Bummed. New P3A Crashed. Pls Advise

[TheRealNick]

McVentures,

I really do not this was your fault. I have been lurking for awhile as I consider a Phantom purchase and this story lead me to sign up. Below are the reasons why I believe it was not your fault, assuming the crash was related to a prop coming off.

1. I have watched almost every video and review of setup and first flight and it doesn't look like any of them do much as far as prop tightening goes. And they seem to have no issue...

2. Having the motor spin in a way that tightens the props is by design. I have big doubts that there is any active braking, if so and if it was severe the prop attachment method seems like a defective design. Furthermore if it was present it seems like DJI would put a lot of emphasis on tightening the props. I also think the air resistance probably is enough to slow the props down that active braking is not needed. All that being said, I am not a complete expert and maybe it does have active braking.

3. Even if it did have active braking, you were not flying aggressive, sounds like barely hovering, seems like there would not have been active braking just before your fall.

4. If it had active braking you would see it in the log file of the motor that lost the prop. You would see a very rapid drop in RPM caused by active braking followed by a rapid increase in RPM after the prop came off. If someone can post each motors speed over the flight I would love to see it.

5. Are we sure the RPM is actual and not target or calculated? Do the motors have at least 3 wires?

6. If would be cool if Steve posts which prop he believes came off from the log files to confirm it is the same one that actually did.

7. If the prop hit something I believe it would only have tightened it.

Don't give up and follow DJI support. I have not purchase yet as I have fear of technical issue and want to see how DJI handles it...

 

[TheRealNick]

Also, just reading back it was not your first flight, but your fourth flight. To me this also makes it less likely that your prop was too loose. It seems like a loose prop would fail on flight one or not at all, each flight would likely tighten the props.

A question to other owners is do you notice if the prop is tighter when you remove it after flight or if you put it in and take it off? If it is tighter after flight then that would confirm that flying tightens the props more than most people install them. I suspect this is the case...

 

[Meta4]

As the prop that was missing earlier has no damages to threads, it means it got separated during flight. And that can happen if motor suddenly stops. If this is correct than its engine failure. Am I correct?

BUT ...

Here is a picture of the prop. I can only see a sliver of plastic out of place. The lower threads appear fine.

That torn thread is not something you see in a normal prop.
I suspect there has been a traumatic impact to do that.

The arms have taken a big hit and there may be IMU damage due to this as well.
Do we know if the detached prop came from one of the damaged arms?
Was there any damage to the blades or leading edges of the loose prop?
If so, it looks even more like it was impact that detached it.

There's incomplete evidence causing a lot of guesswork in this thread.
At least some of the ideas put forward here are wrong.
The only way to be sure is to have DJI look at that dat file.
Upload it to dropbox and contact DJI's online tech support again.
Have them look at it and get back to you after they analyse it.

 

[Meta4]

I have big doubts that there is any active braking, if so and if it was severe the prop attachment method seems like a defective design.

The Inspire and P3 both have active braking and because a small number of early Inspire owners had lost prop issues DJI Propulsion came up with the composite hub solution for the P3.

 

[TheRealNick]

Well I am hoping the torque during active braking is not too high....this gives me an uneasy feeling....critical torque spec with no torque wrench....

A possible test for this would be to install the props barely finger tight. Have someone do a hand launch but not let go and then as it tries to hover rock it around to get each motor to react and see if the props come... I still do not think it is pilot error....

 

[Meta4]

Well I am hoping the torque during active braking is not too high....this gives me an uneasy feeling....critical torque spec with no torque wrench....

There are hundreds of thousands of P3s flying and even in the often conspiracy favouring forums, reports of lost props are rare enough that this isn't considered an issue.
Most owners only set their props finger tight but the grippyy threads in the composite hubs seem to have prevented any problems.

 

[McVenturesRC]

McVentures,

I really do not this was your fault. I have been lurking for awhile as I consider a Phantom purchase and this story lead me to sign up. Below are the reasons why I believe it was not your fault, assuming the crash was related to a prop coming off.

