By the way, be great to hear your feedback on that GPS kit when you get a chance to test it. Let us know on this thread if you don't mind.castlemaster said:
Phantom 2 flyways
- Thread starter dvm71
- Start date
Yeah I'll defo report back on the gps, I'll hopefully have it by the weekend so I'll do a few practice landings out of view and see where the gps reports it as being.
@Jaybee
Sorry to hear about your loss. I have a TK102 which works great. I am in the US but I had a look on amazon.co.uk and looks like they are available in Blighty too. Just add a pay as you go sim card and you are all set. You call the TK102 and it hangs up, then sends an SMS with the current location, which at least in android goes straight into google maps if you click it.
Sorry to hear about your loss. I have a TK102 which works great. I am in the US but I had a look on amazon.co.uk and looks like they are available in Blighty too. Just add a pay as you go sim card and you are all set. You call the TK102 and it hangs up, then sends an SMS with the current location, which at least in android goes straight into google maps if you click it.
Cheers. Does it work well, have you tested it while the Phantom is flying to see if it works as expected? I noticed a few people on Amazon have reported they blew-up on them while charging which is slightly worrying.SARC said:
ianwood
Taco Wrangler
Nice summary of potential causes. Some comments:
3. The drunk guy leaning over. This makes a lot of sense. And I guess the only way to recover from this is to switch to full manual control and guide it to a controlled (or not so controlled) crash.
4. I've watched that video a number of times. Something continues to control the P2V even when the owner said it was out of his control. The gimbal even moves a couple of times in between dodging in and out of power lines and at many times it looks like its actually under control.
Finally, I'd add one more to your list that I'm guessing contributes to at least a few fly aways:
Pilot error under controlled flight: inadvertently switching modes on the controller, orientation confusion, applying the wrong control input. As a new owner of a P2, I am being very conscious of how easy it would be to make these mistakes.
3. The drunk guy leaning over. This makes a lot of sense. And I guess the only way to recover from this is to switch to full manual control and guide it to a controlled (or not so controlled) crash.
4. I've watched that video a number of times. Something continues to control the P2V even when the owner said it was out of his control. The gimbal even moves a couple of times in between dodging in and out of power lines and at many times it looks like its actually under control.
Finally, I'd add one more to your list that I'm guessing contributes to at least a few fly aways:
Pilot error under controlled flight: inadvertently switching modes on the controller, orientation confusion, applying the wrong control input. As a new owner of a P2, I am being very conscious of how easy it would be to make these mistakes.
Ksc said:
ianwood
Taco Wrangler
Jaybee,
If I am counting correctly, yours seems to be the second P2 on this forum that has flown away (the rest are P1 and P2V reports). The first one sounds like it was having issues right out of the box (4 fly aways!). If you don't mind me asking, what are the details of yours?
How many times had you flown it?
Did you crash it all?
Any modifications? Did you open it up at any point?
Where were you? Any potential sources of interference in the area? Had you flown there before?
When did you last calibrate compass or IMU?
How did it fly away? Straight off in one direction? Was it hovering just before or after you lost control? How well could you see it?
What mode were you flying in? Did you try switching modes when you felt it was not in your control?
Sorry for your loss and I hope it turns up somewhere. Anyway, maybe we can learn something from it. I certainly have "phanxiety" over fly aways and am looking for clues as to what is common amongst them. I upgraded to a Futaba radio but I have no real idea if that is going to help prevent anything.
There should be a registry for lost Phantoms! And before you can rebind or physically change out your receiver, you have to plug in your phantom and connect to the registry to verify it first.
If I am counting correctly, yours seems to be the second P2 on this forum that has flown away (the rest are P1 and P2V reports). The first one sounds like it was having issues right out of the box (4 fly aways!). If you don't mind me asking, what are the details of yours?
How many times had you flown it?
Did you crash it all?
Any modifications? Did you open it up at any point?
Where were you? Any potential sources of interference in the area? Had you flown there before?
When did you last calibrate compass or IMU?
How did it fly away? Straight off in one direction? Was it hovering just before or after you lost control? How well could you see it?
What mode were you flying in? Did you try switching modes when you felt it was not in your control?
