CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – Mike

[rilot]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

been doing this heli thing for a while now

Why do people always say something along these lines before stating something that's either misunderstood, or just plain wrong?
I've been flying RC helis since the days of mechanical tail mixing but I never feel the need to tell anyone in order to justify what I say.
In fact, I think that's the first time I've ever stated it on a forum.....

Anyway, as said before by myself and others, you have a configuration issue. Either with stick calibration, stick trims on the throttle, failsafe bind or centre point. No one else can reproduce your issue.

 

[LeoS]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

http://www.rcgroups.com/forums/showthread.php?t=2089211

First two posters zero-ed in to the RAP feature.

 

[750r]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

I don't get why your getting mad ? Seems everyone so far has been trying to help seems that you need to check all your settings . I have 2 phantoms a 1.1.1 and a P2 they both shut down it I leave it sit the P2 was just opened so I could try to get it to act like yours did but it just shut down to . I have been on a few forums and this seems to be extremely rare if you are calibrated correctly . I think if your hell bent on the naza being the problem then maybe you should just move on and find another bird . No reason to get hostile when people are trying to help .
GL !

 

[DeweyAXD]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Hey EL - Look I already seen a post where you made an idiot out of yourself regarding this same matter where another guy was having the same problem while in hover.

Read My lips -

Regardless - no matter how you twist it or what the problem is - the helicopter is; "TAKING OFF BY ITSELF" and going into a flight mode which can result in serious injury.

Bottom line is; the RAP feature needs to be disabled when the machine is on the ground - PERIOD!

Mike your attitude on this is a shame. Everyone in here is trying to help you with something that we all see as being an issue with your individual setup. Whether one day of experiance or 20 years worth things happen that sometimes look like a big problem can actually blind side us from it being something simple.
As I stated before I have the upmost respect for highlighting an issue if it is there. It may well exist but I for one (and I'm sure I speak for the many that read this without reply) believe that this is a TX calibration issue on your individual setup and without seeing a video of failsafe activating with the single flashing atti light (suggesting your TX is at dead stick and complying with the NAZA's RAP rules) then I can only assume it is indeed just that, a setup issue.

If you care enough about highlighting the dangers and this does happen at true 'single flash dead stick' then it is imperitive to demonstrate it visually so others can take heed. Even if you did the TX calibration and it didn't fly off it is still a worthy post and a subject worth highlighting. An experianced user demonstrating a problem that took them by surprise is more valuable than most.

I really hope you demonstrate this to us all. I can't speak for others but a video of the single flash working as it should do would in no way be a deemed a sign of incompatancy on your part.... just a sign that you do care about highlighting the dangers to the community and that you are happy to admit a mistake (that could easily happen to others).

Without the video of this then anything else at this point will, writely or wrongly, be deemed as a demonstration of dented pride/ego for not diagnosing it correctly before posting.

One way or another a prop in the face is no laughing matter and I for one will be looking harder now for that single flashing light and stopping immediatley if i see double flashing on idle in the field. If you show me it happens on a single flash I will seriously consider disabling RAP (and spreading the word a bit) but until then I can only go what I see and hear from others.

Cheers

 

[ElGuano]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Hey EL - Look I already seen a post where you made an idiot out of yourself regarding this same matter where another guy was having the same problem while in hover.

Read My lips -

Regardless - no matter how you twist it or what the problem is - the helicopter is; "TAKING OFF BY ITSELF" and going into a flight mode which can result in serious injury.

Bottom line is; the RAP feature needs to be disabled when the machine is on the ground - PERIOD!

Edit: I'm out, not going to respond to ad hominem attacks. Mike, if you decide to respond substantively to the facts or the methodology/conclusions of the tests, I'm happy to talk.

 

[Mike Mas]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Hey Guys, use the information as you will - I put this out there to help avoid an accident to others that almost happened to me. Rather than take it in consideration as a safety measure - some “keyboard jockey’s” choose to beat me up rather than address the real problem.

If it saves one guy from getting hurt, then I accomplished what I wanted to do. I didn't manufacture this thread and incident, it really happened and I provided a video to show it’s valid. I was on a shoot at a populated area, and as I was looking at the monitor OSD specs, it happened to me and in fact will happen to others – As I mentioned, there are posts from other pilots who experienced this same fault.

The fix is simple- all DJI has to do is use the same firmware routine they use for Home & Heading lock - once the machine is on the ground or if its closer than 10 meters from the home point, the RAP feature & failsafe is disabled.
No matter how you look at it, analyze it, or shape it - there is “Zero” reasons why any machine under any circumstance should lift off unexpectedly.
Fly-Safe!

 

[ElGuano]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Mike,

As long as we're staying on-topic, you've mentioned multiple times that this behavior "also happens with single flashes." I'd like to repeat the multiple requests from multiple users that you show us a video of this happening. It would unequivocally prove your point that something is wrong, and rebut the standing assumption that your TX calibration is at fault. Without it, the only conclusion a reasonable person can reach is that unexpected or not, it's a consequence of user error (improperly configured equipment). Don't you agree that's worth addressing?

 

[ianwood]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

I'd bet 9 out of 10 new Phantom owners are not aware of this risk. And I'd bet half or less are ever checking the calibration of their sticks. Finally, many people might assume the Naza is smart enough to know when it is in the air vs. on the ground. It would seem this is not the case. This all comes down to the need for better documentation.

So if for nothing else, I appreciate the post and effort to duplicate the condition as I am now aware that I should check my sticks more often, make sure the trims are zeroed, etc. And to take off once I get the motors going!

 

[james4u007]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Thanks for the post. I just bought my Phantom Vision and have been worried about the fly away problems so I have been reading as much as I possibly can. I haven't even flown my Phantom yet because I have been reading so much valuable information that has been posted here at Phantompilots.

I have over 100 flights on my Blade 350 QX and never had one issue. My 350QX will not initialize the motors if all sticks are not centered (I guess thats what you call it sorry I am new).

Even though this poster has failed to give us a video of this happening with one blinking light, I am grateful that I was able to learn something from the others here that not having the controls centered could cause this. Its something I would have never thought of considering my 350QX does not have this failsafe built into it and even if it did it won't let you take off unless everything is centered.

Thanks to the original poster and everyone who gave their input to try to help him.

 

[Mike Mas]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Thanks for your positive comments guys - my only concern was the safety of others!

 

[ElGuano]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

It's a credit to this community that people can keep their comments positive! Kudos guys.

 

[Mike Mas]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

I didn't start the thread looking for help. I'm well aware of what the problem is. I happen to have no less than 6 Naza equipped machines. I'm more than familiar with stick centering and the reaction thereof – this was never the point of the post.

I started the thread for the sole purpose of safety reasons, to make others aware this was a unknown potential danger and a safety issue that no one has addressed.

Regardless of how negative some of you guys have been to this alert - no one in this industry is going to prove to me that there is any reason a the RAP to be reactive while the machine is on the ground and cause an unexpected lift off that could injure others.

I repeat – the simple fix to avoid a possible injury is for DJI to use the same firmware routine they use for Home & Heading lock - once the machine is on the ground or if its closer than 10 meters from the home point, the RAP feature & failsafe is disabled.

No one needs the machine to climb to 60 feet then re-land when it on the ground already!!!

 

[LeoS]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

@Mike: You might want to reread your initial post and correct the inaccuracies in it so we have a good problem report and warning then.

1. The first line (advisory) paints a broad stroke that all NAZA V2 equipped crafts will experience this issue with no exception.
2. No correction of the trim issue was mentioned.
3. There's still no mention to RAP feature for the user to look out for.

As it's written now, I have to agree that it is in fact still a FUD piece. It will spread fear among the users based on misinformations.

I agree that there are some default behaviours on NAZA that DJI should be able to improve upon, but how can they take it seriously if the reports are not based on facts?

Be advised if you are using the Naza V2- GPS your machine will take-off with no prior warning - which could result in an accident or serious injury.

Problem: DJi’s new “Non-Signal” firmware which monitors the receiver for a “No Signal” status them automatically engages failsafe has no priority to lock out this feature while on the ground.

Event: Pilot engages and starts the motors – if the pilot does not move the cyclic yaw or throttle sticks for a short period of time - the Naza instantly enters “fail-Safe” mode and in a few seconds goes to full throttle in an attempt to reach failsafe altitude therefore causing the pilot to think the machine is in a fly-away state.

My Event: I set the machine on the runway spooled up the motors, after a few seconds, I walked over to my FPV monitor to check the readout battery voltage - as I began to walk back to the machine, the multi blasted off the ground – As I held low throttle in hopes of reducing the accent - I switched to failsafe with no results - next I flipped to ATTI to dump the GPS and my machine went to low throttle tumbled out of the sky, unknowingly, my 40+ years of reflex instantly added throttle in time to prevent the motors from shutting down and it nearly hit my truck and Linda as it sped back up to the ground.

I recommend that anyone using this Naza be aware of this error and take precautions when on the ground and spooling up the motors – remember if your in GPS mode you’ll have to move to ATTI to dump the fail-safe routine already started. DJI needs to address his problem with a update to prevent fail-safe engagement while the machine is on the ground.

Fly Safe!

Mike Mas

Bench test of error video attached


http://youtu.be/03VRozck72c

 

[mbuk]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Well, I've read and reread this thread and tried to look at it from a logical perspective.

I also debated in my head whether to post a reply but I felt the more info added the better.

I've have tried to replicate this too and have had no success at all, the phantom behaves exactly as it should and as per the manual. I too am using the stock DJI controller.

So as I see it:

Until anyone (Mike or anyone else) can demonstrate the behaviour on single flash (sticks centred) I can only conclude that there is not a problem and I would expect DJi to take exactly the same view.

