I've had my PSP for just a little over one year. Up until the last couple of months it worked very well, was fun and predictable in it's characteristics. However now it is having all sorts of issues. The most recent issue (among others) is it's behaviour on descent. It is rocking from side to side. Almost violently.

I would try an IMU calibration, that might help to stabilise things.

Aardvark Posted at 2017-1-22 09:02
I would try an IMU calibration, that might help to stabilise things.

My local DJI dealer did the IMU calibration 4 flights ago. I suppose I could try that again though.

GuyS Posted at 2017-1-22 09:12
My local DJI dealer did the IMU calibration 4 flights ago. I suppose I could try that again though.

Just make sure it's on a level surface and don't disturb it by walking about etc. Shouldn't take more than five minutes to do.

Have any of the props been damaged at all ?

Aardvark Posted at 2017-1-22 09:26
Just make sure it's on a level surface and don't disturb it by walking about etc. Shouldn't take more than five minutes to do.

Have any o the props been damaged at all ?

I had a couple of tip to one side incidents last summer and the props sustained very minor damage but I flew it many times since then without issue. It is only since the end of Oct of last year all my problems started. The latest unstable descent issue just started in the last few flights.

I always try to keep some lateral speed when decending to avoid your own prop wash

PeteH. Posted at 2017-1-22 09:53
I always try to keep some lateral speed when decending to avoid your own prop wash

Makes sense Pete but this unstableness is new in it's degree. It is very wobbly on descent.

GuyS Posted at 2017-1-22 10:17
Makes sense Pete but this unstableness is new in it's degree. It is very wobbly on descent.

Vortex Ring State can can occur sort of randomly. Does it occur during manual or automatic descent? The two ways to deal with it are maintaining lateral movement as PeteH recommended or reduce the rate of descent. Hopefully it's VRS rather than an aircraft problem.

RSW Posted at 2017-1-22 14:01
Vortex Ring State can can occur sort of randomly. Does it occur during manual or automatic descent? The two ways to deal with it are maintaining lateral movement as PeteH recommended or reduce the rate of descent. Hopefully it's VRS rather than an aircraft problem.

Some things to think about from your and others suggestions. Next time I take it up I'll test it and see if I can discover causal effects.

GuyS Posted at 2017-1-23 12:14
Some things to think about from your and others suggestions. Next time I take it up I'll test it and see if I can discover causal effects.

Are you descending at full down stick? If so, back off a little or reduce the down left stick EXP in the advanced aircraft setting. I run mine at a more moderate rate. That way the dreaded VRS won't spoil my day. Also horizontal motion keeps you out of your own wake when descending.

I had the same problem with my P2 had a nice wobble and drop. On my P3P I learned to add a bit of foward motion
No wobble .  A bit foward as you get lower

RSW Posted at 2017-1-22 14:01
Vortex Ring State can can occur sort of randomly. Does it occur during manual or automatic descent? The two ways to deal with it are maintaining lateral movement as PeteH recommended or reduce the rate of descent. Hopefully it's VRS rather than an aircraft problem.

Aloha RSW,

     It is very hard to put a newer Phantom into VRS (since the P3).  DJI pretty much designed it out of the system.  That is why they were able to increase the descent speeds.  

Aloha and Drone On!

rbryant1492 Posted at 2017-1-23 23:02
I had the same problem with my P2 had a nice wobble and drop. On my P3P I learned to add a bit of foward motion
No wobble .  A bit foward as you get lower

Aloha rb,

     The P1 and the P2 up to the V+ could go into VRS and crash or land hard.  The wobble of death.  But since the P2 Version 3, DJI pretty much designed VRS out of the system and reduced the descent speed allowed by the app.  The P3s and P4s are very difficult to put into VRS.  That is why DJI increased the descent speeds.  

     But still, vertical descents are boring!

Aloha and Drone On!

Rnfaust Posted at 2017-1-23 13:40
Are you descending at full down stick? If so, back off a little or reduce the down left stick EXP in the advanced aircraft setting. I run mine at a more moderate rate. That way the dreaded VRS won't spoil my day. Also horizontal motion keeps you out of your own wake when descending.

