FlightLog discussion of “howfastyes”’s TH crash

howfastyes · Mar 30, 2020

EDIT:

I am opening this to have discussion just on FlightLog analysis. I am starting with @h-elsner & @Ty Pilot posts of their FlightLog review. DoomMeister

Very sad day in my life. I lost my friend, after 2 years. In this crazy times i went very far away from town to fly my Typhoon H. Weather was perfect. Perfect day.

Middle of nowhere this happened:

Typhoon H is now officialy dead.

Brokent parts:
Left landing gear
3x arms with mounts
top and bottom cover, maybe GPS board i don't even care at this point ;(
main frame ;(
probably gimbal ;(

Flight 66:
Battery charged 100%, regular start and little circle at low att and low speed, landing, check and start again. (see picture)
Then i took straight path, turned after 450m and went back. Everything was good to this moment. Recorded this 5 minutes of flight on camera and stopped recording to prepare next shot. It was about 5 minutes of flight.

Before i've done anything i've recived warning Motor Failure and i could clearly see weird aircraft behavior, obviously one of motors shut down.
Immediately tried to lower altitude and bring the aircraft close to me, after few seconds (maybe 20-30m) i've recived a GPS lost Warning!! Low Batterry level 2 warning !! Land immediately !!
At this moment H started to ascend very fast and changed direction as you see on screenshot, i had no controll AT ALL, tried to pull it down but nothing works.
Few second later Typhoon hits big trees and it's dead. There was no time to do anything, i've never seen this crazy ****.

All 6 motors working ;(
I can buy arms and covers but i can't buy Main Frame (one of arms ripped off a piece of main frame).
Second issue is gimbal, video was working but gimbal is lookin' bad i don't even want to check

1 month of warranty left but i can't find the invoice ;(

I've learned how to fly with this aircraft and got my official license passed on this drone, many flights without GPS, never used Smart modes and **** like this, and this happend to me ;(

I almost cried, it was 2 days ago, today i've pulled flight data to check why my Typhoon commited suicide and i think it was some kind of solar blast or virus... joke

Really i have no money but i will buy this drone again as soon as i got some cash - but i want to know what happend. WHY?
I've done crazy things with this aircraft, it supposed to be one of my safest flights ever! No power lines, no wi-fi, nothing. WHY?

I want someone with experience to check my Flight 66 data and tell me what happened.
Data shows me some crazy thing like Max Speed 130km/h or Compass calibration warning with Battery Low level 2 warning at the same time (after 5-6 minutes of flight).

Flight log 66 in attachment, please help me solve the mystery of this suicidal fly away.

h-elsner · Mar 30, 2020

The battery is poor or too cold or both. It the voltage breaks down to 15,3V (~46% remaining capacity) in the first 40sec of the first flight.

The same at the second flight, starting at 13:54:30. Around 14:01 battery starts to die during roll/yaw action. Low voltage warnings start at 14:01:12 after only ~7min flight time.
Then during low battery we got a massive set of Compass Calibration Warning over 17sec.
For some reason the FC initiates RTH landing for <1sec but the drone still ascent (I think this should be the beginning of a low battery fail save or a try for that).
RTH followed by loss of GPS support (Flight mode 4) due to missing GPS module (IMU status and GPS status indicates that). Speed accelerates, it keeps the momentum from the acceleration before because GPS support is absent.

The the drone hits the tree. Pitch shows the impact. 1sec later motor 4 identified as blocked and the FC cut all motors and send Emergency (12).
Note: Excessive high speed at the end is caused by crash, measurement is disturbed when it hits the trees. The same for other sensor values.

Tomorrow I will go on and check the stick inputs.

br HE

Ty Pilot · Mar 30, 2020

I was just getting ready to post my thoughts but I see @h-elsner has already posted his but just for reference here is what I found - pretty much the same thing.

The flight data pretty much matches your memory and even though the first part of the flight looked and felt normal, I am seeing things that are not right, mainly the IMU status is switching from 'All OK' to a status where it is either not reading/using the Compass or GPS data or at times both and this begins early in the flight.

Here is the timeline

13:49:11 - Boot up - IMU status = 225 which means it is not yet ready
13:51:02 - Motor Start
13:51:05 - Takeoff - Voltage is 16.6
13:51:50 - IMU status finally is AOK
13:53:21 - Flight Mode status goes from 'Angle' to 'Acquiring' at the end of the circle you first made
13:54:30 - Flight Mode = Angle but simultaneously IMU status losses the compass and GPS
13:55:23 - IMU status = AOK
13:57:50 - Max Distance - Returning
14:01:12 - Low Batt 1 + Compass Warning
14:01:29 - Low Batt 2 + Compass Warning
14:01:33 - Flight Mode switched to RTH
14:01:34 - Flight Mode back to Angle but without GPS
14:01:40 - Motor 4 OFF + Motors Killed almost simultaneously = Impact

My take on the crash (and this is just my opinion), is that it is possible that the IMU was in a failing state before lift-off. I do not know how the different status's of the IMU plays into the messaging seen on the ST-16 screen such that maybe; everything looked good, but for sure the AOK status was coming and going throughout the flight.