Sorry for your loss and I hope it turns up somewhere. Anyway, maybe we can learn something from it. I certainly have "phanxiety" over fly aways and am looking for clues as to what is common amongst them. I upgraded to a Futaba radio but I have no real idea if that is going to help prevent anything.
There should be a registry for lost Phantoms! And before you can rebind or physically change out your receiver, you have to plug in your phantom and connect to the registry to verify it first.
SARC said:
Which TK102 are you using? I looked on Amazon and see quite a few different ones. Thanks!
I have had two and it's being boxed up and sent back to B&H. One time I was out and the bird flew perfect. I landed it and was happy. When I took off the second time (within 3 minutes of the successful flight), it lifted up and immediately went toward a forest and hit a tree. Just broken props and the camera fell off, other than that it worked.
Second day, a successful flight around my house...no issues.
Third day, a successful flight, no issues.
Fourth day, lifted off and immediately went behind me and into the neighbors house, about 200 ft. away. Luckily no damage to their house and they weren't home. Broken prop and tore the camera plug out (which was lost)
I realize I shouldn't be flying in a neighborhood, but I am convinced something was wrong with this Phantom. I had updated everything...transmitter and receiver and didn't fly it until the wi-fi app was updated.
I had so much fun, but want a product that is solid.
Second day, a successful flight around my house...no issues.
Third day, a successful flight, no issues.
Fourth day, lifted off and immediately went behind me and into the neighbors house, about 200 ft. away. Luckily no damage to their house and they weren't home. Broken prop and tore the camera plug out (which was lost)
I realize I shouldn't be flying in a neighborhood, but I am convinced something was wrong with this Phantom. I had updated everything...transmitter and receiver and didn't fly it until the wi-fi app was updated.
I had so much fun, but want a product that is solid.
Sorry to hear of others who have had flyaways.
You can read my sad tale here.
Question for you fly-away victims:
1. What vendor did you purchase from, and have they done ANYTHING to help you?
2. Did you attempt to contact DJI Support? Via phone or email? Response?
I purchased from B&H Photo/Video, whom I've done business with before and liked a lot. Their response to this has been less than satisfying. Early on I had hopes that they would take the ball to DJI's court on my behalf; at this point, I think they're just hoping I go away. I really don't think I should POOF lose $800 in gear for their defective product. I don't. (And I'm kind of getting tired of the "it was pilot error" mantra espoused on these forums. I'm sure some are. But I can stand before God with 100% confidence that I did everything by the book, and my Phantom had a defect that caused it to leave me forever.)
I've already replaced my GoPro. $300 plus memory card.
I've suggested to B&H that I'd be satisfied with a credit toward purchase of a Phantom 2 (non-vision). I was hopeful that the P2 had updated hardware making the flyaways a thing of the past. Perhaps not.
At any rate, B&H has not responded to my proposed resolution. I'm willing to be patient and keep the heat on, I guess.
I've never had an expensive product just self-destruct after 6 weeks of ownership with no assistance from the manufacturer or vendor.
You can read my sad tale here.
Question for you fly-away victims:
1. What vendor did you purchase from, and have they done ANYTHING to help you?
2. Did you attempt to contact DJI Support? Via phone or email? Response?
I purchased from B&H Photo/Video, whom I've done business with before and liked a lot. Their response to this has been less than satisfying. Early on I had hopes that they would take the ball to DJI's court on my behalf; at this point, I think they're just hoping I go away. I really don't think I should POOF lose $800 in gear for their defective product. I don't. (And I'm kind of getting tired of the "it was pilot error" mantra espoused on these forums. I'm sure some are. But I can stand before God with 100% confidence that I did everything by the book, and my Phantom had a defect that caused it to leave me forever.)
I've already replaced my GoPro. $300 plus memory card.
I've suggested to B&H that I'd be satisfied with a credit toward purchase of a Phantom 2 (non-vision). I was hopeful that the P2 had updated hardware making the flyaways a thing of the past. Perhaps not.
At any rate, B&H has not responded to my proposed resolution. I'm willing to be patient and keep the heat on, I guess.
I've never had an expensive product just self-destruct after 6 weeks of ownership with no assistance from the manufacturer or vendor.
Just a thought.... We are at (or near) the peak of the current 11 year solar cycle.
This will probably be the first solar cycle where the density of devices that are GPS dependent has reached the mainstream consumer.