1. I have watched almost every video and review of setup and first flight and it doesn't look like any of them do much as far as prop tightening goes. And they seem to have no issue...

2. Having the motor spin in a way that tightens the props is by design. I have big doubts that there is any active braking, if so and if it was severe the prop attachment method seems like a defective design. Furthermore if it was present it seems like DJI would put a lot of emphasis on tightening the props. I also think the air resistance probably is enough to slow the props down that active braking is not needed. All that being said, I am not a complete expert and maybe it does have active braking.

3. Even if it did have active braking, you were not flying aggressive, sounds like barely hovering, seems like there would not have been active braking just before your fall.

4. If it had active braking you would see it in the log file of the motor that lost the prop. You would see a very rapid drop in RPM caused by active braking followed by a rapid increase in RPM after the prop came off. If someone can post each motors speed over the flight I would love to see it.

5. Are we sure the RPM is actual and not target or calculated? Do the motors have at least 3 wires?

6. If would be cool if Steve posts which prop he believes came off from the log files to confirm it is the same one that actually did.

7. If the prop hit something I believe it would only have tightened it.

Don't give up and follow DJI support. I have not purchase yet as I have fear of technical issue and want to see how DJI handles it...

So, my flight log is FLY013.DAT on flylog.info
As far as DJI goes, this is Day Two since posting a technical request with them. I have not received an RMA yet. I have requested it twice.

 

[McVenturesRC]

Flight log info:
FLY013.DAT on
flylog.info
for my crash

 

[bobmyers]

I always have a great day. You give yourself far too much credit!

Ignore

 

[III% Streve]

McVentures,

I really do not this was your fault. I have been lurking for awhile as I consider a Phantom purchase and this story lead me to sign up. Below are the reasons why I believe it was not your fault, assuming the crash was related to a prop coming off.

1. I have watched almost every video and review of setup and first flight and it doesn't look like any of them do much as far as prop tightening goes. And they seem to have no issue...

2. Having the motor spin in a way that tightens the props is by design. I have big doubts that there is any active braking, if so and if it was severe the prop attachment method seems like a defective design. Furthermore if it was present it seems like DJI would put a lot of emphasis on tightening the props. I also think the air resistance probably is enough to slow the props down that active braking is not needed. All that being said, I am not a complete expert and maybe it does have active braking.

3. Even if it did have active braking, you were not flying aggressive, sounds like barely hovering, seems like there would not have been active braking just before your fall.

4. If it had active braking you would see it in the log file of the motor that lost the prop. You would see a very rapid drop in RPM caused by active braking followed by a rapid increase in RPM after the prop came off. If someone can post each motors speed over the flight I would love to see it.

5. Are we sure the RPM is actual and not target or calculated? Do the motors have at least 3 wires?

6. If would be cool if Steve posts which prop he believes came off from the log files to confirm it is the same one that actually did.

7. If the prop hit something I believe it would only have tightened it.

Don't give up and follow DJI support. I have not purchase yet as I have fear of technical issue and want to see how DJI handles it...

Some good deduction considering you dont own a P3. Ill address a few things that havent been already

#3.. Its a bit hard to tell exactly from a HealthyDrones report, but active breaking happens most aggressively at the end of a climb, or when the stick is released during lateral movement. Especially at the top of a quick climb, the props slow a LOT. Sounds almost like they go to a flat pitch they are turning so much slower.

#4 on.. We don't have a log file yet that we can examine. But motor RPM and temperatures are things we don't get to see. That data is in the log file ON the P3, and so far, I dont think anyone has unlocked the mystery of displaying that data. The logs in the tablet/phone are very detailed, and would help a lot. But not exactly what you are looking for.

I would love to be able to have guessed what prop flew off. But I will very openly say that I have no way to do that.

#7 Absolutely. But it can be hypothetically tightened to a degree that it strips the threads and flys off. Not the case here as the threads are still perfect.

 

[McVenturesRC]

BUT ...

That torn thread is not something you see in a normal prop.
I suspect there has been a traumatic impact to do that.

The arms have taken a big hit and there may be IMU damage due to this as well.
Do we know if the detached prop came from one of the damaged arms?
Was there any damage to the blades or leading edges of the loose prop?
If so, it looks even more like it was impact that detached it.

There's incomplete evidence causing a lot of guesswork in this thread.
At least some of the ideas put forward here are wrong.
The only way to be sure is to have DJI look at that dat file.
Upload it to dropbox and contact DJI's online tech support again.
Have them look at it and get back to you after they analyse it.

Thank you for the suggestion to context tech support again with the .DAT log. I will today.

Regarding the prop. No, the detached prop did not come from one of the damaged arms and the detached prop was clean as a whistle. No smudges, gouges, nothing.

 

[III% Streve]

Just for giggles I tested the tightening process of the blades by the motors alone. I was able to get 3 different props to come off, and 1 came REALLY close, but letting go of the motor spun that prop back on before it totally came off. Ill try to do a video later. Bruised the tip of my damnit finger yesterday, and broke a prop. So I gave up before there was real carnage!

 

[McVenturesRC]

I have downloaded the .DAT file. But I've tried many times to load it here. It appears to big. How can I get this to those of you interested in looking at it?

 

[III% Streve]

You could put it on dropbox and from there the site will give you a link to share the files with people. Thats the best way to get it to DJI also

 

[McVenturesRC]

Dropbox - FLY013.csv

 

[BudWalker]

I looked at the .DAT some this morning. @McVenturesRC had uploaded it to the flylog.info site and I retrieved it from there. I used a new tool, called DatConverter, that I had made available on the website yesterday. It's an offline version of the website converter. The .DAT has more data than can be plotted by Excel. I used DatConverter to produce 2 .csv files; one for the preflight where a compass calibration was done, and another for the flight itself. This post addresses the flight. The preflight was actually kinda interesting; I'll address that in another post.

The battery was on and providing power for about 180 seconds. It was at roughly 133 seconds when the P3 was launched. When the recording stopped the P3 was still in the air. I'll speculate that the battery disconnected, causing the recording to stop. However, there were no corruptions or partial records that often happens when the battery is removed while still providing power. In fact, this .DAT didn't have any partial records or data corruptions.

The .DAT shows the P3 with a heading of 198 degrees for the entire flight. Is this right? There is also a problem with the way the altitude is presented. I'm looking into that.

You should definitely get DJI to look at this .DAT file. I'd appreciate knowing anything they tell you.

I've included some plots that I made with Excel. First are the X and Y accelerometer data

The horizontal axis is labelled with the clock tick#. The clock runs at 600 HZ so this plot, starting at tick# 80000, coincides with 133 seconds. The event at 81000 is the take off and the events at around 105000 are the maneuvers at the end of the flight.

The barometer is shown here

Note, that the recording terminates with the P3 still at altitude indicating a loss of power in flight.

The battery data is shown here

 

[TheRealNick]

That is awesome Bud! It is weird that the website seems to show data at the end of flight at lower altitude though, where do you think it is getting that from? Can you see motor speeds?

 

[JoBe]

What was happening was that people were not fully seating their batteries after they heard he first click, so during flight they would slide out a fraction. It was just enough to lose main power but not enough to lose data recording.

Sorry for maybe an ignorant first post, but are you saying that the data power and the main power come from different batteries?

 

[III% Streve]

Interesting data. But none of it matches anything from HealthyDrones data we saw early on. Not saying your tool is the problem. Only pointing out the discrepancy.

 

[III% Streve]

Sorry for maybe an ignorant first post, but are you saying that the data power and the main power come from different batteries?

No.. Im guessing he means that the connection is so good that its enough to maintain procession data, but not good enough to power the speed controllers. Like screwing in a lightbulb when the switch is on... there is some arcing, and flickers of light, but not a steady light till its screwed in tighter