Aloha faust,

     VRS has pretty much been designed out of the later Phantoms as I replied in response to the other posts.  Unless you are in very thin air, it should not be a problem.  But still vertical descents are boring!  More fun with lateral movement.

Aloha and Drone On!

GuyS Posted at 2017-1-23 12:14
Some things to think about from your and others suggestions. Next time I take it up I'll test it and see if I can discover causal effects.

Aloha Guy,

     You should do a test with completely new propellers as Aardvark is indicating.  Are you using propguards?  If so, when did you start using them.  Since you did have a propeller problem, the propeller test is very important.  Cracks creep and the blades spin very fast.  You could loose a propeller and crash, so testing with new propellers is very important.  I know someone who never wanted to change his propellers and damaged propguards.  Instead, he repaired them.  Finally he replaced them to make it look better and was amazed at how all his flight problems went away.

     Propguard instability is very rare but propeller problems are the most likely cause of problems like you describe.  They are the cause of crashes.  The increasing problem as you describe matches crack creep problems.

Aloha and Drone On!

Cetacean Posted at 2017-1-24 02:14
Aloha Guy,

     You should do a test with completely new propellers as Aardvark is indicating.  Are you using propguards?  If so, when did you start using them.  Since you did have a propeller problem, the propeller test is very important.  Cracks creep and the blades spin very fast.  You could loose a propeller and crash, so testing with new propellers is very important.  I know someone who never wanted to change his propellers and damaged propguards.  Instead, he repaired them.  Finally he replaced them to make it look better and was amazed at how all his flight problems went away.

I'll check these possibilities out. The problem flight day was colder and the higher air density might have something to do with wobble effect. The denser air gives the props more thrust ...but also generates stronger vortices.

I check the props out frequently. There are no cracks. If I have a little nick I sand both sides of the prop blade ....to maintain balance. I do have another brand new set of props. I might try that if I notice the wobble next flight. Also if the Phantom has it's limitations then practice will help to avoid exceeding them.

GuyS Posted at 2017-1-24 19:38
I'll check these possibilities out. The problem flight day was colder and the higher air density might have something to do with wobble effect. The denser air gives the props more thrust ...but also generates stronger vortices.

I check the props out frequently. There are no cracks. If I have a little nick I sand both sides of the prop blade ....to maintain balance. I do have another brand new set of props. I might try that if I notice the wobble next flight. Also if the Phantom has it's limitations then practice will help to avoid exceeding them.

I have gotten to the point... that any time I notice a performance change such as that I always inspect the propellors first.  90% of the time they seem to be the culprit, then for good measure you should follow up with a IMU calibration.
Im sure you already have,,, but again inspect all the screws make sure they are tight and make sure all the weight is balanced...
If that does not fix it...its time to send it to China,  there is only so much you can do to a phantom on you're own.  
You can not hook one to clean flight and adjust the PIDS among other things... and if you make any major hardware repairs you'll need a way to monitor the effects & possibly  tune them up...
As far as I am aware @ this point in time only DJI can do this.

Good luck im hoping its just the props

Speaking of cracks, the arms tend to develop cracks, particularly in the screw wells behind the motor screws.  That could cause the arms to start flexing more.

DanMan32 Posted at 2017-1-25 06:56
Speaking of cracks, the arms tend to develop cracks, particularly in the screw wells behind the motor screws.  That could cause the arms to start flexing more.

The first few times that I flew my P4p v2.0, it was in rather strong winds. The drone was completely stable. Even while descending straight down from a high altitude. Totally smooth footage.

Fast forward to today. The first 'calm' day of no wind. I took her up to get some views of the downtown area and I noted that on a couple of descents, even slow ones, there was side to side wobble of the camera. And while aloft and flying in a southerly direction, a couple of times I noticed the camera 'twitch' --it lasted 3 frames--in an otherwise smooth flight.