Solar Flares and/or CMEs (Corona Mass Ejections) produce a certain amount of EMF, which can be strong enough to travel into the area that affects Terestrial communications.
Has anyone tried to track the warnings from spaceweather.com and draw any comparisons to failure/inaccuracies in GPS dependent devices?
As a rule, if your car's GPS shows you driving an inch away from the road you are on, there may be something affecting GPS operation, and it may not be a god day to fly.
There have been warnings issued following CMEs.
Again, just a thought.....
This will probably be the first solar cycle where the density of devices that are GPS dependent has reached the mainstream consumer.
Solar Flares and/or CMEs (Corona Mass Ejections) produce a certain amount of EMF, which can be strong enough to travel into the area that affects Terestrial communications.
Has anyone tried to track the warnings from spaceweather.com and draw any comparisons to failure/inaccuracies in GPS dependent devices?
As a rule, if your car's GPS shows you driving an inch away from the road you are on, there may be something affecting GPS operation, and it may not be a god day to fly.
There have been warnings issued following CMEs.
Again, just a thought.....
ianwood said:
Hey buddy, this is a long post but hope it may help others.
How many times had you flown it?
About 8 times. Makes me a novice user, sure, however I am good with technology and fully understand what I am doing in terms of checklists and common sense. Knowing the Phantom has the failsafe feature while flying in GPS mode provides a level of confidence in terms of alleviating natural concerns about flying at a reasonable altitude.
Did you crash it all?
Nope.
Any modifications? Did you open it up at any point?
Nope, only the Zenmuse.
Where were you? Any potential sources of interference in the area?
None, no power lines, source of EMI etc. I even checked where all the mobile antennae are near the area, the nearest was apx. 800m away from my start location and away from the flight area. Here is the mobile antennae stats if anyone is interested:
– Microcell
– Height of Antenna: 8.3 Metres
– Frequency Range: 1800 MHz
– Transmitter Power: 16.9 dBW
– Maximum licensed power: 32 dBW
– Type of Transmission: GSM
Had you flown there before?
Yes, twice.
When did you last calibrate compass or IMU?
The compass; when I first got the Phantom in late December. And the last time I was at this location (previous flight last week). The IMU; never. I’ve never had the need to, it’s always flown very nicely. In fact, the hover on this particular day was the best I’d every seen it — was so stable, almost not moving. No doubt due to the low 4mph wind speed on Sunday.
How did it fly away? Straight off in one direction? Was it hovering just before or after you lost control? How well could you see it?
I had eyes-on and line of sight, but all of a sudden it didn’t respond to the Tx when I tried to bring it home and continued to fly west of my location. It was then out of sight, gone.
About 30 secs or so after this, I killed the Tx and tried to work out where it had gone. It was fairly high the last time I saw it, apx. 30-50m and apx. 300m north of it’s home location. As we know, the advertised range is 1000m in open spaces. Plenty of people seem to have flown out of visual range with FPV and the Phantom engaged RTH when out of range.
If it had entered the failsafe mode when I killed the TX it had plenty of height and was in range / battery for RTH. This is the crux of the problem, for whatever reason this just didn’t happen.
What mode were you flying in? Did you try switching modes when you felt it was not in your control?
GPS as always. No, I killed the TX to try and get the thing home.
_____
I understand anyone with anxiety, it’s natural to feel like that, especially when we hear that Phantoms have flown away. The cost is fairly significant too with a GoPro 3+ Black, memory card, and gimbal. My setup cost me around £960 ($1577). To be honest I felt a bit of anxiety to begin with but, as any Phantom owner would, I thought my care and attention to the important details would prevent a flyaway caused by human error.
"If I make sure to stick to the check list and didn’t make any glaring mistakes, everything will be fine as the failsafe is only a Tx power-off away." This is a comforting thought and a major selling feature, Without this feature, it’s far more likely things could go wrong. This is why it’s obviously massively disappointing if the failsafe fails to do it’s job.
_____
Here’s a list of possible human error causes and how they relate to my particular situation:
Not waiting for the rapid green lights then steady green before take off
I’m 100% sure I the sequence was correct as per usual.
Flying straight off without testing hover and responsiveness
I always make sure I test the hover / responsiveness and fly about nearby for a couple mins to ensure everything is in order before getting to any altitude.