I also do not know if the rapid decrease in voltage towards the end had anything to do with the crash but obviously it did not help.

From Low Batt 1 to the crash was 28 seconds. Your recounting of the ascent lines up with the RTH being turned on and then off - (you may not have known you did it but in these moments we can do things like this). Obviously the RTH began to work as it is normal to first climb to the RTH altitude but without GPS it is likely this would not have worked.

DoomMeister · Mar 30, 2020

I will echo the results of both @h-elsner and @Ty Pilot. I would also add that the 2.4GHz RC link seemed very weak. There were way too many 0’s in the fsk_Rssi column indicating a switch to the 5.8GHz redundancy for control.

IMU Status - 97, missing controller GPS and AC compass

IMU Status - 225, missing AC compass

IMU Status - 229, all good

The single largest problem I see here is the weakness of the battery. I don’t think temperature was a factor as I believe it was in the fifties and sixties ( Fahrenheit) that day. The way the voltage dipped right after takeoff with both flights I would expect to find a high IR on at least one cell of the battery if not more.

The only other time I have seen batteries act like that when it wasn’t the battery at fault was when motors needed replacing from high number of flight hours (400+).

Mrgs1 · Mar 31, 2020

Was the OP using the original battery charger supplied for the Two years or a 3rd party charger? I have found the Yuneec to be very hit and miss, and a contributor to battery problems.

h-elsner · Mar 31, 2020

I agree with @DoomMeister, the battery seems to be the root cause of all the problems. All works well untill the voltage drops down at the end of the second flight (3). The battery is weak that's proof with (1) and (2).

Let's check the last 30sec. During this time all the strange things appear until the drone hits the tree. This is the part of the second flight, we are talking about.

With the turn in the red circle the voltage drop-down increases. Then Compass Calibration Warnings (CCW) appear, the longest over 17sec. The strange intermission of flight mode 14 is during a CCW block, followed by switch off GPS support. GPS flags in IMU status and GPS status falls to 0 but the GPS in Pressure-compass -status persists, means HW is OK and we still see plausible coordinates. Only the FC rejects the GPS. Why? I don't know. But my thoughs go to the power supply. There may drop peaks that we not see in the voltage measurement samples. Those transients may disturb the FC.

After GPS lost the drone will fly with its momentum and possibly with the wind. After some maybe erratic flying during CCW the drone still reacts to the stick inputs but probably slower than expected due to the missing power. The battry is very low at this time. However, the drone tries bravely as we see in pitch chart.

It is now ~200m away from the pilot and has no GPS support. I think @howfastyes is a good pilot when I see how he mixes yaw and roll to fly a turn to have a nice looking flight. But at this point he is not aware that he loss GPS support and the drone moves currently away from him and accelerate (pitch give between the both red lines). After that he tries a lot, mainly with pitch but he got only 5sec left before a jumping tree (that we all know) catches the drone.

Attached all the pic and the last 30sec cut from the flight log for analysis.

br HE

Photo · Mar 31, 2020

Ty Pilot said:
I was just getting ready to post my thoughts but I see @h-elsner has already posted his but just for reference here is what I found - pretty much the same thing.

The flight data pretty much matches your memory and even though the first part of the flight looked and felt normal, I am seeing things that are not right, mainly the IMU status is switching from 'All OK' to a status where it is either not reading/using the Compass or GPS data or at times both and this begins early in the flight.

Here is the timeline

13:49:11 - Boot up - IMU status = 225 which means it is not yet ready
13:51:02 - Motor Start
13:51:05 - Takeoff - Voltage is 16.6
13:51:50 - IMU status finally is AOK
13:53:21 - Flight Mode status goes from 'Angle' to 'Acquiring' at the end of the circle you first made
13:54:30 - Flight Mode = Angle but simultaneously IMU status losses the compass and GPS
13:55:23 - IMU status = AOK
13:57:50 - Max Distance - Returning
14:01:12 - Low Batt 1 + Compass Warning
14:01:29 - Low Batt 2 + Compass Warning
14:01:33 - Flight Mode switched to RTH
14:01:34 - Flight Mode back to Angle but without GPS
14:01:40 - Motor 4 OFF + Motors Killed almost simultaneously = Impact

My take on the crash (and this is just my opinion), is that it is possible that the IMU was in a failing state before lift-off. I do not know how the different status's of the IMU plays into the messaging seen on the ST-16 screen such that maybe; everything looked good, but for sure the AOK status was coming and going throughout the flight.

