My P3P crashed, DJI not cooperating

[GadgetGuy]

WTF? He had strong signal for the entire episode.

Regards, Jim

Maybe you missed the part that it was the dumb Smart Return Home that activated during his hand catching,which ultimately led to the crash, and his attempt to regain control of an apparent flyaway, by moving the sticks, cancelled the ascension part of the Smart Go Home RTH, so it couldn't clear the rocks.

 

[Chris_P]

Maybe you missed the part that it was the dumb Smart Return Home that activated during his hand catching,which ultimately led to the crash, and his attempt to regain control of an apparent flyaway, by moving the sticks, cancelled the ascension part of the Smart Go Home RTH, so it couldn't clear the rocks.

My understanding is that even my attempts to regain control didn't much matter because the the RTH engaged as I was landing, I was applying negative throttle while landing and that automatically cancelled the RTH altitude rise to my preset level of 100'. It then flew off at what it believed to be approx 54' which was in fact approx 15' and hit the rocks.

 

[Diehard]

Thanks, I wasn't aware of the 20m activity when hitting the throttle. Can you tell throttle position at all during the flight from the logs?

 

[Diehard]

On the DJI app flight record click on the flight in question and when it pops up look at the bottom left hand side and you will see an icon of an RC. Click on that and a visual of the control sticks will pop up. Press the play button and you will then see every movement that was made throughout that whole flight.

 

[GadgetGuy]

My understanding is that even my attempts to regain control didn't much matter because the the RTH engaged as I was landing, I was applying negative throttle while landing and that automatically cancelled the RTH altitude rise to my preset level of 100'. It then flew off at what it believed to be approx 54' which was in fact approx 15' and hit the rocks.

The best thing you could have done, and should do in the future, is turn OFF the dumb Smart Return Home feature in the app. If you are already properly monitoring your battery level and your remaining flight time and the flight time needed to return to the Home Point, as you were, it is at best an annoyance, at at worst, can be the precipitating cause of a crash, even if not the proximate cause! in this case, any stick initiated elevation change, after the SRH activated, cancelled the Ascension. Applying elevation manually, or cancelling the Smart Return Home, after it activated, using a single press on the RTH button on the remote, in a timely manner, would have prevented the crash. However, Smart Return Home led to the crash. Turn it OFF!

 

[flyNfrank]

The best thing you could have done, and should do in the future, is turn OFF the dumb Smart Return Home feature in the app. If you are already properly monitoring your battery level and your remaining flight time and the flight time needed to return to the Home Point, as you were, it is at best an annoyance, at at worst, can be the precipitating cause of a crash, even if not the proximate cause! in this case, any stick initiated elevation change, after the SRH activated, cancelled the Ascension. Applying elevation manually, or cancelling the Smart Return Home, after it activated, using a single press on the RTH button on the remote, in a timely manner, would have prevented the crash. However, Smart Return Home led to the crash. Turn it OFF!

You know back before the P3 when it was all P2V+ we always mentioned how nice it would be to have a feature such as RTH with the ability to cancel at our own request. It's so easy to miss the a/c's safe to return home point and in my opinion the smart RTH is a wise feature to have. The only real downfall here was not cancelling the RTH in this particular situation. It's might be better to educate others on disarming the call rather then advising to turn it off.

 

[henick]

Yeah turning it off just causes losses in another already resolved method. It's a great feature in principle I just honestly rarely use it. I bet the day I need it I'm going to be pretty glad it's there.

There are many situations proven where it should ask what to do in a more intelligent or dynamic manner though. Going behind a building is a big culprit if height isn't considered. Walking on a sidewalk under trees is another if unexpected and it raises into them. Then the trying to RTH with a bad compass causes all kinds of nasty. Newbies are obviously more prone to make the mistakes but the common ones are probably more common than you think...

In theory it's awesome though.