GoPro WiFi ON
I’m 100% sure mine was off as always.
S1 is in GPS mode not ATTI
I’m 100% sure mine was set to GPS.
S1 gets flipped out of GPS mid-flight
I am 100% sure this didn’t happen.
Flying in strong winds
As I mentioned, on this day I had the calmest winds out of any previous flight. It was around 4 mph according to weather reports and I can verify that there was no wind or sudden gusts and that the Phantom was extremely responsive and stable, even at a high altitude.
Flying near a strong source of EMI
None in the area (see above regarding a single mobile antennae at ground level, well away from the point when I lost control).
Compass calibration required (from the manual):
(1) When Compass Data is abnormal, the LED flight indicator will blink alternating between red and yellow.
I am 100% sure my Phantom wasn’t blinking red/yellow
(2) Last compass calibration was performed at a completely different flying field/location.
The last time I flew was the same location.
(3) The mechanical structure of the aircraft has changed, i.e. changed mounting position of the compass.
I’d changed nothing here.
(4) Evident drifting occurs in flight, i.e. the aircraft doesn’t fly in straight lines.
Flight and hover was very stable.
Flying with a low battery
I am 100% sure I had a full battery to begin with. Before my last flight path, I’d flown the Phantom back around to check the battery. It had 3 green lights, this is around 2-4 mins before it flew off.
_____
Possible reasons for this fly away
• Something messed with the compass and GPS causing the Phantom to fly away. From my research, many seem to dismiss cell towers cause problems. However, the Phantom manual states: "6.1 (5) Try to avoid interference between the remote controller and other wireless equipment. (No base stations or cell towers around).” The video with Colin Guinn advising how to avoid flyaways also suggests to avoid cell towers. I’ve read suggestions that cell companies actually use drones to check their towers (the big ones). Others suggest flying near cell transmitters with 2.4Ghz drones doesn't matter (check RC Groups for more). Clearly we need more conclusive info. Either way, I almost certain this has anything to do with my flyaway.
• The system crashed, i.e, for whatever reason the code running had an error and the failsafe didn’t engage. Quite plausible. Here’s an interesting point of view from http://www.rcgroups.com/forums/showthread.php?t=1872000&page=66
• The battery ran out before I killed the Tx: impossible, it had 3 lights around 2-4 mins before I lost control.
• Prop came loose / fell off. Hardly likely with the self-tightening mechanism, and considering the apparent stability of my Phantom in this instance.
• iPhone / smart phone in pocket causes Tx issues. No idea here, but I’ve never had any issues myself. If there was a problem, I’d expect the Phantom to RTH as it would with a broken / obstructed Tx signal.
• GoPro malfunction causes it to switch it’s WiFi on. Never heard of such a situation, but I guess it could happen in theory.
• A malfunctioning / faulty motor. This would certainly cause unexpected flight behaviour. I’d suggest that if things are normal at the start of a flight and if you’ve not crashed prior to taking off again, this is highly unlikely.
_____
Preventative measures going forward
• Install some form of GPS tracker. Test fully, even while in-flight so if the worst happened you could kill your Tx and quickly phone the tracker to try and get a location, especially if out of visual / or heading out of visual at a high altitude. Test often.
• Mark the Phantom with contact details. You never know, someone may find a lost craft and get in touch. Easy to do.
• If the concern / fear of loss is so great, simply don’t fly high and far away. Keep the craft nearby. Not going to be much fun, but a safer bet.
• Only fly in the countryside where you are surrounded by fields for at least 1000m. Same as above.
• Don’t use the stock controller. I don’t have any experience with this one, but many suggest the third party controllers are far more reliable. Either way, I’d still expect a RTH situation if my stock controller’s signal is; obstructed, turned off, looses power.
• Install a more robust third-party Rx onboard the craft. Again, I don’t have any experience with this either but same as above.
• Always make sure you can get eyes-on the Phantom’s lights, so you can monitor any potential compass failures and perhaps switch to ATTI to manually fly home. Altitude and distance will be fairly limited if deciding to practise this one. Or get a spotter with binoculars.
• Fly ATTI only. Very risky when flying distances / high altitudes. I wouldn't even attempt this mode unless I was flying in close range. I bought the Phantom 2 mainly for filming. The failsafe mode is the dealmaker for users like me.
