Fly Away Phantom

Phantom 1 Chat, Queries and Advice.

Moderators: 1275mini, Richard

Fly Away Phantom

Postby Captain Tobias » Mon Oct 14, 2013 11:03 pm

This is bugging me as I now own a Phantom. Reading the various posts here, one gets the impression the Phantom has a tendency to "fly away" due to various reasons. The poor quality stock transmitter is one of many. I don't get it. If the Phantom is properly configured and all the pre-flight checks were done, it should go into "failsafe" mode soon after the transmitter signal is lost is it not? So, it should always return home, provided you haven't purposely flown it beyond the range of your battery!

OK. The new v4.0 firmware isn't exactly well written in the "Range" area. I set my right switch to GPS-ATT-Failsafe and a horizontal range of 20 meters. As predicted, my right stick was disabled and I could only increase or decrease throttle when my copter hit the 20 meter barrier. There was nothing I could do to bring back the Phantom except to switch to Failsafe. This work flawlessly as the copter found it's way back to the spot it was launched. Some people claimed that the Failsafe switch didn't work but mine did. If the Failsafe switch doesn't work, why can't you just simply switch the transmitter off?

I would like to hear more about "Fly Away" scenarios just so I can avoid them myself.
Captain Tobias
 
Posts: 66
Joined: Fri Oct 04, 2013 10:51 am

Re: Fly Away Phantom

Postby dhardjono » Tue Oct 15, 2013 3:23 am

ok i will start my story i just change to new landing skid made out of carbon fiber except for compass plate it look like hard cardboard type but i had it on attitude mode it was acting weird than switch to gps i flew into the sunset than return full throttle i stop pull it forward to stop but nope keep going hit hte ground and broke a small crack on my arris 2000 .So next day i try again with compass inside the landing skid and re calibrate i keep it low started run away again so now i switch back to original landing skid and no problem . http://youtu.be/rLeaA51Mn6Q this video show crash than after that was before the crash in attitude mode it was acting weird drifting when there no wind so i switch to gps causing the run away. oh i useT 8J RENOTE
let fly
User avatar
dhardjono
 
Posts: 91
Joined: Mon Jul 15, 2013 4:58 pm

Re: Fly Away Phantom

Postby G6EJD » Tue Oct 15, 2013 5:16 pm

Flyaway - there are many theories, mine is that it's down to pilot error in most cases, due to a lack of flying experience and orientation errors. Panic sets in, the sticks are pulled in all directions and before you know where you are it's gone, but not a flyaway!
User avatar
G6EJD
 
Posts: 3541
Joined: Thu Mar 28, 2013 11:29 am
Location: Bath, UK

Re: Fly Away Phantom

Postby Slartibartfast » Tue Oct 15, 2013 6:21 pm

I agree, most flyaways are pilot error. Or as we say in my industry 'picnic' problem in chair not in computer
Slartibartfast
 
Posts: 363
Joined: Sat Jun 08, 2013 4:29 pm
Location: High Wycombe, Bucks

Re: Fly Away Phantom

Postby Pufuleschi » Tue Oct 15, 2013 6:44 pm

I have experienced two types of almost "fly-away".

One seems to be related to the compass being very sensitive to potential magnetic interference. One time I was flying my Phantom near a small stones trail only to find itself having a mind of its own. I've managed to switch to manual and immediately land. After power-cycling the LED pointed out that compass calibration was required. I've tried calibrating in that area without success. Later I've found that the stones used on that trail were used in an iron melting plant and than I understood why my Phantom started to misbehave. Please note that I'm compass calibrating my bird before start flying it for the first time on each location. Iron from the rocks probably impacted the magnetic field used by the compass.

Another one is related to poor quality of the RX module. It seems to be very sensible to 2.8 GHz interference. I was on the beach in Thassos (Skala Potamia) and tried flying my Phantom several times during my stay. Above 10 meters in altitude my Phantom would start ascending on it's own and every-time I had to keep the throttle stick down fighting to get the Phantom back down. Looking through the area I've noticed a huge hub of microwave antennas - some used for GSM signal but some could had been for WiFi. I've noticed a similar behavior in other locations when I was near antenna towers.

I would not put the firmware v. 4.0 limit bug in the "fly-away" basket but I had my share of scary thoughts the first time I've tried that functionality only to find my Phantom drifting away on its own; away from what was supposed to be home location.

