P4P moving up down on its own after takeoff.

Oh, just to update, on mine it is happening at 40-70'. However, the speculation on the air pressure/temp gradients may not be too far off. I've noticed it most when flying at home, where I'm taking off from about an acre of clearing of a hillside surrounded by trees, and I'm typically on the leeward side of the hill so I can have pretty different winds at ground level. My standard protocol is to launch a cpl feet, everything is stable then immediately get up above the treeline. This is where I've noticed it, after that launch up, then watching the altimeter stay the same while I watch the bird drift down and see the trees rise up on the screen.

I’m glad it sounds logical to you guys too. Knowing the altitude is only set by a barometer, when you say the altitude reading doesn’t change but it drifts up and down, I’m only seeing changes in air pressure or density as the cause. Birds ride updrafts, so it’s no secret air moves a lot,we just can’t see it. I’d be very interested to see someone test it in an enclosed warehouse at different elevations, using VPS and without it. I bet it’s pretty stable.

My bet (only a guess) is that that they adjusted the algorithm for altitude sensitivity to be more sensitive, with more of a curve smoothing factor or more time to average the change before making an adjustment. If that’s true, we should make a lot of noise to DJI so they readjust it for the better for the next update.

FWIW I’m new to drones but have a mechanical engineering degree. Controls always interested me! I have two flights on my P4P but I’m haunting these forums to know what to expect to avoid crashes and issues.

Is DJI good about responding to issues like this?
 
I believe KevMo and Bakersfield said they are seeing this slight up down wandering at low altitudes less than 20'. Theoretically the VPS should kick in at around 30' AGL, and at that point the barometer is no longer the priority for altitude hold. The sonar from VPS should hold the craft rock solid if the VPS is enabled. When my P4P craft is in a hover at 6' AGL it's always rock solid for altitude. I hand catch 90% of the time, and I always watch my craft hover at 6' for 5 seconds or more before I grab it, so I would notice any instability in altitude when hovering at low altitudes. Maybe a firmware upgrade has messed with your VPS somehow. This is a long-shot, but have you guys tried calibrating the sensors? Although the down facing cameras shouldn't control altitude for VPS, maybe it's tied into the sonar for altitude somehow, and there's some uncertain calculation going on the affects altitude stability. During the sensor calibration it calibrates the downward cameras, but I doubt if this does a calibration of the sonar, but I could be wrong.

@KevMo Photog @Bakersfield Quad , what craft firmware are you flying with?
 
Last edited:
I believe KevMo and Bakersfield said they are seeing this slight up down wandering at low altitudes less than 20'. Theoretically the VPS should kick in at around 30' AGL, and at that point the barometer is no longer the priority for altitude hold. The sonar from VPS should hold the craft rock solid if the VPS is enabled. When my P4P craft is in a hover at 6' AGL it's always rock solid for altitude. I hand catch 90% of the time, and I always watch my craft hover at 6' for 5 seconds or more before I grab it, so I would notice any instability in altitude when hovering at low altitudes. Maybe a firmware upgrade has messed with your VPS somehow. This is a long-shot, but have you guys tried calibrating the sensors? Although the down facing cameras shouldn't control altitude for VPS, maybe it's tied into the sonar for altitude somehow, and there's some uncertain calculation going on the affects altitude stability. During the sensor calibration it calibrates the downward cameras, but I doubt if this does a calibration of the sonar, but I could be wrong.

@KevMo Photog @Bakersfield Quad , what craft firmware are you flying with?
I have not tried calibrating the sensors yet John. Prob wouldn't hurt either. And I still hand catch. It seems like this up and down is only going on at take off. When I come back it's not doing it. Also not doing it when you stop out away from you on a flight out. And I am using the last update firmware whatever that is. LOL. And it's been ok except for the stupid GeoFencing even around small airports! Big hassle for sure at times. But it did fix that slight gimbal twitch that always had first few months.
 
I’ve had 2x P4P and both have twitchy gimbals for still photos. It’s incredibly frustrating. No combination of settings or calibrations so far have made a difference. I wonder if it’s more pervasive than people realize but that maybe those doing video or less technically exacting photography wouldn’t notice it.
 
How I would describe what I have seen is "Settling" a gentle drift downward, if I try to move along under 10 feet slowly. I can see the craft settling, maybe 1 foot at most, this hasn't been a problem, just have to watch it if I am very low. Seems a bit more at say 5-6 feet than 10 feet for me.
Not a problem, just something I have noticed when flying close and low to the ground.
 
