Flyaways?

[cdronefly]

Here is the last word on flyaways.....
Fear of Flying 2016 - The Last Word on Drone Flyaways! - Drone Flyers

 

[flyNfrank]

I'm in the camp that believes that calibrating every flight will increase risk of a bad calibration. There is no logical reason to calibrate at every flight location, much less every battery. The exceptions are:
1. If you your flight location is over 100mi from the last flight
2. The sensor mod value is out of range
3. You've just loaded new firmware.

Calibrating every flight location introduces the risk of being in an area that you have NO IDEA what's below you for magnetic anomalies, and if you do have something strange at the area you can get a bad calibration and not know it. If the Mod value is 1450 to 1550, I always fly the same calibration. If it's beyond this value, I'll move the bird about 40-50' to another location and 90% of the time the mod value is within the 1450-1550 range and I fly it. In the 7 months I've been flying my P3P, I've had only one compass error (about 6wks ago) and I had to bring the bird back to recalibrate. I deem that as pretty solid, and I've never had a hint of a fly away. I do know that P2 craft had many reports of fly-aways (pre-lightbridge), however many P2 owners swear by them. That's why I waited until July to buy the P3P, waiting for the reports, or lack of.

I could get into a lengthy conversation with you based on your quoted response here. But I wont and don't plan to. I will repeat some of what I have said in the past which is....most know I'm a dji product and beta tester. We have been told more then once by those that develop the firmware the compass should be calibrate each time prior to start up. It only takes 15 seconds max to do the process. I think if I were on one of my 2 miles out from home point and happen to have a compass error I would probably have heart failure. I do know I would be lighting a cigarette up for sure. Anyway, the last time the engineers mentioned calibrating the compass I asked for an explanation as to why it should calibrated before each start up. And they never did/would give a response.

Like I said I don't want to hash on this subject with you, and don't want you getting upset either. So let's move on.... Thanks!

 

[syotr]

Here is the last word on flyaways.....
Fear of Flying 2016 - The Last Word on Drone Flyaways! - Drone Flyers

More like the latest BS. Consider the source.

 

[Msw46]

Judging by the amount of flight put on to you tube that do not conform to flight regulations and without the knoledge before take of its not surprising these are fly aways

 

[N017RW]

Last word??? Pretty narcissistic.

I've always contended flyaway is a verb not a noun.

 

[dirtybum]

Sooo, yer saying dont worry bout calibrating right Frank![emoji41]

QUOTE="flyNfrank, post: 681879, member: 7198"]I could get into a lengthy conversation with you based on your quoted response here. But I wont and don't plan to. I will repeat some of what I have said in the past which is....most know I'm a dji product and beta tester. We have been told more then once by those that develop the firmware the compass should be calibrate each time prior to start up. It only takes 15 seconds max to do the process. I think if I were on one of my 2 miles out from home point and happen to have a compass error I would probably have heart failure. I do know I would be lighting a cigarette up for sure. Anyway, the last time the engineers mentioned calibrating the compass I asked for an explanation as to why it should calibrated before each start up. And they never did/would give a response.

Like I said I don't want to hash on this subject with you, and don't want you getting upset either. So let's move on.... Thanks![/QUOTE]

 

[Sky King]

I have had 2 P2 flyaways. One was lost never to be seen again and the other was recovered due to using a Trackimo. After putting 160mi on my P3P, and always using a Trackimo, (just in case) I have not had even a close call on a flyaway. The P3P is a much more reliable machine and the only problem I had with it was a FW update which DJI took care of completely. I have great confidence in this machine. I always auto-take off and hover a few moments, climb, hover a moment and then fly my mission with confidence!

 

[Fantasmagorico]

Flyaways are very rare with the Phantom 3. In most (all?) cases, it was probably due to pilot error (e.g. too much wind and/or confusion after auto flipping into ATTI mode). If you take the time to gain a deep understanding of how your Phantom works, this should never happen to you.