From a "human error" point of view I would suspect "fly-away" to happen due to IOC. If you play with the IOC control you can make the Phantom change its "home" location. Another possible fly-away reason could be related to bad GPS signal. If Phantom is loosing GPS signal in flight, my understanding is that it might drift in order to find satellites - though I would not understand why it would become unresponsive to commands and so it can be again a human error.

One thing I've learnt is that if my bird (my Phantom is now converted into a F450) would start acting weird I would switch to ATTI or Manual mode and land asap.
Pufuleschi
 
Posts: 29
Joined: Thu Oct 10, 2013 8:05 pm

Re: Fly Away Phantom

Postby The Editor » Tue Oct 15, 2013 8:11 pm

I agree with the comments in this thread - The vast majority of 'Fly Aways' can be attributed to pilot error.

The very small percentage of 'True' fly aways (if we can call them that) seem to be either linked to the very poor Phantom Rx which appears to be very susceptible to 2.4g interference or the well documented problems the the altitude barometer within the Naza when the sun hits the left side of the unit. (This is why the LHS of my Naza has black insulation tape applied to it !).

Nearly EVERY case of fly aways reported via the forums and Youtube etc seem to have one thing in common and that is stock TX/RX. even when you look through the Youtube recorded fly aways you usually get a glimpse of the operator/pilot on the ground at some point in the video and you can spot the tell tale white transmitter in their hands.
Self Built 12 foot Hex powered by 6 lawnmower engines, 36"props giving 14 days flight time and a range of just over 3,000 miles. Transmitter commands come from thought inputs connected to my frontal lobes via a colander on my head.
The Editor
 
Posts: 396
Joined: Tue Jul 02, 2013 7:39 pm
Location: Sort of in the middle

Re: Fly Away Phantom

Postby Avant_adam » Tue Oct 15, 2013 8:46 pm

In 6 months flying the phantom several times a week i've never had any occasion where i lost contol. The only occasion it did anything weird was in the house on the table one day after fitting something. I powered it up (motors spinning) but then it took off and was out of control. I grabbed it and then pulled the battery off incurring a few cuts on the forearms in the process. I attributed it to the house wifi.

I realise it'll now fly away on the next flight( hence why i tested it post V4 install with my flight case attached. All good, flew very stable).

I must be just lucky or away from eradio transmitters. I am careful where i fly.

Adam
Phantom V1 -flying for fun
Avant_adam
 
Posts: 312
Joined: Wed Apr 24, 2013 9:48 pm

Re: Fly Away Phantom

Postby Captain Tobias » Wed Oct 16, 2013 8:26 am

@The Editor:
What I don't understand is that if it was down to poor TX/RX, surely turning off the TX will bring back the Phantom in Failsafe mode? My understanding is that assuming that the home point GPS has been properly recorded (rapid green flashes), the Phantom will always switch to Failsafe mode if it looses contact with the TX?

@Avant_adam:
A silly question, did you turn your TX on before you plug the battery to your Phantom?
Captain Tobias
 
Posts: 66
Joined: Fri Oct 04, 2013 10:51 am

Re: Fly Away Phantom

Postby Avant_adam » Wed Oct 16, 2013 9:14 am

Always. Tx on then gopro then phantom. I turn the camera on so I am not working against the gimbal when it is powered up.

When I land it's the exact reverse.

Adam


Just a thought - you have got return to home set in the Naza assistant as the default option in case of lost reception? Thre are opportunities to turn this off... I guess you answered that in the first post saying RTH did work.
Phantom V1 -flying for fun
Avant_adam
 
Posts: 312
Joined: Wed Apr 24, 2013 9:48 pm

Re: Fly Away Phantom

Postby Captain Tobias » Wed Oct 16, 2013 10:08 am

Yes, I have set the thrid option on the right hand switch to Failsafe as I wanted to test it. I will be changing it to Manual as Failsafe can be achieved by turning off the TX.
Captain Tobias
 
Posts: 66
Joined: Fri Oct 04, 2013 10:51 am

Re: Fly Away Phantom

Postby G6EJD » Wed Oct 16, 2013 11:24 am

Good point, why enable fail-safe when TX off will do that.
User avatar
G6EJD
 
Posts: 3541
Joined: Thu Mar 28, 2013 11:29 am
Location: Bath, UK

Re: Fly Away Phantom

Postby The Editor » Wed Oct 16, 2013 11:39 am

Captain Tobias wrote:@The Editor:
What I don't understand is that if it was down to poor TX/RX, surely turning off the TX will bring back the Phantom in Failsafe mode? My understanding is that assuming that the home point GPS has been properly recorded (rapid green flashes), the Phantom will always switch to Failsafe mode if it looses contact with the TX?



