Can someone explain Flyaways?

Joined
Nov 19, 2013
Messages
103
Reaction score
1
Whats up guys...

So I just got my PV, Three flights, no issues except very minor ones related to the Camera... What I would like to know is about the infamous flyaways...

These issues apparently are present since the first Phantom... So what do we know about it?


What causes it normally?
Are cell towers a problem? Or maybe 5.8ghz from newer wi-fi routers?
What exceptions are made for it to happen?
Will a Futaba transmitter cure it?
In case it happens switching to manual mode will get you back control?

There are many experts on this forums... I hope we can get enough info to put a sticky for everyone else
to avoid flyaways...

Thanks!
JC
 
I don't know too much about it since I don't own one yet, but I've been lurking enough to take a stab at it. And besides, this is the internet so I can play "expert."

My understanding is it's due to signal interference. The stock transmitter has a weaker signal compared to the Futaba or some other, nicer-than-stock transmitters. The weaker signal is vulnerable to interference while the Futaba and others can power through it. That's really just a guess based on some comments that nobody has seen a flyaway occur when using a non-stock transmitter. Seems to make sense though.

then again, I'm pretty sure some folks have had these happen in very close proximity to the transmitter so...I'm probably getting a DJI 550 instead!

PhantomFan and others also have a good idea, that some flyaways could be due to vibration causing solder joints and wires to crack/loosen up (especially the antenna receiver wire). Good recommendation is take the ship apart every now and again and check for loose wires and cracked solder joints.
 
First thing to do, is separate genuine 'flyaways' from craft acting in strange ways due to pilot error etc.

As a rule, loss of control signal, whether from faulty wiring or interference will cause erratic flying, crashing or or forced Return To Home.

Flyaways however unfold as a situation where the craft will, of its own accord, take off in a specific direction. The pilot will rarely be able to intervene. Even turning off the transmitter [to force a RTH] has no effect. I have seen some claims that a motor shutdown, both sticks back and together, might arrest the flyaway [causing it to land heavily], but I have my doubts.

The theory gathering pace as to the cause of genuine flyaways, is that under certain conditions the Naza will ignore any inputs and revert back to an inbuilt 'home location' This default location is in China!
 
I will just add to the comments already posted that:

1. Some feel that a true fly away is caused by interference from the TX/RX. While this is surely possible, I don't think it is correct because the Phantom/Vision is programmed to RTH if the signal from the TX is lost. You could theorize that maybe the signal from the TX is overridden by some interference which tells the model to go somewhere else, but even in this case, the model would eventually fly too far from the interference to have an effect, and then would RTH, right. Well this does not seem to occur.

2. Most believe that a true fly away is the result of an error in the Naza, the gps, the compass, or a combination of the 3. Switching to Manual mode on the TX has been shown to recover a fly away in progress, even if it resulted in a crash.

What to do is up to you. I think it is very unlikely that a true fly away will occur. But I think there is enough evidence that it does happen.
 
This may be completely out in left field, but could solar storms potentially cause a fly away?

Here's an excerpt from an article last year:

"How does this affect my GPS?: According to the Space Weather Prediction Center at the National Oceanic and Atmospheric Administration, GPS satellites could be disrupted by the radiation from the flares and increased drag from the additional solar activity could up the chances that your GPS will calculate that you’re in the wrong location."

Source - http://articles.washingtonpost.com/...s-space-weather-prediction-center-solar-storm
 
I'm sure everyone got to view this alleged P2V youtube video... if not, it's located here: http://www.phantompilots.com/viewtopic.php?f=7&t=3914

In any event, to support my theory (noted above), his test flight occurred on 10/8/13... and according to the Spaceweather.com website that day, it was noted:

"CME IMPACT, GEOMAGNETIC STORM: An interplanetary shock wave, possibly the leading edge of a CME, hit Earth's magnetic field on October 8th at approximately 2015 UT (1:15 pm PDT). The impact has sparked a minor (G1-class) geomagnetic storm that could intensify in the hours ahead".

Could it be that this caused the alleged fly away? or perhaps it's just coincidental... Interesting I think.
 
It's possible that there isn't a single cause of "flyaways" and the fact that they seem more common with Phantoms could simply be due to the large number of Phantom owners when compared to the DIY quadcopter user base.