Mike, the video you currently have posted only proves that when you think you have centred sticks on your Tx - the phantom doesn't think you have - you are correct that this is dangerous. But it is not a widespread fault with the Naza or the Phantom. It may be calibration or an isolated fault with your set up.

Mike, with respect, if you are experiencing the behaviour on single flash, as you say, you are actually doing the opposite of your objective. By not posting a video to prove it you are risking people's safety as the error will not be acknowledged as genuine and DJi will absolutely not do anything about it.

Irrelevant of if the video you hopefully will post shows the correct or incorrect behaviour on single flash, I am really glad you did start the thread as it has highlighted some very important points which I am sure will help keep people safe, but as yet we have not got to the root of the issue which is vital to everyones safety. As you are the only person that appears to be able to demonstrate I think we would all urge you to do so.

 

[FatKevinSmith]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

No one needs the machine to climb to 60 feet then re-land when it on the ground already!!!

While that's true, you have to appreciate the nature of how programming works. The Phantom really has no idea of its relation to the ground. It does not have proximity sensors. So all it knows is that in order to avoid potential vertical obstacles between it and "home" it should go 60' up from where it thinks the ground might be, get its bearings, and proceed to the x,y home coords, then descent until the z coord stops changing.

Because the phantom will not descend FIRST during a RTH (because if it landed on a roof, you'd kind of be hosed), it will never know where the ground is, and the only logical thing it can do first is ascend to 60'. At that point, it assumes that it would have good LOS to the GPS and away from most EMI, where it can attain an accurate bearing.

Hope this helps.

 

[DeweyAXD]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Agreed with all these positive comments... it has at least open all our eyes to the possibility of this happening with a TX ever so slightly out of calibration. Soooo easy to knock the trims on non stock TX's too.

 

[Mike Mas]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Thanks for the replys guys - here's what I'm thinking - is I have a subtrim value right on the threshold and it changes from one flash to two - perhaps at a single moment the Naza see's the two flashes and goes into the failsafe mode.

Here's the way you can re-create this error:

1) Plug in your machine and let it get happy! (one flash / gps or ati)

2) Go to any one of your "cyclic" sub-trims (Naza don't not care where the throttle is positioned) and move say the aileron or elevator sub trim up or down from center and the single flash will become two flash's - this is the Naza telling you its not happy any more and the machine is not trimmed.

3) Weight the machine down or pull the props and start the motors (hands off the trans sticks) and in 30 seconds she will go into failsafe and take off.

Events - The Naza thinks you’re flying around with the expected two grn or yel flash's during stick movement – then all of a sudden there is no stick movement from the pilot which starts the clock ticking. Next, the Naza activates failsafe to bring your pup back home.

This is exactly what is happening on the ground or in hover should the trims not be calibrated. It is not that unusual for a pilot to be playing with his transmitter and change a sub trim in error to create an event that could cause serious injury. I have two separate Naza's doing the same exact thing.

A simple fix to avoid a possible injury is for DJI to use the same firmware routine they use for Home & Heading lock - once the machine is on the ground or if its closer than 10 meters from the home point, the RAP feature & failsafe is disabled.

Enjoy!

 

[stevefly]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

I flew my Vision 2 successfully for the first time, landed and shut rotors off by pulling left stick all the way down and holding it down. I turned my back to get something and my wife yelled "It's taking off by itself." I couldn't get control so pulled S1 down and then had control for a soft landing. I never figured out what happened.

 

[ElGuano]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

2) Go to any one of your "cyclic" sub-trims (Naza don't not care where the throttle is positioned) and move say the aileron or elevator sub trim up or down from center and the single flash will become two flash's - this is the Naza telling you its not happy any more and the machine is not trimmed.

Events - The Naza thinks you’re flying around with the expected two grn or yel flash's during stick movement – then all of a sudden there is no stick movement from the pilot which starts the clock ticking. Next, the Naza activates failsafe to bring your pup back home.

A simple fix to avoid a possible injury is for DJI to use the same firmware routine they use for Home & Heading lock - once the machine is on the ground or if its closer than 10 meters from the home point, the RAP feature & failsafe is disabled.

It sounds like it was a tx issue after all, glad we could get that cleared up! The broader implication of this development is--if you're using a tx function that sends a non-zero input when your hands are off the stick, you are risking losing control of your copter at any time, on the ground or in the air (anytime you hover). If you use a subtrim in this way, the best thing to do is simply to disable RAP, as it is fundamentally incompatible with how the feature operates (it is a time bomb waiting for you to go hands-off at any point for 30 seconds).

Merely geofencing RAP wouldn't stop it from "unexpectedly" activating at any other time.

 

[mbuk]

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

Merely geofencing RAP wouldn't stop it from "unexpectedly" activating.

Good point! I'd also guess that DJi would have limited interest in trying to find any solution as their stock Tx doesn't have any subtrim adjustment so no issue.

Do you guys find you use the sub trim with the Phantom? I do with my Heli's but never missed it on the phantom.

If you don't, are you able to disable the subtrim in the Tx setting when assigned to the phantom?