Well, it happened to me today and I don't think I made any error. My P3P had taken off from a big, empty parking lot near Fifth Street in Miami Beach and I flew it above and slightly off Alton Road toward the area of South Pointe, a distance of 2,389 feet (according to my flight log), less than half a mile, on a clear line of sight without obstructions. I had flown this route previously without incident. I lost RC signal and the aircraft initiated RTH a few seconds later. I was using the DJI Go app with the latest firmware (2.8.0) and had set it to Smart Return to Home. I was using Home Lock as the intelligent flight mode and, just in case, had set the failsafe altitude to 400 feet because there were buildings in the area and I know from experience that RTH does not always take place in a straight line. In any case, no building was in the flight path. I lost sight of the bird momentarily as it rose, but made eye contact again after I got the message "Aircraft Descending." Unfortunately, it was descending over the busy intersection of Alton Road and Fifth Street, about two hundred yards from the home point. I took over manually and cancelled RTH. However, no matter how much I tried, the aircraft would not move toward home. It kept drifting to the sides and toward a building. I managed to keep it away from the building and moved it toward the channel of Miami Beach Marina to prevent it from dropping on the McCarthur Causeway or on the boats in the Marina. Again, I had a clear line of sight without obstructions. From there, I kept trying desperately to bring it back, but it was as if there were an invisible barrier that it couldn't cross at about 200 yards from me. I could move it right or left on an arc, or away from me, but not toward me. The reactions to the stick were erratic, so that if I moved it to the right, it would turn left first, and then right, for example. The moment I let go of the controls, it wouldn't hover, but drift away fast. There was an 18 mph wind at 300 feet according to my UAV Forecast app, but only 10 mph at 30 feet according to my Drone Buddy app, and it was not blowing against the bird, but rather slightly across, and should have blown it closer to me, not farther. At that point, the bird was at a height of about 100 feet. I turned it to face every which way and moved the stick in every direction, but nothing worked. It just refused to return home. Eventually, the battery ran out and my dear, five-month old Phantom found a watery grave in Biscayne Bay. I am very upset and it seems that this is definitely a software problem, not pilot error. In fact, I had noticed a slightly different version of this problem before. I like to take control of any RTH maneuver toward the end for a softer landing; if you let the P3P land on its own, it tends to land too hard. Sometimes, when the bird was about 20 feet from me, it would refuse to come closer. I accounted for that eventuality by taking off only from areas with plenty of room to land. I thought that was quirky, but never imagined that the Phantom would refuse to get closer than 200 yards. There is definitely a factory problem. What a bummer!!!

 

[msinger]

Fantasmagorico, it sounds like you were flying in ATTI mode during the part of your flight where it seemed you had no control over your Phantom. Please upload your TXT flight log here and post a link back here if you'd like us to review it and help explained what happened.

 

[John Locke]

I could get into a lengthy conversation with you based on your quoted response here. But I wont and don't plan to. I will repeat some of what I have said in the past which is....most know I'm a dji product and beta tester. We have been told more then once by those that develop the firmware the compass should be calibrate each time prior to start up. It only takes 15 seconds max to do the process. I think if I were on one of my 2 miles out from home point and happen to have a compass error I would probably have heart failure. I do know I would be lighting a cigarette up for sure. Anyway, the last time the engineers mentioned calibrating the compass I asked for an explanation as to why it should calibrated before each start up. And they never did/would give a response.

Like I said I don't want to hash on this subject with you, and don't want you getting upset either. So let's move on.... Thanks!

This is a point, what happens if I'm two miles out and I get a compass error? I know the craft switches to ATTI automatically, but what about telemetry?

In 10mos of flying my P3P I've only had one compass error, which was about 4mos ago, and it was only about 175' away at the beginning of a flight. I returned in ATTI and landed fine. However, everything happen so quick I didn't take the time to observe the screen telemetry, I was highly focused on flying the craft LOS in ATTI.

Does anyone know what happens to telemetry data when the craft automatically enters ATTI during a compass error? Do you still have the radar screen info? Do you still have the map with the green line that show how to get home? I'm thinking the radar goes away but the map with return path should still work. Anyone know?

Usually it's possible to find your way to home using FPV, but I've been in mountain areas where that's difficult to do, especially if I'm not familiar with the terrain.

 

[Fantasmagorico]

Fantasmagorico, it sounds like you were flying in ATTI mode during the part of your flight where it seemed you had no control over your Phantom. Please upload your TXT flight log here and post a link back here if you'd like us to review it and help explained what happened.

Thank you. I was never in ATTI mode, but it seems from what I have read after the incident that switching to it and then back to GPS mode, or staying in ATTI mode to fly with the arrow, might have helped. I attach the flight log. You can also access additional flight data from Healthy Drones at this link:
HealthyDrones.com - Innovative flight data analysis that matters

 

[Trackman1]

From following crash and fly away threads, I notice most occur during the first 90 seconds of flight. Usually in the first 60.

Let it hover and make sure it is stable. Check your sensor screen mod values. Get use to what is normal.

Check the battery screen. Check satellites (number and signal strength)

If you lose GPS DO NOT hit return home as it does not know where to go. But it does go somewhere. A lot have called this a fly away.

Make sure your return to home altitude is set higher than anything you could run into.

