Drone smashed into the ground - why?

Hi everyone

I'm new to this and have a Phantom 4.

I used RTH and the P4 came back to me and started landing at 60M in the air. I saw a guy with a dog approaching so I cancelled the auto landing at approx 30M. I started to lower the drone manually and at approx 25M it started going down, FAST.

I pushed the stick upwards and got no response. It smashed into the ground and bounced back up, at which point I was able to control it again, bringing it to a safe landing.

However, a piece broke off and a component was hanging out of one of the legs. I've put it back in and everything works fine, but it's been making a creaking noise since when I touch it, and the piece has been trying to fall out again.

WTF happened here? I've reviewed the logs and it just cuts out at 25M to the ground and doesn't show anything after that.

I have the video footage of it hitting the ground which has me partially in shot and I can see that I wasn't even on the throttle so wasn't me that smashed it into the ground. Bit strange that the DJI Go app has no info after it was at 25M in the air, isn't it? Does that suggest a Phantom problem to you?


something similar happened to me with my P2V+ a long time ago, i hit RTH when about 30 to 40 ft up i canceled RTH but still came screaming down very fast , fortunately it responded to the ascend command other wise it would have come down pretty hard and fast, wierd how it continued on its downward path at high speed. Not sure why.
 
Have you checked if altitude readings are correct? Known case in context of RTH where altitude reading is higher than reality so the bird is late to reduce descend speed when approaching to the ground.


Sent from my iPhone using PhantomPilots mobile app
 
I've tried my best over the past few days to understand how to use Dashware and overlay the controls but I've failed :( I installed the DatConverter and located the video but can't seem to get the right data file... so unless it's absolutely crucial, I think I'll have to give it a miss.

What I can do, however, is record the screen of my tablet playing the footage, with the sticks overlayed, to prove that I didn't do something stupid.

Have you checked if altitude readings are correct? Known case in context of RTH where altitude reading is higher than reality so the bird is late to reduce descend speed when approaching to the ground.

Sent from my iPhone using PhantomPilots mobile app

I'm no expert but it seems to be accurate. It has also landed correctly all the other times I've used RTH.

Do you have a gimbal guard on it?

Nope. I'm surprised the gimbal didn't break tbh.
 
This is what my Phantom looks like when it's holding itself together:

IAxYmc7.jpg


and this is what it looks like after it lands. I just push it closed again, but it comes loose after a landing:

cwe35r9.jpg


This bit pulls off:

tmG98Bs.jpg


I don't know what's inside, but it's a thin card type CPU thing that reminds me of RAM. It fell out in the crash and I just pushed it back in. Seems to function fine.

What I want to know is, based on the information that I've provided, and the state of my P4, is it worth contacting DJI to repair this under warranty, or should I just put a bit of sellotape on it and move on?

I'm sure I didn't do anything wrong, so it appears to be a malfunction and they should repair it for free, right? Is it worth losing my Phantom for however long, though?
 
Ever since the last firmware update I've noticed my elevation isn't accurate when I land at times. Sometimes it thinks it's 50' higher than reality. I never do auto landing so I'm unsure how this might affect the craft with RTH enabled. What's strange is the VPS is suppose to be effective below 30' high, and when my craft is 6 to 10' off the ground it shows 50+' elevation sometimes. It's appears VPS isn't working for some reason, and there's no way to turn off VPS from the app. Has anyone else noticed this elevation error? This elevation inaccuracy isn't consistent, sometimes it seems close, but other times it's way off. If VPS has been disabled somehow with the last firmware, or VPS is intermittent, this could explain this hard landing. The craft may be depending only on the barometer, which may have a wide tolerance for accuracy.

From my recollection, in the past my P4 was always very accurate close to ground because of VPS, but not anymore.

Krieg, when you cancelled the RTH, did you press and hold the button, or did you short press the button? If you press and hold I don't think that will disable RTH. You have to short press the RTH button to disable. Guys, correct me if I'm wrong. I never use RTH but that's my recollection using the RTH button.
 
Last edited:
I cancelled RTH using the app (the big red X on the left hand side) rather than the RC. There's then a popup which asks you to swipe to cancel, which I did.

You can actually see on the log that RTH was indeed cancelled.

Oh, and yes, I have noticed an issue with elevation being inaccurate, but it was only once. Showed as being 50m in the air when it was hovering in front of me. I landed and took off again, and it fixed itself. This was recently, not when this issue occurred.
 
You can actually see on the log that RTH was indeed cancelled..

OK, I didn't see the RTH cancel. The log data I saw didn't note the RTH cancel. Maybe I'm blind.

Did you take off with only 38% battery? That log appears to be a portion of a log, starting at 9min 7.9sec. I've never seen that from a log like this before. Did you land after a 9min flight, leave the motors running, and reset the HOME point, then take off again?

I guess I'm a bit paranoid about flying without 90+% battery charge, after all of the mid air power shut down stories with P3P. Those incidents seemed to be triggered with less than 95% battery at take-off, in discharge cycle, etc. I know the P4's smart battery algorithm is more conservative, so I don't think that's an issue anymore. However, I still don't suggest starting a new flight with 38% battery at take off, but that's just me.
 
I cancelled RTH using the app (the big red X on the left hand side) rather than the RC. There's then a popup which asks you to swipe to cancel, which I did.

You can actually see on the log that RTH was indeed cancelled.

Oh, and yes, I have noticed an issue with elevation being inaccurate, but it was only once. Showed as being 50m in the air when it was hovering in front of me. I landed and took off again, and it fixed itself. This was recently, not when this issue occurred.
I've only had my P4 for a short while, but I've noticed the altitude seems off by 50 ft or more. compared to the VPS? I, too, wonder about that. I also 'tested' RTH once and when it seemed like it was going to crash descending too fast, I cancelled it and landed in 'P' mode. I haven't had the courage to try S mode yet.
 
