our FC's running 8k/8k on an F3 board the FC would get overloaded and shut the whole thing down putting you into the death roll.....DJI marketing makes it seam like anyone can just take one of their products of the shelf and get amazing cinematic quality footage.. That is far from the truth.
Look at DJI's marketing videos though, of course users want to fly like that.
@@FilipRadelic That's the main problem with DJI they make it seam like anyone can grab one of their products off the shelf and get stellar shots with no skill at all like someone who has been flying for years. That is a flat out lie. Fact is certain crafts are made for certain purposes. Cinewhoops are not made for freestyle/chase flight they are underpowered and top heavy and are made for very slow close proximity indoor or around actors. Freestyle /chase quads are much more powerful with a minimum 10:1 power to weight ratio and do 120mph in less than 1 sec. Not even the DJI FPV drone is capable of that.