 

[flyNfrank]

Thank you. I was never in ATTI mode, but it seems from what I have read after the incident that switching to it and then back to GPS mode, or staying in ATTI mode to fly with the arrow, might have helped. I attach the flight log. You can also access additional flight data from Healthy Drones at this link:
HealthyDrones.com - Innovative flight data analysis that matters

I downloaded and converted your flight log and scanned over it briefly for redflags. I saw some things I never seen before which got me interested. I then came back here to see what you may have said in your post. I will now pay more attention to areas of the flight log I haven't really looked at yet.

At this time I feel like the number 1 reason most all of this took place centers around the fact the you started this flight with a 75% charged battery. You only have a short time before the battery starts into set cycles per the remaining charge. Then based on what you have the app set at when it reaches the more critical stages will determine how you will be able to fly the a/c. Btw, the temp on the battery also reached 143.5 degrees. That is close to melt down.

 

[Fantasmagorico]

I downloaded and converted your flight log and scanned over it briefly for redflags. I saw some things I never seen before which got me interested. I then came back here to see what you may have said in your post. I will now pay more attention to areas of the flight log I haven't really looked at yet.

At this time I feel like the number 1 reason most all of this took place centers around the fact the you started this flight with a 75% charged battery. You only have a short time before the battery starts into set cycles per the remaining charge. Then based on what you have the app set at when it reaches the more critical stages will determine how you will be able to fly the a/c. Btw, the temp on the battery also reached 143.5 degrees. That is close to melt down.

1. The 75% charge is due to the fact I started to fly using VR goggles for the first time using the FPV Camera app. I didn't like the goggles and landed the bird, then took off again using the DJI GO app.
2. I noticed that about the battery and wondered too why it happened. I have no idea. I never pushed the throttle hard. But that was not the reason for the crash. The battery just ran out of juice. Also, note that, despite starting with 75%, the battery had enough charge for about 10 minutes more when the problem started, which is the time I struggled to regain control.
3. Thanks for your interest in the matter.

 

[msinger]

Fantasmagorico, you can see your flight in my log viewer here.

At 17m 29s into your flight, your battery hit the critical low level and your Phantom started to auto land. While there is no way to cancel that landing, you can give your Phantom full throttle up and steer it to a safe landing location. If you do this for a long enough period of time though, the battery will eventually die and your Phantom will drop like a rock (as seen at the end of your flight).

Other than being in ATTI mode while being pushed by the wind, I'm not sure why you were not able to bring your Phantom close to the home point. Is it possible that you were disoriented by the forced auto landing?

It's important to watch your battery voltage while flying (see more details here). Your Phantom was about 392 feet in the air when the voltage was at 3.6V and 146 feet when you hit the critical low battery level.

 

[flyNfrank]

The a/c was in the process of climbing up to 400ft when the battery expired. The flight log shows the a/c spent most of the end of the flight climbing in altitude rather then being landed. It was attempting to RTH when you were trying to take over. When you stopped it would go back to climbing to the 400ft height. All of that activity drained the battery that much quicker.

 

[msinger]

It was attempting to RTH when you were trying to take over.

The log shows it was attempting to auto land at its current location.

 

[flyNfrank]

No doubt that msinger's site helps to simplify what takes place during most crashes. When you hover the cursor over one of the white or red circles provides even more detailed info at that point. What is seen here is included with the uploaded flight log details which really helps when diagnosing a bad flight. Nice work msinger!

 

[flyNfrank]

The log shows it was attempting to auto land at its current location.

.....AND???

My reference was pretty much from the 18.13 point. Even though flight details show "Autolanding", the flight went to a RTH attempting to go up to the 400ft set point in the app. When the user manually over took the controls the climb paused until the controls went to neutral once again. And once that happened it returned to the 400ft climb.

The altitude data shows the a/c increasing altitude to the end, not landing.

 

[msinger]

My reference was pretty much from the 18.13 point. Even though flight details show "Autolanding", the flight went to a RTH attempting to go up to the 400ft set point in the app. When the user manually over took the controls the climb paused until the controls went to neutral once again. And once that happened it returned to the 400ft climb.

The altitude data shows the a/c increasing altitude to the end, not landing.

Prior to the auto landing, he was flying in home lock mode. Perhaps that's why he noted the odd flight behavior here:

I could move it right or left on an arc, or away from me, but not toward me.

Then, the mode changed from home lock to auto landing. I don't think it's possible that it could have been trying to RTH and auto land at the same time. It appears like he was manually increasing the altitude to attempt to counteract the auto landing. I did not confirm by checking the stick positions, but that would be the next step if we wanted to dig into this further.