My quad loses height just after lift off. I normally go straight to 4-5 m.
The VPS gives the correct height, but the barometric height can be off a few m.
The first picture is just after lift off. The (barometric) height is 4.2 m. The VPS shows 3.8m.

The second pitcure is two minutes later, after moving the quad, the height is still 4.3 m, but the VPS shows the correct 0.8 m.
The difference is almost 3 m/10 feet.

I have calibrated all sensors. Today the weather is stable and there is not much wind.I have seen other posts about this effect, but this seems worse.

Is this normal?

"Is this normal?"
That can happen, normally I would expect it to stay closer than that normally but there is no guarantee that the barometer will stay accurate during a flight.  

That is one of the reasons it has ultrasonic sensors, they are nearly always accurate, but only work up to a height of 15m, the barometer is needed for above 15m.

Nigel_ Posted at 2017-4-1 07:10
"Is this normal?"
That can happen, normally I would expect it to stay closer than that normally but there is no guarantee that the barometer will stay accurate during a flight.  

The VPS might be more accurate, but the actual height is the barometric. If I would have waited longer the quad might touch the ground, despite the VPS.

Rene vd Meer Posted at 2017-4-1 07:57
The VPS might be more accurate, but the actual height is the barometric. If I would have waited longer the quad might touch the ground, despite the VPS.

Yes, I think that is so that if you are flying over rough ground, it doesn't rise up whenever it flies over a bump or a rock and give bumpy video.  

Actually, I think normally I would prefer it to fly up and over things, but that would have to be optional.

Just to clarify, do you have any sort of aftermarket brace at the bottom of the unit?

My bird is in the shop right now because altitude showed 94 feet when I hit the 60 foot tree about 3/4 of the way up.

{:4_142:}

DJI-Jamie Posted at 2017-4-1 08:25
Just to clarify, do you have any sort of aftermarket brace at the bottom of the unit?

No nothing.

Rene vd Meer Posted at 2017-4-2 01:28
No nothing.

Are you noticing this right out of the box, or is it after a firmware update or hard landing?

DJI-Jamie Posted at 2017-4-2 10:46
Are you noticing this right out of the box, or is it after a firmware update or hard landing?

I bought the drone in December last year. I did notice height variations then, but I thought it was caused by the gimbal protection (which I removed).
A month ago I had to fly close to the ground. Then I noticed this behavior the first time.  
I have seen that the VPS height can be very low, shouldn't VPS prevent the drone from touching the ground?

Rene vd Meer Posted at 2017-4-3 05:08
I bought the drone in December last year. I did notice height variations then, but I thought it was caused by the gimbal protection (which I removed).
A month ago I had to fly close to the ground. Then I noticed this behavior the first time.  
I have seen that the VPS height can be very low, shouldn't VPS prevent the drone from touching the ground?

The VPS readings can be affected depending on the lighting conditions, color and the solidity of the surface it's flying over. On pages 22-23 of the P4 manual, it goes over the conditions that can effect the performance of the sensors:

https://dl.djicdn.com/downloads/ ... ser+Manual+v1.4.pdf

If the VPS isn't in the optimal conditions, the performance will be off. Given the last screenshot you provided, it should have been a little more accurate barring if the ground is becoming elevated or if you were above a very soft patch of ground where the sonar like component is easily absorbed  (like sand) that we can't see. If this continues to be an issue, you may want to look into having the unit sent in for further evaluation and potential repair.

DJI-Jamie Posted at 2017-4-3 12:07
The VPS readings can be affected depending on the lighting conditions, color and the solidity of the surface it's flying over. On pages 22-23 of the P4 manual, it goes over the conditions that can effect the performance of the sensors:

https://dl.djicdn.com/downloads/phantom_4/20170327/Phantom+4+User+Manual+v1.4.pdf

Thanks DJI-Jamie for the response.
The VPS readings are OK. It is the barometric height that gives a ~3m to high value. As you can see in the picture, the quad is close to the ground. The VPS reads 0.8 m, but the barometer 4.3 m.
Is a deviation of 3 m of the barometer normal?

Rene vd Meer Posted at 2017-4-3 22:33
Thanks DJI-Jamie for the response.
The VPS readings are OK. It is the barometric height that gives a ~3m to high value. As you can see in the picture, the quad is close to the ground. The VPS reads 0.8 m, but the barometer 4.3 m.
Is a deviation of 3 m of the barometer normal?

It is within tolerance. If you're not experiencing any flight performance from the aircraft, then you would be fine.

DJI-Jamie Posted at 2017-4-4 15:34
It is within tolerance. If you're not experiencing any flight performance from the aircraft, then you would be fine.

The concern I have over this discrepancy is whether the drone will fly into the ground if I were to fly behind my barn at 2 meters indicated and  the actual decreased while it was momentarily out of my view. The same concern arises whenever it is in a situation where my perspective will not let me accurately determine true height above ground. I would think that downward sensing would update above ground height whenever the AC goes to two meters or less.

fansad6ed80e Posted at 2017-4-4 17:47
The concern I have over this discrepancy is whether the drone will fly into the ground if I were to fly behind my barn at 2 meters indicated and  the actual decreased while it was momentarily out of my view. The same concern arises whenever it is in a situation where my perspective will not let me accurately determine true height above ground. I would think that downward sensing would update above ground height whenever the AC goes to two meters or less.

In reality, you would want to ensure that you have far more than 3m of clearance as that's not a lot of margin of error to begin with in the grand scheme of things. Depending on where you are, the altitude density can affect the barometer along with uneven ground that makes it seem like there's a height change when there isn't. The barometer will calculate it height based on the home point. It will not notice a height change should you fly over a cliff, for instance. If you feel as though it really is an issue, sending it in for evaluation would be the only thing I could recommend since you've already tried calibrating the unit and resetting to factory defaults.