I've only had my P4 for a short while, but I've noticed the altitude seems off by 50 ft or more. compared to the VPS? I, too, wonder about that. I also 'tested' RTH once and when it seemed like it was going to crash descending too fast, I cancelled it and landed in 'P' mode. I haven't had the courage to try S mode yet.

Note this being already discussed in other threads, as well see my earlier post in this discussion.

Important is to cxl RTH on time. Anyway I use RTH when lazzy to get the bird back and then land manually once cxl RTH. But problem will be in case of major incident were complete RTH needed: bird might descend too fast and crash and with possible vortex.

Re inconsistent altitude readings current findings is that affected P4 are from last production batch + last firmware.


Sent from my iPhone using PhantomPilots mobile app
 
The problem with measuring the height lies in the software , not the hardware ( VPS )!
Dji will correct this in next firmware update as i was tould from Dji EU.
 
Hi everyone

I'm new to this and have a Phantom 4.

I used RTH and the P4 came back to me and started landing at 60M in the air. I saw a guy with a dog approaching so I cancelled the auto landing at approx 30M. I started to lower the drone manually and at approx 25M it started going down, FAST.

I pushed the stick upwards and got no response. It smashed into the ground and bounced back up, at which point I was able to control it again, bringing it to a safe landing.

However, a piece broke off and a component was hanging out of one of the legs. I've put it back in and everything works fine, but it's been making a creaking noise since when I touch it, and the piece has been trying to fall out again.

WTF happened here? I've reviewed the logs and it just cuts out at 25M to the ground and doesn't show anything after that.

I have the video footage of it hitting the ground which has me partially in shot and I can see that I wasn't even on the throttle so wasn't me that smashed it into the ground. Bit strange that the DJI Go app has no info after it was at 25M in the air, isn't it? Does that suggest a Phantom problem to you?
.
...Krieg, I'm speculating a bit here, but i'm guessing DJI's thought is that a RTH is typically an Emergency Situation so the drone needs to get back home as soon as possible.. i do agree with you that the P4 comes down FAST.. i've tested RTH a few times and have NEVER let it actually touch down because of its speed.. seems like DJI could put the VPS to use in this situation and lower the speed once it gets within meter of the ground.. but then again.. i suppose they want to sell parts too hhhaha.. i'm a lil hesitant to try RTH in Sport Mode, my last RTH test it did the YoYo thing at approx 12'' from the ground in p-mode once i halted it,, attached below is an image of the Notifications (of Your Flight), you might glean something from it... BTW what was the Firmware Version as Received and what was the Version you Updated to..?
.
IIIDaemon
www.GasRecovery.net

upload_2016-8-4_2-10-59.png
 
Last edited:
Are we in agreeement that with latest/current firmware P4 does come down much faster than with previous firmware in RTH mode?

Most likely since altitude readings not acurate for some of us with last firmware and bug fix welcome, let's hope.

Previously in RTH we had no problem to auto land as bird reducing descending speed at the right moment. Now we cxl RTH when coming close to ground to calm the bird, for the reason you can imagine. It's actually scary seing that!

Just to highlight it has been not always the case and hope rather malfunction than intended design :)


Sent from my iPhone using PhantomPilots mobile app
 
Last edited:
Are we in agreeement that with latest/current firmware P4 does come down much faster than with previous firmware in RTH mode?

Most likely since altitude readings not acurate for some of us with last firmware and bug fix welcome, let's hope.

Previously in RTH we had no problem to auto land as bird reducing descending speed at the right moment. Now we cxl RTH when coming close to ground to calm the bird, for the reason you can imagine. It's actually scary seing that!

Just to highlight it has been not always the case and hope rather malfunction than intended design :)


Sent from my iPhone using PhantomPilots mobile app
.
...Remy, I have no idea about the Current Firmware Version, or the 'Previous' you're speaking of, I received my P4 with Firmware v01.01.0301 and can see Absolutely No Reason or Advantage to Change or UpDate it, in fact the evidence online shows that Updating from v01.01.0301 is Risky, period.. now you may not believe that the wide spread issues with updating firmware is 'Not Intended', but in my mind then that only leaves one other alternative.. Complete and Utter Incompetence.. I happen to think its Both, reason: The Right Hand Doesn't Know What the Left Hand is Doing over there at DJI, AND they are deathly afraid of loosing market share (which i sincerely hope happens sooner than later)...
.
...as far as the Altitude Variance Issue, it is an OnGoing, Known Issue (with the P4) by DJI and as per DJI '..they are working on it..' ..IF I can purposely land my P4 within inches of my Take-Off Point Manually (all VPS Systems are functioning) THEN, theres no reason why it shouldn't be able to work in an Automatic Mode/Situation aside from Failure or Damage Related Failure...
.
IIIDaemon
www.GasRecovery.net
 
I appreciate what you said and seen others reporting similar. If they resolve this particular regression then I'm happy so far and next time I will wait longer by observing others user experience. I know some possible hidden fixes sometimes such as wrong conversion factor in log etc. but if every time there is regression impacting too much then I need to calm down the mania of getting last version.
Given the subject possibly at least 1 or 2 month if it's possible.


Sent from my iPhone using PhantomPilots mobile app
 
I am pretty sure RTH landing goes "fast" until it gets to a certain lower altitude, then it slows down, adjusts and continues. I would encourage everyone to not use RTH auto landing, in my opinion, RTH should only be used until the drone is visibily controllable. THe only time auto landing should be used is if your controller is inoperable. Just my 2 cents.
 

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,358
Members
104,936
Latest member
hirehackers