  • Like
Reactions: DaRana
@KevMo Photog Gimbal twitch? I'm on firmware 1.3.509 from May and I haven't seen any gimbal twitch. Do you recall the firmware version number where you had that problem?
I don't remember for sure John. But as I recall it is whatever version that I was on before the big NFZ update a few months ago. But I don't think that everyone was having the little twitch. I'm still getting a weird issue with taking stills here and there since the last update too btw. It's like the app or camera freezes for a few seconds to almost a half minute randomly. And then you will hear it rapid fire off 2 or 3 shots in the app and then it's unfrozen and shoots a pic as soon as you snap it. Kind of a bother.
 
I don't remember for sure John. But as I recall it is whatever version that I was on before the big NFZ update a few months ago. But I don't think that everyone was having the little twitch. I'm still getting a weird issue with taking stills here and there since the last update too btw. It's like the app or camera freezes for a few seconds to almost a half minute randomly. And then you will hear it rapid fire off 2 or 3 shots in the app and then it's unfrozen and shoots a pic as soon as you snap it. Kind of a bother.

KevMo what you are describing with the delays sounds like what has been happening to me, but it’s actually a precursor to go4 starting to crash. It starts with the pictures getting slower like that, and then it actually got slower believe it or not and occasionally even missed shots ( which is quite a bother when you’re shooting 100 shot HDR panorama let me tell you!), but eventually got to the point where the app is crashing during flight.

Anyways, uninstalling the app and reinstalling it has fixed it twice now. Twice being the operative word, as in it slowly has been getting worse until I have to do the reinstall again. This started with the update to iOS 11 and there was a simultaneous update to Go4.

It’s a bother bc it doesn’t keep the settings, or at least not all of them, so you have to go through and reset everything. And, no, doing the space saving delete the app keep the settings new feature with an iOS 11 did not fix it. [emoji3]
 
KevMo what you are describing with the delays sounds like what has been happening to me, but it’s actually a precursor to go4 starting to crash. It starts with the pictures getting slower like that, and then it actually got slower believe it or not and occasionally even missed shots ( which is quite a bother when you’re shooting 100 shot HDR panorama let me tell you!), but eventually got to the point where the app is crashing during flight.

Anyways, uninstalling the app and reinstalling it has fixed it twice now. Twice being the operative word, as in it slowly has been getting worse until I have to do the reinstall again. This started with the update to iOS 11 and there was a simultaneous update to Go4.

It’s a bother bc it doesn’t keep the settings, or at least not all of them, so you have to go through and reset everything. And, no, doing the space saving delete the app keep the settings new feature with an iOS 11 did not fix it. [emoji3]
Same here exactly DaRana! App never crashed once until the update a few months ago. And while taking pics I have had the screen go black & white and then lock and then crash. I'm so surprised that this has not been fixed after several months.
 
Same here exactly DaRana! App never crashed once until the update a few months ago. And while taking pics I have had the screen go black & white and then lock and then crash. I'm so surprised that this has not been fixed after several months.

Unfortunately, while there are many of us, I think those of us that are using the P4P for primarily photography and are very focused on those technical aspects are still in the minority. I'm honestly a little bit torn right now, I love the P4P camera quality, but I continue to have such struggles with the gimbal. This is the second unit with the exact same behavior. I flew my Mavic yesterday to re-remind myself that I'm not crazy, and it is a completely different experience. I can spin the mavic in a 360 circle in sport mode, stop, wait 2 seconds and the gimbal sits right down. Take a photo or 5 shot AEB at 1/10th second and after the shot the gimbal has not moved a single pixel. Whereas on the P4P, I can take off, turn a few degrees very slowly, let it sit for 30 seconds as I watch the gimbal slowly drift one way or the other, maybe finally stopping/settling down, take a shot with the same settings, and at the end of the 5 shot AEB the gimbal has jumped a couple % of the whole frame. Every. Single. Time. Except for the 1 in 50 when it doesn't for no reason at all I can discern ... even if I take another shot without touching the sticks, it'll jump again the next time after the exposure. Just utterly infuriating and frustrating. In the end my large panos come out far better and more reliably with the Mavic despite the much smaller sensor - since with the P4P it simply isn't stichable half the time, or jumped between frames enough that it causes blurs.

/frustrated rant off

Anyways, point being, I wonder how widespread this is. I would not be surprised if it is something that many wouldn't notice.
 