• Fly FPV with iOSD / telemetry. This was going to be my next upgrade. I guess it further complicates debugging a potential flyaway as it introduces more variables, but at least you could get an idea of the flight path / erratic behaviour prior to any flyaway issue - providing you’re in range of your FPV’s signal.
• Film your flight from a tripod mounted camera / friend with a cam etc. Not always practical and not always going to explain stuff, but may help rule out certain causes and/or help others to avoid mistakes.
_____
In conclusion
I’d love to know what happened to my P2. I strongly suspect it’s a compass / GPS thing and it tripped out and never initiated a RTH. The cause will no doubt be an unsolved mystery. I really wonder if the flight software had something akin to a kernel panic.
With the power that comes with these cool devices, we should all be continually conscious of the risks. No doubt, safety improvements will follow in years to come. I’d hope eventually these crafts could somehow communicate realtime flight data (black box style, not just telemetry). This way you could know for sure what went wrong in the case of a mechanical / software error. This is critical to rule out human error gong forward. A built in GPS tracker would be a great feature too.
Hope some of this is useful to others. I'll add updates if I get any more info. Cheers.
ianwood
Taco Wrangler
Solar activity is easy to track. We've had an event the other week but nothing big enough to cause noticeable issues with GPS or other sensitive services. Maybe slightly diminished RF performance. I'm guessing but I don't think it's an issue. Anyway, don't bother spaceweather.com. They charge you for alerts when you can get them for free here:
https://pss.swpc.noaa.gov/LoginWebForm. ... service%2f
And don't trust your car to evaluate GPS performance. It too has a magnetometer compass in it. When's the last time you calibrated that? I would guess, like me, never. Android has a nice little app called GPS status. Anyway, I've never seen GPS issues in good weather. And I've never seen dramatic jumps in position. I've seen drift but it was very very slow moving drift. Your Phantom doesn't really care about atmospheric drift much as it is all relative and it certainly wouldn't cause a fly away.
https://pss.swpc.noaa.gov/LoginWebForm. ... service%2f
And don't trust your car to evaluate GPS performance. It too has a magnetometer compass in it. When's the last time you calibrated that? I would guess, like me, never. Android has a nice little app called GPS status. Anyway, I've never seen GPS issues in good weather. And I've never seen dramatic jumps in position. I've seen drift but it was very very slow moving drift. Your Phantom doesn't really care about atmospheric drift much as it is all relative and it certainly wouldn't cause a fly away.
Dalite said:
ianwood
Taco Wrangler
Jaybee, great post. Lot's of detail. Some comments. A 1.8GHz microcell isn't going to be a problem unless it was broken and scattering RF all over the place and/or you parked your phantom right next to it. It's simply too far away spectrum wise for it to achieve power levels that far out of band. Same with the 2GHz towers. It would need to be in band at the 2.4GHz range. And having your phone in your pocket might not be a good idea, but people have ben strapping cell phones to Phantoms (not advisable) and flying them.
(Side note: Does anyone know if the stock P2 transmitter/receiver uses frequency hopping or not? I have a basic 2.4GHz analyzer. I should have done a test before I pulled out the stock receiver. Maybe I can power it up externally and check.)
So, interference could have caused it but there's certainly no smoking gun. And outside of firmware/hardware issues in the Phantom, I don't think GPS was your issue either. Your issue was it didn't respond. So either your inputs weren't being received or the Phantom was busy dealing with another issue like a motor problem. Remember that the Phantom's priority is to maintain attitude in both GPS and ATTI modes. All other functions are secondary. (Someone correct me if I am wrong.) So if one corner of your Phantom was suddenly generating less lift, it would likely result in your Phantom rapidly heading off in one direction trying to maintain attitude. Apparently, manual mode is the only way to escape from the bad motor fly away scenario and likely you're going to crash it. I wonder if a rapid descent is possible in GPS or ATTI modes.
Flawed hardware, frozen firmware is possible. I've seen video of at least two recent fly aways that were just truly bizarre like the machine was possessed (one P1 and one P2V). I don't know anything about Naza but I'm wondering about debugging and logging data. There's two audio channels on most FPVs that don't get used. I'm wondering if you could modulate OSD and debug data onto them and then demodulate and record it into an Android app.