I also do not know if the rapid decrease in voltage towards the end had anything to do with the crash but obviously it did not help.

From Low Batt 1 to the crash was 28 seconds. Your recounting of the ascent lines up with the RTH being turned on and then off - (you may not have known you did it but in these moments we can do things like this). Obviously the RTH began to work as it is normal to first climb to the RTH altitude but without GPS it is likely this would not have worked.

Ty Pilot, does this mean that the H "found" enough nbr of sats but not "locked" up to them ? 13:49:11 - Boot up - IMU status = 225 which means it is not yet ready

DoomMeister · Mar 31, 2020

Photo said:
Ty Pilot, does this mean that the H "found" enough nbr of sats but not "locked" up to them ? 13:49:11 - Boot up - IMU status = 225 which means it is not yet ready

It booted up with IMU status=97, I believe that was a typo on the part of TP.

Mrgs1 · Mar 31, 2020

So this is likely to happen more as the battery stock ages, and the useful charge cycles are approaching the end.

h-elsner · Mar 31, 2020

Photo said:
Ty Pilot, does this mean that the H "found" enough nbr of sats but not "locked" up to them ? 13:49:11 - Boot up - IMU status = 225 which means it is not yet ready

As I understand the IMU_status representsn the sensors used by flight contoller. It is a binary list of flags for the sensors. I did a lot of research in the ST10/ST16 firmware and tests with Q500 and Typhoon H. Below is my interpretation. If it is correct, I don't know. There is no documentation from Yuneec for the telemetry. All is just guessing.

Code:

dec | binary
----+-----------------
 65   0100 0001
 97   0110 0001
225   1110 0001
229   1110 0101
      |||| ||||
      |||| |||+-IMU
      |||| ||+--n/a
      |||| |+---Compass
      |||| +----n/a
      |||+------Sonar?
      ||+-------GPS
      ||        ..................................
      |+--------C-Compass  |   same as "cgps_used"
      +---------C-GPS      |   at telemetry Q500
                ..................................

My idea 225 means GPS is ready, 229 means the compass values are taken into account by FC.
What I know: With IMU status 225 the ST16 shows 'Ready' and you can arm the drone in all GPS assisted flight modes modes.

br HE

howfastyes · Mar 31, 2020

Thank you guys for all information.

But still something is not clear.

Few seconds before crazy things happened i stopped recording and prepared for next shot - aircraft was not moving and i was looking at the screen setting camera angle.
First warning that poped up in my eyes was Motor failure land immediately, i look at drone and i can clearly see it's fighting to get balance but i had control and aircraft was responding to my moves for few second as i remember it made maybe 20-30 meters in my direction. Then i felt lack of response to RC and aircraft started to ascend and fly at full speed in opposite direction. I've never pulled RTH switch and there was no time to do anything.

First warning was Motor failure, where is battery warning when voltage was so low?

I'm little confused now, here is the 20 minutes flight on this battery i've done 11 days before crash. I've drained it on purpose flying last minutes few meters away from me and maybe 1m above ground just to see how long i can fly (19:59 by log). For me looks like regular voltage drop but i don't know.

DoomMeister · Mar 31, 2020

I am surprised at the initial voltage drop seen during liftoff and before the full power climb to 120 meters. To me it is indicative of IR imbalance in the cells. Do you have a charger that has the capability to measure the IR?

howfastyes · Mar 31, 2020

I'm using Yuneec charger. This flight was done in cold conditions, around 2*C (35*F).

Bad flight 3 days ago during 25*C (77*F).

Mrgs1 · Mar 31, 2020

The one supplied or the Yuneec branded A10?

howfastyes · Mar 31, 2020

Genuine one that came with aircraft.

Thomas111 · Mar 31, 2020

I left the original yuneec charger in its box on day 1 bought the evpeak charger

howfastyes · Apr 1, 2020

Just discovered something:

I can start all motors but when i rev it up like take off - one of the motors starts making sound like hammer hit (very loud and metalic sound) then motor shuts down. That's the one that is not broken due to crash. All 6 motors can run on idle speed without issues and weird sounds. Edit: Maybe i should not make take off without props?

I will connect H to PC soon, have to stick the broken arms with tape to safely run motors. EDIT: All 6 motor spin in test mode.

It is possible that the failing motor could drain battery so fast and caused crash? Like i said before, first warning was motor fail and there is no evidence of that in flight log.

Here is what i remember: