DJI repair takes 8 weeks -- then it crashes again in two days

[catware]

My first crash occurred 5 days after purchasing the Inspire 1, and the repair was covered by DJI under warranty due to a firmware bug. It took 8 weeks to get repaired and returned by DJI. Then today, just two days after receiving it back, it crashed again for a different reason.

Incident 1: 15 March I took off from a parking lot on the edge of a 500' cliff and flew the aircraft down over the edge of the cliff, which caused control and video signals to cut out. I pushed the throttle forward, attempting to vertically raise the aircraft back into LOS to regain control. The aircraft went into RTH mode, and a firmware design at the time caused the preprogrammed minimum RTH height climb to be *cancelled* due to my vertical stick input, although the horizontal portion would still be automatically performed. The aircraft then flew itself horizontally into the cliffs at 100' below take-off elevation and crashed. Pieces were recovered and sent back to DJI. I was told that this firmware issue was fixed in the latest firmware update.

After 8 weeks, the Inspire 1 was finally repaired and shipped back to me. I updated the firmware on the aircraft and remote, which were not up to date on the aircraft when DJI returned it to me. I calibrated the IMU and compass before the first flight, and did not recalibrate either for any flights afterward. I then performed 7 flights from my property in the last couple days, averaging about 30,000 ft travelled each flight. I have a big smile on my face as it seems to be performing well.

Incident 2: 19 May Today I took off from the same location and flew out to 1 km away. I noticed the horizon in the camera view seemed tilted. I took my hands off the sticks and the aircraft was clearly still flying on its own (i.e., not just hovering). I quickly decided to get it back fast, and was able to fly it back to take-off point despite this problem. But hovering with my hands off the sticks, it would fly tilted at about a 45 degree angle while slowly turning. I lowered the landing gear and was able to do a controlled crash into the field by my house -- it was too difficult to control with the constantly shifting stick offset to land controllably. Luckily there was no obvious damage to the aircraft. The camera was ripped out of its rubber mounts and the cables disconnected, but this was easily put back together.

In the flight playback on the DJI pilot app, it is easy to see the aircraft flying itself, with no stick input being given while it is moving. I have read of several other people having this issue. I am not sure how to prevent this second problem or what caused it, and it has resulted in a very low confidence level for this product.

 

[catware]

I will add that I also have the Phantom 2 with Lightbridge installed, and with over 200 flights on it in extremely challenging mountainous terrain I have never had a single issue with it not performing perfectly.

 

[Shazbot4]

so you break LOS on the first one, bet you still had the antennas pointed in the up position when you went below your take--off altitude instead of pointing the either horizontally or down toward the aircraft.
there is a couple pages in the manual that talk about the range of the remote and how it works. looks like after your second crash you still didn't read any of it.

so you get it back and figure... 8 weeks wait time... what the hell lets take this thing 1k out just for the heck of it?!?! who was paying you to film something over 3/4 of mile away and why on earth would you take that risk?????

you deserved what you got buddy thank your lucky stars you didn't break it AGAIN. have you flown something other then a DJI product? fixed wing aircraft? a helicopter? i dont think you have because your doing things any experienced r/c pilot wouldn't consider doing unless their drunk, or lot of money on the line for a bet.

DJI's flight software does 99% of everything needed to keep it's self in the air and still people like you go out of their way to push the limits and crash it. i guess 3-4000 dollars isn't a lot of money to you because what i read on your post you sound like you really don't care about what your doing.

go to a r/c field and talk to the people there and maybe learn something before you do something really stupid, ground your inspire until then.

 

[amkorp]

^ what he said.

You dont sound like you know what you're doing.

What were your mod values before you took off for that flight. I'd bet my left nut that they were out of the acceptable range and probably required a compass calibration and maybe even an IMU calibration prior to the last flight where you "control crashed" your aircraft because you couldn't land it safely while it was drifting.

I also would bet my right nut that you didn't check the mod values before taking off.

Read the manual, go through this site and read the help and suggestions provided. Practise in atti mode in preparation for times when you dont have gps signal so you dont have to "control crash".

 

[pops52]

Where's the love?

 

[catware]

Incident 3: 19 May I re-calibrated the compass, and made another flight in which all went well. Then I made another flight (the 3rd of the day), just flying around in the field in front of my house. Again (as in Incident 2) the control went haywire -- the aircraft would actively attempt to fly away in a direction on its own. However I was able to counteract with moderate (not slight) stick input. The direction it wanted to fly would shift repeatedly -- one time it would fly in one direction, then I'd stop it by finding the spot to hold the stick off center, then I'd release the stick, and it would try to fly off in a different direction (it would change every time). After 4 or 5 times like this, with me continually readjusting in this manner, I was able to bring it down and land in the field.

If you saw the video a few months ago of the person who took off and his Inspire 1 flew headlong into the garage door, this is what it was like. Several other people have reported this same issue. This is a fast aircraft and easy to get away from the pilot if it flies on its own without pilot stick input.

This is really unfortunate because it appears there is a persisting problem in the airframe. If it's software related then I could address it on my end, but if this is hardware it will have to go back to DJI. The previous repair experience was very poor -- despite me calling several times, no information was provided by customer service until the email "it's been fixed!" right on the last day of their 6 week repair timeframe guarantee. Then it sat for nearly two weeks in their shipping dept until they got around to sending it back to me.

If anyone else has suggestions I'm open to hear them. I'll address the two people that already replied in my next post.

 

[catware]

@Shazbot4 : I am actually an engineer, and I have posted my flight resume earlier in this forum, which is extensive. I also reiterated in the post directly above yours that I have over 200 flights on the Phantom 2. As I wrote above, Incident 1 was covered by DJI under their warranty, i.e. they admitted it was a firmware design bug on their part. Furthermore, they took the effort to address and fix this bug in a later firmware release. So you going on trying to find blame in what I did is a rather moot point. And the 1 kilometer away I was in Incident 2, did not play a factor in the incident itself, in addition to 1 km being well within the stated design specs of the aircraft.

I purchased the P2 and now the I1 in order to scout new canyoneering routes high in the mountains. These new prospects are difficult to get a team together to descend without advance information provided by these aircraft. And indeed in one instance the recon saved our team from entering a narrow slot with a high water situation. But this is well beyond the scope of this forum -- I'm only stating my purposes for flying 1 km away or more. Earlier this year I posted a YouTube link to a recon video that I did.

@amkorp : I have a screenshot of the mod values after the Incident 2 flight, before powering down the aircraft. I also have the mod values of the flight immediately prior to Incident 3, in which the phenomenon happened again. Can you tell me what values I am looking for? There is nothing in the DJI Inspire 1 manual that states we as pilots need to check any values before flight. And if these values are fluctuating enough to produce an uncontrollable situation from one flight to another, this is a clear error in the design of the system; the onus is not on the end user.

Also, just to be clear I normally fly in ATTI mode. I only use GPS mode when I want it to hold its position. I am very familiar with the flight characteristics between the two modes. This is not what was happening when the aircraft. As I mentioned above, the horizon was tilted in the display which was my first clue that something was going wrong.

 

[JimPA]

Catware, take a look at this thread http://www.inspirepilots.com/threads/a-bit-of-a-scarry-launch-today.2856/

I had a scary moment and in the end it was 100% my fault for NOT checking my MOD values before flight. Read the whole thread especially the part where I went back a few days later and took MOD readings that were WAY out of spec.

But, at the end of the day, what I experienced with lack of control and having to adjust to keep the AC in position sounds just like what you're saying.

DJI has went to great levels to make these things so easy to fly, but we easily forget or in many cases just don't understand how they really work and how important checking the MOD values is. I remember reading about someone who launched from the back of a pickup truck and the Inspire piled itself into a nearby field. Launching from a metal bed of a truck is probably not a good idea.

My first 70 or so flights I NEVER read my MOD values. After the one bad launch I had changed my mind on that. I now check them EVERY time. You can never tell what may be underground where you launch. I'm sure even out in the middle of a field, there could be an iron deposit.

Moral of the story, ALWAYS check them before each flight.

 

[catware]

Thanks @JimPA , I'll read through that thread. There's so much traffic on this forum these days it's hard to catch important thread like that

 

[sdjackaerial]

Where's the love?

Seriously. I guess it's fun to *try* to make someone feel dumb. And I did say TRY.

 

[ISP5557]

Very suspicious three flights like this. What electronics are you carrying on your person? Bluetooth on your phone? This seems like a localized issue. I could be wrong though.

 

[sirnikolas]

so you break LOS on the first one, bet you still had the antennas pointed in the up position when you went below your take--off altitude instead of pointing the either horizontally or down toward the aircraft.
there is a couple pages in the manual that talk about the range of the remote and how it works. looks like after your second crash you still didn't read any of it.

so you get it back and figure... 8 weeks wait time... what the hell lets take this thing 1k out just for the heck of it?!?! who was paying you to film something over 3/4 of mile away and why on earth would you take that risk?????

you deserved what you got buddy thank your lucky stars you didn't break it AGAIN. have you flown something other then a DJI product? fixed wing aircraft? a helicopter? i dont think you have because your doing things any experienced r/c pilot wouldn't consider doing unless their drunk, or lot of money on the line for a bet.