  • Like
Reactions: KevMo Photog
I flew my P4P+ three batteries worth today in 11mph winds and it was a joy. Short of having to reboot the AC because the camera remained pointed to the ground and would not respond to the dial on the first battery, it was flawless. There was no downward drift or settling. I purposely tries low altitude hovering but it maintained altitude perfectly. No gimbal problems, no beeping except high wind warning, just smooth flying. It behaved the way it should EVERY time. Made my day.
 
My P4P just started moving up and down on it's own a little bit after takeoff. Is this usually fixed by an IMU calibrate or is it a VPS calibrate? The app is not calling for either one btw. It does not seem to do it when I am up a little higher. Thoughts anyone?
Are you close to a large metal building or any large metal objects if not delete the firmware and reinstall it and then recalibrate your compass and Imu it should fix it do not fly it can't tell where it is supposed to be which means it might not listen to you or hear in sorts farther away kind of hard to explain just make sure it hovers correctly before you fly or you may never see it again and for everyone get a gps tracker when these go there gone forever for most people there's my five cents fly safe and leagle it's up to us no one is going to help to many people doing ******* stuff to take this hobby away from use be proactive and educate so we can hold onto this hobby
 
You don't have a gimbal guard on the P4 do you? Seems no matter what kind I found to put on it messed with the VPS downward sensors. I would float around up and down and different altitudes. Removed it and problem solved
 
This is a known problem caused be the heating up of the drone and the altimeter. Normally only affects the first 10 mins of flight then stabilises. If you look at your take off and landing altitudes in any log file you will see that the landing altitude is approximately 3m higher than take off.
I have proven this by carrying a RTK gps on my P4 and. Ompared flight paths.
 
Are you close to a large metal building or any large metal objects if not delete the firmware and reinstall it and then recalibrate your compass and Imu it should fix it do not fly it can't tell where it is supposed to be which means it might not listen to you or hear in sorts farther away kind of hard to explain just make sure it hovers correctly before you fly or you may never see it again and for everyone get a gps tracker when these go there gone forever for most people there's my five cents fly safe and leagle it's up to us no one is going to help to many people doing ******* stuff to take this hobby away from use be proactive and educate so we can hold onto this hobby
Def not close to anything. I calibrated the IMU last night and am testing now.
 
You don't have a gimbal guard on the P4 do you? Seems no matter what kind I found to put on it messed with the VPS downward sensors. I would float around up and down and different altitudes. Removed it and problem solved
Def no gimbal gaurd on man. Calibrated the IMU last night. Testing now.
 
This is a known problem caused be the heating up of the drone and the altimeter. Normally only affects the first 10 mins of flight then stabilises. If you look at your take off and landing altitudes in any log file you will see that the landing altitude is approximately 3m higher than take off.
I have proven this by carrying a RTK gps on my P4 and. Ompared flight paths.
Not sure about that. My P4P did not do this at all for 6-7 months. just started this a month ago. I calibrated the IMU and am testing now.
 
This is a theory, but I am pretty confident in it for this purpose. If you buy a $10 cheap helicopter, it will hover just find when it reaches equilibrium, but if you put your hand under it it will move up because you're changing the equilibrium. What I think is happening is that you're getting changing air pressure stability under the craft. If the air pressure rises a little, it will move it up until the computer realizes it and adjusts, and allows it to drift back down. That's why it will work better in open areas higher up, because the air is less turbulent and more predictable for the computer.

Mine will do it slightly. More in tighter areas, less or none in open areas. More with a breeze, less or none in dead calm conditions. It'll do it on my back porch even hovering as high as 5' where the downdraft can get turbulent around the grill and table and chairs, but it will be a lot better in the yard where it's flat for 3'-5' away, hovering as low as 3' or so.

TL;DR- if the air from the props or breeze isn't steady but is turbulent, the computer can't predict it as well and you're seeing slight overcompensation.

What seems obvious is that there are enough folk noticing the change recently and reporting it, myself included (and yes after an update). The common factors are recent and global... based on these 2 factors I'd conclude a recent algorithm modification, most likely in our drones.
 
  • Like
Reactions: KevMo Photog
I just took my bird out a little bit ago after an IMU calibrate. My bird is now rock steady again in a low hover. I like that!
 
Flew again tonight! My bird is rock solid now and even the camera app and taking pics seems to be fixed now as well. I have shot pics at night yet. But it seems like a new bird after IMU calibrate! So many have said not to calibrate IMU's if it is not asking for it. But if you are having any kind of issue with drifting up and down or sideways then I would go ahead and calibrate the IMU's. Mine is flying the best it ever has again!
 

Recent Posts

Members online

Forum statistics

Threads
143,095
Messages
1,467,610
Members
104,981
Latest member
Scav8tor