Shaky-cam footage of shaky P3P.

[Morgon]

I've always thought my P3P was shakier than it should be, especially after a number of pictures taken on the 4th of July came out blurry. I did IMU calibration a couple of times today, ending up with a surface that couldn't have possibly been any more level, but that didn't do much - oddly, the Acceleration Z-axis is steadily at -1.0; the other values are okay.

A while ago, @Tahoe Ed posted a video of his super-stable P3, so I wanted to document how mine stacks up. I realize my footage is also a bit shaky, but compare the wobble of the background to the wobble of the landing gear and gimbal. Watching the preview on my iPad, I could clearly see the movement (I wish I would have included a shot of that in the video).

Thoughts? I tried to return this to Adorama, but they're requiring me to go through DJI, which is really aggravating.

 

[flyNfrank]

Put your spare set of props on and see if anything changes for the better.

 

[Jbundy63]

WOW your gimbal is not compensating? Did it do that before the IMU calibration? Did you do a gimbal calibration as well?

 

[johnp44]

I've always thought my P3P was shakier than it should be, especially after a number of pictures taken on the 4th of July came out blurry. I did IMU calibration a couple of times today, ending up with a surface that couldn't have possibly been any more level, but that didn't do much - oddly, the Acceleration Z-axis is steadily at -1.0; the other values are okay.

A while ago, @Tahoe Ed posted a video of his super-stable P3, so I wanted to document how mine stacks up. I realize my footage is also a bit shaky, but compare the wobble of the background to the wobble of the landing gear and gimbal. Watching the preview on my iPad, I could clearly see the movement (I wish I would have included a shot of that in the video).

Thoughts? I tried to return this to Adorama, but they're requiring me to go through DJI, which is really aggravating.

It looks like air turbulence. If you have ever flown a real heli near the ground, you would know why I think that. Try it outside when wind is calm and see what it does. Any hover near the ground will most likely do exactly what you are experiencing. Especially indoors. Even an A/C duct blowing air in the room can cause that amount of drifting.

 

[ScottyT]

That looks pretty standard for the latest firmware and not too bad for flying indoors with reflective air turbulence in what I am guessing is P-Opti mode?. You still have VPS enabled?

 

[Jbundy63]

Agreed that it seems normal, especially being indoors, I was looking at the gimbal, it kind of looks like it's not compensating with the movement of the P3? I could be wrong?

 

[Morgon]

Thanks for all the responses so far.

Put your spare set of props on and see if anything changes for the better.

I've mixed and matched all the props - I even bought another set (two pair) from Amazon, and set them on the balancer. The majority of them seem to be weighted on the one side of hub (vs either of the blades themselves). I would have considered this an issue with my balancer or placement, but I do have a few that are reasonably stable all around, which I'm using.

WOW your gimbal is not compensating? Did it do that before the IMU calibration? Did you do a gimbal calibration as well?

My gimbal does compensate as best it can, but the vibrations seem to exceed the capabilities. There were several shots from the 4th that should have been much, much better. My hope was that the IMU calibration would help (and maybe it's helped), but it's certainly not where it should be. I've done several gimbal calibrations... it's never been fully perfect.

It looks like air turbulence. [...] Any hover near the ground will most likely do exactly what you are experiencing. Especially indoors. Even an A/C duct blowing air in the room can cause that amount of drifting.

I'm not doubting that it's being affected by its own wake, but definitely not to the degree that it's in. I specifically turned off the AC (which is above the window you see in the video) for this test.

--

Anyway, the vertical movement is less important to me than the vibration of the gimbal. I know there aren't many 'still' shots, but just try to concentrate on the edge of the gimbal peeking from the landing gear, you can visually see movement. I'll try to upload some of my pics or videos to see how this translated to the camera.

I'm trying to find @Tahoe Ed's video for comparison (assuming I didn't dream this up); it was a P3 flying in a warehouse, rock-solid. I think it was on RCGroups, but there are just too many posts to fish through.

 

[ScottyT]

Oh right, sorry, wasn't looking at the gimbal. Hard to tell from the video. I'd have to fire my one up to compare, also. Where's the link to Ed's video?

As you say, perhaps sample video from the camera might be better.

 

[Jbundy63]

By any chance were they night time shots? Just curious

 

[ScottyT]

One thing you could try is removing all the props and binding the P3 to a table (heavy plank between the legs, cable ties etc) then revving up the motors to see if there is any noticeable vibration on either of the arms. You can measure by sight or touch or you could possibly use a suitable phone app.
If ok, proceed to place one prop on a motor at a time and measure any difference, then place multiple props on.
If nothing obvious it might be a faulty gimbal, especially if you've already swapped all the props.

 

[Morphious]

Make sure packing foam behind gimbal is removed. I sent back my first P3P for the same reason.

 

[Morgon]

By any chance were they night time shots? Just curious

The worst offenders were, yeah. I am fully aware that the low-light capabilities are pretty substandard, but I have also seen recent images and video from folks on this forum whose quads whose P3s are worlds ahead of what I'm able to capture. (I highly doubt they were post-processing big blobs into something sharp )

I have an example video that shows yet another issue (lots of turbulence when lowering altitude; moreso than P2) which I'll need to edit the relevant parts from a larger file. Until then, a couple of example shots..

You might need to zoom in on this to make sense of it, but the trees are blurry. Before you think I'm off my rocker, I went and compared similar shots from my P2, and there's a difference between fuzziness from quality and the fuzziness from instability; this is the latter.

An example of a still image during fireworks. Admittedly, this is the worst one out of several others that are still blurry, but I will mention that there was only a very slight breeze the entire night.

This one should have been my crowning achievement of the evening. Because it's such a slight blur, it's hard to see in my daytime shots, but it really shows here. Understand that I'm not referring to the noise; as I said, I accept that. But someone else posted a nighttime shot similar to this recently, making it clear mine should have been better.

 

[Morgon]

A video. I am moving too much, but you can see the shake between 3-4 seconds, and then (relative) craziness when I'm descending.

It does seem rather unstable when coming down.. I thought it would have been more stable than the P2V+, but I've learned never to descend straight down; I always move horizontally, but the props still make unsettling noises.

 

[ScottyT]

A video. I am moving too much, but you can see the shake between 3-4 seconds, and then (relative) craziness when I'm descending.

It does seem rather unstable when coming down.. I thought it would have been more stable than the P2V+, but I've learned never to descend straight down; I always move horizontally, but the props still make unsettling noises.

I'm no multimedia buff, but I can definitely see the shudder in the video.

 

[wordsmith]

You've probably already checked this, but is your gimbal mode set to "follow" or "fpv"? From the first video you posted it doesn't look like the gimbal is compensating for the motion of the P3 at all, which is the kind of behavior it has in "fpv" mode. All the vibration and shakiness of the craft is automatically transmitted to the camera that way.
Just a thought.

 

[Morgon]

You've probably already checked this, but is your gimbal mode set to "follow" or "fpv"? From the first video you posted it doesn't look like the gimbal is compensating for the motion of the P3 at all, which is the kind of behavior it has in "fpv" mode.

No, it's on Follow. But you reminded me that I haven't seen any difference between the two on this P3 (whereas the P2V was very obvious), which I always found weird.