DJI's flight software does 99% of everything needed to keep it's self in the air and still people like you go out of their way to push the limits and crash it. i guess 3-4000 dollars isn't a lot of money to you because what i read on your post you sound like you really don't care about what your doing.

go to a r/c field and talk to the people there and maybe learn something before you do something really stupid, ground your inspire until then.

^ what he said.

You dont sound like you know what you're doing.

What were your mod values before you took off for that flight. I'd be my left nut that they were out of the acceptable range and probably required a compass calibration and maybe even an IMU calibration prior to the last flight where you "control crashed" your aircraft because you couldn't land it safely while it was drifting.

I also would bet my right nut that you didn't check the mod values before taking off.

Read the manual, go through this site and read the help and suggestions provided. Practise in atti mode in preparation for times when you dont have gps signal so you dont have to "control crash".

Can you 2 please be a little nicer to our members! Jeez!

 

[catware]

A picture might help. Here is a screenshot of the Incident 2's flight playback (what a fantastic feature by the way). You can see that there is no stick input, and it is flying at 24.6mph in a large circle. The playback shows that the aircraft was pointed backward, but actually it wasn't -- it was pointing forward. Where the path curves to the left again and goes off the bottom of the image is where I retook control and flew it back to base. During the rest of the flight playback, again the data shows the orientation was backward, but it wasn't -- camera and aircraft were facing forward. This error occurred midway through the flight and is when I noticed the horizon had tilted -- the first minute or so has the orientation correct.

As far as Mod values, the compass had Mod 1475 after I controllably crashed it in Incident 2 before I powered it off. It had 1527 after doing a recalibration, and 1480 just before takeoff on Incident 3. Interestingly though, the compass X value was -500 to -600 adjacent to both incidents, but +479 on the middle flight that had no problem. I will monitor that value in the future to determine if it's relevant.

Inspire1Loop

• catware
• May 21, 2015

 

[jon b]

A picture might help. Here is a screenshot of the Incident 2's flight playback (what a fantastic feature by the way). You can see that there is no stick input, and it is flying at 24.6mph in a large circle. The playback shows that the aircraft was pointed backward, but actually it wasn't -- it was pointing forward. Where the path curves to the left again and goes off the bottom of the image is where I retook control and flew it back to base. During the rest of the flight playback, again the data shows the orientation was backward, but it wasn't -- camera and aircraft were facing forward. This error occurred midway through the flight and is when I noticed the horizon had tilted -- the first minute or so has the orientation correct.

As far as Mod values, the compass had Mod 1475 after I controllably crashed it in Incident 2 before I powered it off. It had 1527 after doing a recalibration, and 1480 just before takeoff on Incident 3. Interestingly though, the compass X value was -500 to -600 adjacent to both incidents, but +479 on the middle flight that had no problem. I will monitor that value in the future to determine if it's relevant.

Inspire1Loop

• catware
• May 21, 2015

Thanks for posting and updating your experiences. could there be magnetic issues at this flight location? Have you moved to another open place and repeated the process you follow. Seems like the compass?

 

[YeeaaBoii]

Honestly, that first crash sounded pretty bad considering you had to gather up a bunch of pieces to send back to DJI to fix. Even if DJI successfully "fixed" it, I would never trust that aircraft again. You should have had it repaired and sold it upon getting it back but obviously noting in the description to the buyer that it had been crashed and repaired by DJI. Lose a few hundred bucks and get a brand new one.

 

[pops52]

Seriously. I guess it's fun to *try* to make someone feel dumb. And I did say TRY.

My reply was directed to the two posters below catware's initial post. I thought they were a little quick to judge catware's experience.

 

[Shazbot4]

@Shazbot4 : I am actually an engineer, and I have posted my flight resume earlier in this forum, which is extensive. I also reiterated in the post directly above yours that I have over 200 flights on the Phantom 2. As I wrote above, Incident 1 was covered by DJI under their warranty, i.e. they admitted it was a firmware design bug on their part. Furthermore, they took the effort to address and fix this bug in a later firmware release. So you going on trying to find blame in what I did is a rather moot point. And the 1 kilometer away I was in Incident 2, did not play a factor in the incident itself, in addition to 1 km being well within the stated design specs of the aircraft.

I purchased the P2 and now the I1 in order to scout new canyoneering routes high in the mountains. These new prospects are difficult to get a team together to descend without advance information provided by these aircraft. And indeed in one instance the recon saved our team from entering a narrow slot with a high water situation. But this is well beyond the scope of this forum -- I'm only stating my purposes for flying 1 km away or more. Earlier this year I posted a YouTube link to a recon video that I did.