When I got home, I cleaned off the propellers. Some dead insect parts stuck to the blades and did an IMU calibration, twice, to make sure I did it correctly, then did a gimbal calibration. I'll probably take it up tomorrow and see if the vibration has cleared.

My hunch is that without the strong winds blowing away the vortex beneath the rotors, VRS was making the drone vibrate more than usual on descent in the calm air.

Mark Weiss Posted at 2018-5-28 18:09
The first few times that I flew my P4p v2.0, it was in rather strong winds. The drone was completely stable. Even while descending straight down from a high altitude. Totally smooth footage.

Fast forward to today. The first 'calm' day of no wind. I took her up to get some views of the downtown area and I noted that on a couple of descents, even slow ones, there was side to side wobble of the camera. And while aloft and flying in a southerly direction, a couple of times I noticed the camera 'twitch' --it lasted 3 frames--in an otherwise smooth flight.

My hunch is that without the strong winds blowing away the vortex beneath the rotors, VRS was making the drone vibrate more than usual on descent in the calm air.
The P4 pro is quite stable and doesn't suffer from VRS at all, even in calm air.
Not the motors are offset from the vertical - that is what keeps VRS away.

Mark Weiss Posted at 2018-5-28 18:09
The first few times that I flew my P4p v2.0, it was in rather strong winds. The drone was completely stable. Even while descending straight down from a high altitude. Totally smooth footage.

Fast forward to today. The first 'calm' day of no wind. I took her up to get some views of the downtown area and I noted that on a couple of descents, even slow ones, there was side to side wobble of the camera. And while aloft and flying in a southerly direction, a couple of times I noticed the camera 'twitch' --it lasted 3 frames--in an otherwise smooth flight.

It is not to be rude but do you have to drag out very old threads just to make comments that just don't fit the thread?
This thread was initially about P3's so the experience you have on you brand new P4P v.2.0 brings no value to those old threads.

Labroides Posted at 2018-5-28 18:29
My hunch is that without the strong winds blowing away the vortex beneath the rotors, VRS was making the drone vibrate more than usual on descent in the calm air.
The P4 pro is quite stable and doesn't suffer from VRS at all, even in calm air.
Not the motors are offset from the vertical - that is what keeps VRS away.

The only change to configuration of the craft was setting the camera from 16:9 to 4x3. I am wondering if that reduced the in-camera image stabilization range to the point where it can't cancel out all of the vibrations?

Mark Weiss Posted at 2018-5-29 05:16
The only change to configuration of the craft was setting the camera from 16:9 to 4x3. I am wondering if that reduced the in-camera image stabilization range to the point where it can't cancel out all of the vibrations?

I am wondering if that reduced the in-camera image stabilization range to the point where it can't cancel out all of the vibrations?
There is no in camera stabilisation to wonder about.
And changing the image resolution makes no difference to the way the gimbal keeps the camera stable.

Labroides Posted at 2018-5-29 05:25
I am wondering if that reduced the in-camera image stabilization range to the point where it can't cancel out all of the vibrations?
There is no in camera stabilisation to wonder about.
And changing the image resolution makes no difference to the way the gimbal keeps the camera stable.

Is this a certainty that there is no digital image stabilizer? A 20MP sensor outputting 8.8MP video has a lot of room for digital image stabilization. Since the gimbal doesn't provide any lateral stabilization (I noted this while holding the drone and watching the image on the tablet screen while shaking it slightly in a lateral mode to simulate drone vibrations) it would have to be cleaned up with in camera stabilization. However, I could see the lateral shake when I hand held the drone and simulated aircraft vibration with my hand, or setting on the table and shaking the table +/- 1mm.

There is a dramatic difference in the smoothness of footage between the last three days (in high winds) and last night (in calm air). My first thought was I was seeing the effects of VRS because no wind to blow the vortex out from under the drone, but I was told in another post that the 2.0 doesn't have VRS problems due to the motor tilt.

So the only other thing that changed besides calmer air was the aspect ratio. I know it doesn't make any sense, but in software, sometimes changing something that seems unrelated can cause another feature to be affected in unexpected ways.