 

[Meta4]

The main take home lesson is that out over water is an unforgiving environment for flying.
The pilot has to be completely familiar with operating the Phantom and aware of all aspects of the aircraft status at all times.
This includes such things as home point location, accuracy of altitude readings, RTH status, battery levels, wind conditions, obstacle heights, any warnings in the app etc etc.
As well as the usual hazards that are normally present, this flight was complicated by having nearby obstacles (trees and shoreline).
In either situation, one thing goes wrong and your Phantom can be out of action.
Combining over water flight and nearby obstacles means that the pilot has to be doubly sure of all aspects of the flight operation.

 

[Chris_P]

Did your video record settings in the app also record an .SRT sidecar file, along with the video, that you can make available?

I assume it would be on the microSD card in the same directory as the video? If so, I just checked and i don't see any files with the extension .SRT

 

[impilot51]

If you don't have the video caption enabled in the video options section, that file won't get created.


Sent from my iPhone using PhantomPilots mobile app

 

[GadgetGuy]

After well over 300 flights, with Smart Return Home turned off, I have never missed it, nor ever needed it. I fly over water all the time from great distance. Only time I had Smart Return Home on was when I first got my P3P, before I learned how to disable it, and then later, when a firmware update turned it back on. I personally find it extremely annoying. The app already clearly tells you at all times whether you are ever farther away than the remaining battery capacity, based upon a very conservative calculation. If I turned around every time Smart Return Home would activate, my flights would all be at least 5 minutes shorter, and every minute counts, when you have a 6 mile round trip to reach your subject! The two battery settings for Low battery and Critically Low battery already warn the pilot in plenty of time to land if they are flying within VLOS. If you are flying beyond VLOS, you better know what you are doing, and depending upon Smart Return Home to remind you, seems to me like a fool's folly. It's one of these settings that, in my opinion, is too "smart" for its own good. YMMV.

 

[Thho]

After well over 300 flights, with Smart Return Home turned off, I have never missed it, nor ever needed it. I fly over water all the time from great distance. Only time I had Smart Return Home on was when I first got my P3P, before I learned how to disable it, and then later, when a firmware update turned it back on. I personally find it extremely annoying. The app already clearly tells you at all times whether you are ever farther away than the remaining battery capacity, based upon a very conservative calculation. If I turned around every time Smart Return Home would activate, my flights would all be at least 5 minutes shorter, and every minute counts, when you have a 6 mile round trip to reach your subject! The two battery settings for Low battery and Critically Low battery already warn the pilot in plenty of time to land if they are flying within VLOS. If you are flying beyond VLOS, you better know what you are doing, and depending upon Smart Return Home to remind you, seems to me like a fool's folly. It's one of these settings that, in my opinion, is too "smart" for its own good. YMMV.

I am a private helicopter pilot since 1985. As far as I am concerned there is no barometric height, it is only pressure altitude. Pressure is changed with temperature. My theory is, there is nothing wrong with the barometric device inside the Phantom. It is just influenced by the rising temp inside the Phantom shell. If you let the quad settle for 3 minutes or so before liftoff, the temp inside has likely increased. At a point (probably longer than 3 minutes), there will beequilibrium and that is the proper time to lift off, because then the home point height is calculated to 0 ie "the reference pressure". The pressure needs to be measured inside the body, because otherwise the reading will be faulty due to the movement and speed of the quad. So IMHO, there is very little to state towards DJI and this is just plain physics. Of course the SRTH which uses the pressure to calculate height is very unsecure and the only way to do it right is to use the same technique as all aircraft pilots do when landing; Get the proper QNH (corrected pressure at sea level) from the tower, adjust your altimeter with this value and you will land at the right indicated altitude ASL (above sea level). In order to use this technique, the RC must have a barometric device inside (not affected by heat) and send this value (QNH) to the drone before kicking in the SRTH. This works of course only when the home point is at the same position as the RC. Resetting home point manually is not going to work with this technique. To be able to use accurate height measurement, you need a microwave doppler radar as bigger aircraft use (or ultrasonic doppler with very high power), as GPS height indication is very insecure. I have no ideas how to make a secure SRTH when you update the quad home point manually during the flight... The ordinary RTH will probably work because you cannot alter throttle stick when out of radio contact!