@amkorp : I have a screenshot of the mod values after the Incident 2 flight, before powering down the aircraft. I also have the mod values of the flight immediately prior to Incident 3, in which the phenomenon happened again. Can you tell me what values I am looking for? There is nothing in the DJI Inspire 1 manual that states we as pilots need to check any values before flight. And if these values are fluctuating enough to produce an uncontrollable situation from one flight to another, this is a clear error in the design of the system; the onus is not on the end user.

Also, just to be clear I normally fly in ATTI mode. I only use GPS mode when I want it to hold its position. I am very familiar with the flight characteristics between the two modes. This is not what was happening when the aircraft. As I mentioned above, the horizon was tilted in the display which was my first clue that something was going wrong.

read your reply completely and im sticking to my original opinion of you. 200+ flights on a phantom gives you zero credit as a r/c aviator. i don't care if you wrote the software for the f-35 fighter jet. you really don't have a clue what your doing and reading more reply's from you confirms that.

your going to take a p2 and "scout" ahead before you fly your inspire on that route? lol ok. anyone who defends you as being a competent inspire pilot is as clueless as you if they think that is necessary.

here is what you should do. walk into a hobby store (this will probably be your first time) and buy helicopter kit, one you have to put together and program the servos and transmitter and trim, the whole smack. then fly it and crash it about 30 times and fix it. don't come back till you post a video of you doing full 3d flight with it successfully. then i want you to think about how you fly that aircraft and compare it to what your doing with your inspire, what you will figure out is you wont do 90% of the things with your helicopter that your inspire does without any practice or effort, then your going to think back about how you were flying your p2 and I1 and then understand.. oh now i see what shazbot was talking about, i should be more careful.

nothing on any of your reply's or statements i read takes accountability. your blaming your crashes on everything and everyone other than yourself. software, equipment... everything but YOU! your ignorance, pride and vanity is putting your inspire in situations where it's probably going to crash. you dont know anything about how your inspire works and until you figure it out im positive im going to see "i crashed my inspire for the 3rd time first day after 6 week repair" real soon.

 

[turbodronepilot]

read your reply completely and im sticking to my original opinion of you. 200+ flights on a phantom gives you zero credit as a r/c aviator. i don't care if you wrote the software for the f-35 fighter jet. you really don't have a clue what your doing and reading more reply's from you confirms that.

your going to take a p2 and "scout" ahead before you fly your inspire on that route? lol ok. anyone who defends you as being a competent inspire pilot is as clueless as you if they think that is necessary.

here is what you should do. walk into a hobby store (this will probably be your first time) and buy helicopter kit, one you have to put together and program the servos and transmitter and trim, the whole smack. then fly it and crash it about 30 times and fix it. don't come back to you post a video of you doing full 3d flight with it successfully. then i want you to think about how you fly that aircraft and compare it to what your doing with your inspire, what you will figure out is you wont do 90% of the things with your helicopter that your inspire does without any practice or effort, then your going to think back about how you were flying your p2 and I1 and then understand.. oh now i see what shazbot was talking about, i should be more careful.

nothing on any of your reply's or statements i read takes accountability. your blaming your crashes on everything and everyone other than yourself. software, equipment... everything but YOU! your ignorance, pride and vanity is putting your inspire in situations where it's probably going to crash. you dont know anything about how your inspire works and until you figure it out im positive im going to see "i crashed my inspire for the 3rd time first day after 6 week repair" real soon.

Hey Shazbot4. .
You may or may not be correct in your assessment of catware,but the way you thrashed her I don't appreciate. .
in fact if you don't have anything positive or productive to say then keep it to yourself. .
turbo. .

 

[Shazbot4]

Hey Shazbot4. .
You may or may not be correct in your assessment of catware,but the way you thrashed her I don't appreciate. .
in fact if you don't have anything positive or productive to say then keep it to yourself. .
turbo. .

read the Manuel
learn to fly other aircraft
talk to other r/c pilots
go to a hobby store
go to a r/c field
dont take stupid risks
dont push the limits of the aircraft and it's software
learn how the inspire works first

your right, no good advice there

the guy is looking for excuses every where for why is inspire goes nuts except him self. he needs his inspire taken away before he gets us all in real trouble.

 

[turbodronepilot]

read the Manuel
learn to fly other aircraft
talk to other r/c pilots
go to a hobby store
go to a r/c field
dont take stupid risks
dont push the limits of the aircraft and it's software
learn how the inspire works first

your right, no good advice there

the guy is looking for excuses every where for why is inspire goes nuts except him self. he needs his inspire taken away before he gets us all in real trouble.

Ya man..
Well this is a lady you're talking to
not a guy..
I was hoping you would get what I'm saying to you but apparently you dont..
yes you gave some good advice but you also insulted her more than once..
turbo. .