Failsafe mode will depend on:
1. The Rx loosing connection with the TX (presumably loss of carrier wave and/or data stream/frames)
2. No Stick input for a given period of time

I believe what is happening with the 'true' fly aways is that the RX is being swamped by RF from somewhere and then giving erroneous data to the Naza which is interpreting it as stick inputs. The Naza will never go into failsafe as it 'thinks' it is getting commands from the RX (albeit full throttle or full pitch etc). The Naza is saying to itself "I'm ok, I'm doing what I'm being asked to do !
The other theory of course is that again, stray RF swamps the Naza and crashes the code running. Once the firmware has crashed it is then anybodies guess what the output will be to the esc's and ultimately the motors!
Last edited by The Editor on Wed Oct 16, 2013 12:00 pm, edited 1 time in total.
Self Built 12 foot Hex powered by 6 lawnmower engines, 36"props giving 14 days flight time and a range of just over 3,000 miles. Transmitter commands come from thought inputs connected to my frontal lobes via a colander on my head.
The Editor
 
Posts: 396
Joined: Tue Jul 02, 2013 7:39 pm
Location: Sort of in the middle

Re: Fly Away Phantom

Postby Pufuleschi » Wed Oct 16, 2013 11:43 am

Fail-safe triggers when the value for GPS/ATTI/Manual channel has a value outside GPS(~75%), ATTI (50%) or Manual (~25%). When loosing TX the value would automatically drop to 0%. NAZA needs about 3 seconds to "confirm" that TX was lost in order to switch to Fail-safe. RX would need another 3-10s to sync back with TX when TX gets powered back on.

I think there is a point on why set-up fail-safe instead of using TX off switch. Setting the ~25% value to fail-safe would simple enlarge the range of values that trigger fail-safe. Having the TX already connected when switching to "Fail-safe" would ensure that "valuable" time would not be lost.

If you really plan on flying in manual than you should probably consider upgrading from stock RC. I'm using a Futaba T8J and have the "Fail-safe" as a mixer triggered by a switch.
Pufuleschi
 
Posts: 29
Joined: Thu Oct 10, 2013 8:05 pm

Re: Fly Away Phantom

Postby Pufuleschi » Wed Oct 16, 2013 11:47 am

The Editor wrote:believe what is happening with the 'true' fly aways is that the RX is being swamped by RF from somewhere and then giving erroneous data to the Naza which is interpreting it as stick inputs. The Naza will never go into failsafe as it 'thinks' it is getting commands from the RX (albeit full throttle or full pitch etc). The Naza is saying to itself "I'm ok, I'm doing what I'm being asked to do !
The other theory of course is that again, stray RF swaps the Naza ans crashes the code running. Once the firmware has crashed it is then anybodies guess what the output will be to the esc's and ultimately the motors!


And that's another very good reason to switch to a different RX/TX system ;-)
Pufuleschi
 
Posts: 29
Joined: Thu Oct 10, 2013 8:05 pm

Re: Fly Away Phantom

Postby Captain Tobias » Wed Oct 16, 2013 12:55 pm

The Editor wrote:I believe what is happening with the 'true' fly aways is that the RX is being swamped by RF from somewhere and then giving erroneous data to the Naza which is interpreting it as stick inputs. The Naza will never go into failsafe as it 'thinks' it is getting commands from the RX (albeit full throttle or full pitch etc). The Naza is saying to itself "I'm ok, I'm doing what I'm being asked to do !
The other theory of course is that again, stray RF swamps the Naza and crashes the code running. Once the firmware has crashed it is then anybodies guess what the output will be to the esc's and ultimately the motors!


That's a very good point. I've heard DIY quad/tri copter owners complaining that flying their craft too close to telecom masts caused all kinds of problems. My Futaba has FASST technology which I believe would overcome such interference but I'm not sure if the cheap stock TX has similar?
Captain Tobias
 
Posts: 66
Joined: Fri Oct 04, 2013 10:51 am

Next

Return to Phantom 1

Who is online

Users browsing this forum: No registered users and 2 guests