 

[GadgetGuy]

I am a private helicopter pilot since 1985. As far as I am concerned there is no barometric height, it is only pressure altitude. Pressure is changed with temperature. My theory is, there is nothing wrong with the barometric device inside the Phantom. It is just influenced by the rising temp inside the Phantom shell. If you let the quad settle for 3 minutes or so before liftoff, the temp inside has likely increased. At a point (probably longer than 3 minutes), there will beequilibrium and that is the proper time to lift off, because then the home point height is calculated to 0 ie "the reference pressure". The pressure needs to be measured inside the body, because otherwise the reading will be faulty due to the movement and speed of the quad. So IMHO, there is very little to state towards DJI and this is just plain physics. Of course the SRTH which uses the pressure to calculate height is very unsecure and the only way to do it right is to use the same technique as all aircraft pilots do when landing; Get the proper QNH (corrected pressure at sea level) from the tower, adjust your altimeter with this value and you will land at the right indicated altitude ASL (above sea level). In order to use this technique, the RC must have a barometric device inside (not affected by heat) and send this value (QNH) to the drone before kicking in the SRTH. This works of course only when the home point is at the same position as the RC. Resetting home point manually is not going to work with this technique. To be able to use accurate height measurement, you need a microwave doppler radar as bigger aircraft use (or ultrasonic doppler with very high power), as GPS height indication is very insecure. I have no ideas how to make a secure SRTH when you update the quad home point manually during the flight... The ordinary RTH will probably work because you cannot alter throttle stick when out of radio contact!

Thanks for the input. My takeaways from your post are that the most accurate barometer calibration requires a stable operating temperature before lift off, which likely requires the motors running and the props spinning, and hence a second battery, especially if it requires more than 3 minutes, as you suggest, to maximize flight time by starting the battery at 100%. Internal operating temperature will remain the same during the battery swap out. There may also be something to flyNfrank's earlier post that in warmer weather, the operating temperature will be achieved faster, requiring less barometer calibration time. Smart Return Home leaves the pilot with full override capability, unlike the Failsafe RTH, where you have no control, until the Failsafe RTH aircraft movement restores signal and control, which is really its primary purpose in normal flight, so you can then cancel it and move on. As I stated earlier, the very new flyers who will benefit most from Smart Return Home, are the ones who are already flying within VLOS, and for those who are flying beyond VLOS, or are aware of their app telemetry, it is at best an annoyance, and at worst, can lead to a crash, as it did here for the OP. The OP should turn Smart Return Home OFF! It's only "smart" if you are "dumb"! The OP was responsibly handling the very issue Smart Return Home was designed to solve (landing at his position before the battery dies), and like an airbag designed to save your life if you aren't wearing your seatbelt, it can also possibly kill you if it activates inappropriately while wearing your seatbelt. He still had 31% remaining battery while trying to handcatch and land, when the Smart Return Home activated during his attempted landing! Dumb Smart Return Home!

 

[Chris_P]

Its been over 4 weeks since my crash and after several phone calls and pages and pages of text in their RMA log, DJI has taken the final position that their 20% offer stands .....

Here is their ridiculous reasoning.... They have concluded that the Phantom crashed due to the RTH home engaging after a low battery (30%). The Phantom impacted an object as it was returning home. They have completely disregarded all the evidence I provided regarding the faulty barometer and the incorrect altitude readings. They claim that from the logs they only see a RTH heading then a crash. The altitude is irrelevant. They indicated that although the DAT files contain additional data that could assist, they are disregarding that as corrupt info since the Phantom went briefly in the water. I get that info recorded after impact may be from damaged sensors therefore considered corrupt but certainly the info recorded before the impact can be used. Its a stupid line of reasoning. The also claim that the additional logs that I supplied after taking the Phantom back to the area are useless as they are unreliable coming from a damaged aircraft therefore unusable. I have to say that these reasons are beyond ridiculous. They are clearly doing everything they can to avoid admitting some fault relating to the altitude readings. I will not let this rest until the entire world is aware of their unprofessional attitude when dealing with repair issues . I will be starting a Facebook and Twitter campaign to educate prospective buyers on DJI's underhanded customer service tactics. Buyer Beware !!!