As for steps to avoid it, definitely upgrade your receiver and transmitter. The P1 version was clearly prone to issues. The P2 version may be better, but if it isn't frequency hopping, it could still easily get knocked out and who knows if it properly filters out spurious TX on the same band that could cause it to lock up.
(Side note: Does anyone know if the stock P2 transmitter/receiver uses frequency hopping or not? I have a basic 2.4GHz analyzer. I should have done a test before I pulled out the stock receiver. Maybe I can power it up externally and check.)
So, interference could have caused it but there's certainly no smoking gun. And outside of firmware/hardware issues in the Phantom, I don't think GPS was your issue either. Your issue was it didn't respond. So either your inputs weren't being received or the Phantom was busy dealing with another issue like a motor problem. Remember that the Phantom's priority is to maintain attitude in both GPS and ATTI modes. All other functions are secondary. (Someone correct me if I am wrong.) So if one corner of your Phantom was suddenly generating less lift, it would likely result in your Phantom rapidly heading off in one direction trying to maintain attitude. Apparently, manual mode is the only way to escape from the bad motor fly away scenario and likely you're going to crash it. I wonder if a rapid descent is possible in GPS or ATTI modes.
Flawed hardware, frozen firmware is possible. I've seen video of at least two recent fly aways that were just truly bizarre like the machine was possessed (one P1 and one P2V). I don't know anything about Naza but I'm wondering about debugging and logging data. There's two audio channels on most FPVs that don't get used. I'm wondering if you could modulate OSD and debug data onto them and then demodulate and record it into an Android app.
As for steps to avoid it, definitely upgrade your receiver and transmitter. The P1 version was clearly prone to issues. The P2 version may be better, but if it isn't frequency hopping, it could still easily get knocked out and who knows if it properly filters out spurious TX on the same band that could cause it to lock up.
- Joined
- Dec 17, 2013
- Messages
- 32
- Reaction score
- 1
jthorstad, please message me about returning to BHPhotoVideo. I posted a response earlier that was removed. I've been in numerous forums, and this is a first for me. No profanity was used and I did not intend for my observations or opinions to be inflammatory. I only expressed my thoughts on fly-a-ways, disappointment that a small number of owners who have not yet had a fly-a-way assume the fault is always with the operator, and disappointment in DJI's firmware and sub-par customer service. I was under the impression that the exchange of ideas and opinions was the purpose of forums.
Thanks in advance to the forum admins if this post makes it to the forum.
Thanks in advance to the forum admins if this post makes it to the forum.
There is one big difference: Your TV will not be flying aroundJimmy Jim said:
There certainly is a problem with the quality control of Chinese mass production items like the DJI Phantom and other RC flying stuff.
A Phantom 2 with Zenmuse and GoPro has a mass of about 1.3 Kg.
If this falls suddenly out of the sky it can really hurt of even kill someone.
That's why I have opened my P2 and checked all solder joints.
Maybe I should even open the box of the NAZA after watching this video http://www.youtube.com/watch?v=sRk6OLbNZ2s.
Last week I had a problem with my FPV TX. My monitor was showing signal reception, but there was no video from my Hero3 and iOSD mini.
After checking the installation I found out that the yellow video wire did not make contact with the pin in the plug of the TX. Obviously some Chinese worker (bored and tired?) had forgotten to remove the isolation from the wire before crimping it in the pin of the plug...
ianwood
Taco Wrangler
DuoDiscusxT said:
Oh man, that's no good at all! Any other reports of bad/missed solder joints from DJI? Putting my OSD inside this week, gonna take a close look around with a magnifying glass.
Jaybee said:
Big Ben said:
This listing says the v1 Phantom does frequency hopping too:
http://www.redrockethobbies.com/DJI-Pha ... hantom.htm
So maybe they all do but maybe it's not as resilient a form of frequency hopping. The v2 has a diversity antenna which is much better and you can use a Futaba radio with the stock v2 RX which you couldn't on the v1 so there are definitely differences. I swapped to the Futaba RX before I even took my first flight with the P2 simply because I've yet to find a Futaba TX/RX fly away with a Phantom. I'm sure it has happened but looks like the odds are much better. It's a very unscientific conclusion but going on what I can figure out.
Similar threads