Just to be on the safe side, when I got home, I did an IMU calibration, twice, to make sure I was doing it correctly and did a gimbal calibration as well. I plan to take it up later today and try a couple of descents in both 4:3 and 16:9 and see if there is any change in camera stability.

I can't imagine what else could be causing the unstable footage, unless there's some interference flying over town that I didn't have in the ball field.

Mark Weiss Posted at 2018-5-29 06:17
Is this a certainty that there is no digital image stabilizer? A 20MP sensor outputting 8.8MP video has a lot of room for digital image stabilization. Since the gimbal doesn't provide any lateral stabilization (I noted this while holding the drone and watching the image on the tablet screen while shaking it slightly in a lateral mode to simulate drone vibrations) it would have to be cleaned up with in camera stabilization. However, I could see the lateral shake when I hand held the drone and simulated aircraft vibration with my hand, or setting on the table and shaking the table +/- 1mm.

There is a dramatic difference in the smoothness of footage between the last three days (in high winds) and last night (in calm air). My first thought was I was seeing the effects of VRS because no wind to blow the vortex out from under the drone, but I was told in another post that the 2.0 doesn't have VRS problems due to the motor tilt.

Is this a certainty that there is no digital image stabilizer?

Yes ... I wouldn't have said it if I wasn't certain.
The gimbal does all the stabilising and it does a great job.
The drone can be subjected to all kinds of buffeting and the gimbal keeps the camera stable.

So the only other thing that changed besides calmer air was the aspect ratio. I know it doesn't make any sense
That's right.  That makes no sense.


Just to be on the safe side, when I got home, I did an IMU calibration, twice
It's not necessary and not advisable to go messing with calibrating this and calibrating that.


I can't imagine what else could be causing the unstable footage, unless there's some interference flying over town that I didn't have in the ball field.
Interference won't have any effect on the stability of the gimbal either.

How is the horizon?  Is it perfect?  

Are you sure the plug in the gimbal is fully inserted?  If it wasn't originally inserted until it clicked, it could have begun to work its way out which would eventually cause an intermittent connection problem - especially while the AC is shaking in a straight down descent.  You wouldn't be able to tell by looking at it.  You'd need to carefully push down on the top of the plug fully with a tool of some sort.  

You might also clean the connection there.

Good luck

Mark The Droner Posted at 2018-5-29 07:19
How is the horizon?  Is it perfect?  

Are you sure the plug in the gimbal is fully inserted?  If it wasn't originally inserted until it clicked, it could have begun to work its way out which would eventually cause an intermittent connection problem - especially while the AC is shaking in a straight down descent.  You wouldn't be able to tell by looking at it.  You'd need to carefully push down on the top of the plug fully with a tool of some sort.  

I checked the ribbon cable on the gimbal and it appears to be fully inserted. I pressed on the connector shell that surrounds the cable with a sharp pointy tool, but it would not move, so it seems fully inserted.

I'm going to edit down the footage and make a separate short video showing the odd behaviors. I noticed that some times when I turn the drone while it's looking down and stop turning, there's a sudden backlash in camera rotation. In another down-facing shot, the camera abruptly repositioned to the left by a few pixels, subtle but noticeable on a 4K display.

As for calibrations, I was under the impression that this should be done on a regular basis. Is that not the case?

If the camera is looking straight down, it's fully rotated down, right?  So if it needs to adjust for smoothness, there's no room for it to move further in that direction.  So you'll see that roughness in the vid.  That could be an explanation.

Re calibration, Labroides is pointing out that it's not necessary or worthwhile to calibrate excessively.  Once you have a good calibration, it will likely stay that way, and re-calibrating risks turning a good calibration into a bad calibration.  If you have a solid reason to calibrate, that's fine, but you shouldn't do it just to do it.  So no, it should not be done on a regular basis.

Good luck

My P3SE does wobbles when full stick down to descend ... when take it moderate down..half stick down...it slows and descending with more stable look........