 

[impilot51]

Well, I can't say that I am surprised. While there is, in my opinion, clearly a barometer that gave false information, it is also pretty clear that this was entirely avoidable through an understanding of how RTH works and thus disabling it due to the type of flight you were doing (over water from a boat with home point moving throughout the flight). I think that their offer of 20% is their acknowledgement that the altimeter was a problem. The other 80% is, in their opinion, your responsibility due to pilot error. And there were numerous errors here on your part.

 

[syotr]

I expect that they still make money selling one at 20% off. That's about what I would guess they sell to dealers for. So every crash is like a new sale, great for business.

 

[Reed L]

Well there selling them for $709 and 20% off brings it down to $570, so no complaining is warranted.

Its been over 4 weeks since my crash and after several phone calls and pages and pages of text in their RMA log, DJI has taken the final position that their 20% offer stands .....

Here is their ridiculous reasoning.... They have concluded that the Phantom crashed due to the RTH home engaging after a low battery (30%). The Phantom impacted an object as it was returning home. They have completely disregarded all the evidence I provided regarding the faulty barometer and the incorrect altitude readings. They claim that from the logs they only see a RTH heading then a crash. The altitude is irrelevant. They indicated that although the DAT files contain additional data that could assist, they are disregarding that as corrupt info since the Phantom went briefly in the water. I get that info recorded after impact may be from damaged sensors therefore considered corrupt but certainly the info recorded before the impact can be used. Its a stupid line of reasoning. The also claim that the additional logs that I supplied after taking the Phantom back to the area are useless as they are unreliable coming from a damaged aircraft therefore unusable. I have to say that these reasons are beyond ridiculous. They are clearly doing everything they can to avoid admitting some fault relating to the altitude readings. I will not let this rest until the entire world is aware of their unprofessional attitude when dealing with repair issues . I will be starting a Facebook and Twitter campaign to educate prospective buyers on DJI's underhanded customer service tactics. Buyer Beware !!!

Honestly, it's not ridiculous, the crash was user error all the way. You woulda/ coulda/ shoulda done a lot of things different, I have thought from day 1 that 20% was fair being that the crash was 100% your fault. Plain and simple, you just were not familiar enough with the controls. Heck I've crashed twice on my own and never expected to try and blame it on DJI or have them pay for anything. You should be happy with the 20% because the reality is... You shouldn't have asked them to pay for your mistakes in the first place.

 

[Chris_P]

Well there selling them for $709 and 20% off brings it down to $570, so no complaining is warranted.


Honestly, it's not ridiculous, the crash was user error all the way. You woulda/ coulda/ shoulda done a lot of things different, I have thought from day 1 that 20% was fair being that the crash was 100% your fault. Plain and simple, you just were not familiar enough with the controls. Heck I've crashed twice on my own and never expected to try and blame it on DJI or have them pay for anything. You should be happy with the 20% because the reality is... You shouldn't have asked them to pay for your mistakes in the first place.

Did u miss the entire part of this thread where it's abundantly clear that the Phantom calculated the wrong altitude? The reported altitude at the time of impact was off by over 40 feet. This was certainly a contributing factor in the crash. As I've said before, I did make mistakes, however there is also an element of equipment failure in this crash. Had the phantom calculated the proper altitude it would have never hit that rock.

DJI is refusing to acknowledge any issue with the barometer sensors. Others on the forum have conducted independent testing and have confirmed incorrect altitude measurements after a few minutes of flight.

 

[GimbleOut]

Awesome thread about the altitude readings ...and thanks to you folks for digging deep to help us learn.


Sent from my iPhone using PhantomPilots mobile app

 

[msinger]

Did u miss the entire part of this thread where it's abundantly clear that the Phantom calculated the wrong altitude? The reported altitude at the time of impact was off by over 40 feet. This was certainly a contributing factor in the crash.

Here are some other things that contributed to the crash:

• You did not disable Smart RTH prior to takeoff
• You did not cancel RTH when you were prompted to do so
  
• You moved the throttle stick while your Phantom was trying to ascend to your set RTH altitude
• After your Phantom started returning home, you gave it full throttle down (instead of up) and descended toward the rocks