With that said, I almost lost my Phantom during a flight on a windy day. I can't be sure what really caused it but my best guess is that my phantom entered an area of high winds at a higher altitude and the hover in place wasn't enough to keep in in place as it was quickly carried away with the sticks at neutral. I struggled to bring it back but it was hard to see which direction it was flying and as I hadn't yet enabled the IOC Home Lock/Course Lock, I couldn't be sure I wasn't making it worse. I ran across the park to get a better view and eventually brought it back to me. (EDIT: Mine was an unmodified Phantom 1)

What I learned:
* Learn to fly your quadcopter well -- don't assume that a GPS, Compass, and, Gyroscope will alone be enough to keep you out of trouble.
* Enable the IOC modes and practice using them.
* Enable manual mode and practice using it in a large open space with sufficient altitude to begin with. (be sure NOT to activate it accidentally; and for the record I don't fly in manual mode but I can if I needed to)
* Practice flying it back toward you from a distance without IOC when you can't see which direction it's facing.
(Example: when you pull the stick left and it goes right -- know exactly which way to move the stick next to bring it towards you)
* Label your phantom with your phone number and mention a reward (apparently people DO return them)
* If you are so inclined, paint your Phantom such that you can see it's orientation better from a distance.
* Use caution when flying in the wind -- assume high currents could be present at higher altitudes.


In any case, I really like how this site is collecting user reports of flyaways in an attempt to narrow in on a solution:
http://flyawayclub.com/
 
rbarghini said:
This may be completely out in left field, but could solar storms potentially cause a fly away?

Ummmm...no. You are reading far too much into the effects of solar irradiance on low Earth satellites. The influence of solar phenomena over GPS satellites has a completely different mechanism than you envision.

What typically happens is as follows: When solar activity is high solar storms and other phenomena like coronal mass ejections heat up the upper atmosphere by a tiny increment proportional to the extra energy received. The atmosphere (being a gas, albeit a tenuous one at that altitude) expands ever so slightly, and reaches further out into space where the increased atmospheric density ever so slightly increases the drag on ANY satellite in low Earth orbit. This retards its position slightly over predicted. For GPS satellites, this orbital retardation results in a timing error which translates into a POSITIONAL error on Earth of a small magnitude. FAR too small to cause a flyaway since the signal is NOT interrupted whatsoever - only slightly degraded in accuracy, probably a few tens of feet in a "bad" storm. Consequently this phenomena is NOT the cause of Phantom fly-aways.

PF
 
I flew up to a cell tower and checked it out from about 20 feet away with no problems
 
PhantomFan said:
rbarghini said:
This may be completely out in left field, but could solar storms potentially cause a fly away?

Ummmm...no. You are reading far too much into the effects of solar irradiance on low Earth satellites. The influence of solar phenomena over GPS satellites has a completely different mechanism than you envision.

What typically happens is as follows: When solar activity is high solar storms and other phenomena like coronal mass ejections heat up the upper atmosphere by a tiny increment proportional to the extra energy received. The atmosphere (being a gas, albeit a tenuous one at that altitude) expands ever so slightly, and reaches further out into space where the increased atmospheric density ever so slightly increases the drag on ANY satellite in low Earth orbit. This retards its position slightly over predicted. For GPS satellites, this orbital retardation results in a timing error which translates into a POSITIONAL error on Earth of a small magnitude. FAR too small to cause a flyaway since the signal is NOT interrupted whatsoever - only slightly degraded in accuracy, probably a few tens of feet in a "bad" storm. Consequently this phenomena is NOT the cause of Phantom fly-aways.

PF

PhantomFan,

In any event, I had to put it out there... perhaps I've seen too many doomsday movies ;)

Thank you for the detailed response.

Ron
 
No sweat, Ron. I actually like the way you think. Unfortunately, the cause of Phantom fly-aways are probably multi-factorial. In some instances, pilot error, in others Tx to Rx interference. Still others are probably due to component failure - motor, ESC. or even prop. Sometimes I wonder (especially now that it is winter) whether a static electricity discharge onto/into a Phantom can upset the compass or some stored-on-board navigational algorithm or waypoint from a spark just from walking across the carpet in my home! Who knows??? <shrugs shoulders>

It's pretty hard to establish a causal connection without recovering the errant Phantom and without a "black box" which stores data on the incident. The closest thing to that is on-board video and audio in a bunch of instances - many of which are posted to YouTube. Eventually, with more people doing FPV and using/recording their OSD output, more objective data will become available and perhaps, in certain instances, some real insight into those particular fly-aways may become available.

Until then, your guess is as good as anybody's.

PF
 

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,359
Members
104,936
Latest member
hirehackers