GerdS said:
Gerd, if you are using 3.0.6, would you be willing to try the experiment of a fast forward run and then let go of the sticks, observing altitude change?
Uncommanded altitude drop in v3.06
- Thread starter damoncooper
- Start date
Gerd, if you are using 3.0.6, would you be willing to try the experiment of a fast forward run and then let go of the sticks, observing altitude change?
I have done this already, full speed in both GPS and ATTI mode, up to almost 20m/s and then stop. However, I rarely release the sticks from one second to the other but go back to center positions more smoothly to avoid any aircraft instability, as I am used to from scale-like flying my helicopters. I never have seen any altitude change exceeding the normal range of maybe 2m.
Regards, Gerd
Regards, Gerd
GerdS said:
A 2m drop is still pretty significant and I would not call that "normal".
I'm guessing by the time delta between my request and your post you are going on memory and judging 2m vs 4m drops over distance from memory is probably less than exact. Actual Flytrex data as I posted above shows the magnitude of the issue as more of a 3-4m drop.
But even a 2m drop is significant, especially if you are flying a few meters above water.
I am always flying FPV with the Phantom and due to my experience 2m was the maximum altitude amplitude I have seen when hovering. No jumps, just smooth change due to maybe wind influence and so on. I also have never seen a significant loss in altitude after stopping, neither visually nor by looking to the altitude as displayed by the iOSD mini.
Flying just a few meters over water is always a high risk, especially as a barometric sensor is not that exact due to atmospheric influences. For such purpose I would recommend using a different copter, equipped with a radar or ultrasonic altimeter.
Regards, Gerd
Flying just a few meters over water is always a high risk, especially as a barometric sensor is not that exact due to atmospheric influences. For such purpose I would recommend using a different copter, equipped with a radar or ultrasonic altimeter.
Regards, Gerd
I've just been flying this afternoon in a local park. I was practising flying circuits, at around 4m height. The battery was pretty low (2nd level warning - double blinking red). I'd got the P2 about 30m away from me and did a sharp right turn. It rapidly lost height (no throttle input from me), bounced off one of the prop guards on the ground, and flipped over on the grass. Absolutely no damage done, but slightly unnerving. It's a pretty windy day, so I don't know if it was just caught by a strong gust of wind (I guess prop guards must make it catch the wind more), or something to do with the very low battery state, or even the firmware. I guess the moral of the story is to fly higher?
ianwood
Taco Wrangler
I think up to 2m during abrupt transitions is not unreasonable. Flying hard without easing the sticks, you could replicate a close to 2m vertical change before 3.06. With easing, it should be a lot less though.
In general, I think easing is important and will probably mitigate the issue. When you push a car to its limit, you ease between accel and braking so not to upset the balance even if you're doing it very fast. I think the same should be true with flying the Phantom aggressively. Apply a little residual pressure when letting off so the sticks smoothly transition to their center point.
In general, I think easing is important and will probably mitigate the issue. When you push a car to its limit, you ease between accel and braking so not to upset the balance even if you're doing it very fast. I think the same should be true with flying the Phantom aggressively. Apply a little residual pressure when letting off so the sticks smoothly transition to their center point.
I guess I'd consider a return to center position a non-aggressive flying maneuver and full opposite stick an unreasonably aggressive flying maneuver.
Regardless, I agree that if it were 2m, I could live with it. But it's measurably more severe than that, as data in my original post demonstrates. It's on the order of a 6m+ drop.
For simply centering sticks I do consider that an unacceptable drop.
Regardless, I agree that if it were 2m, I could live with it. But it's measurably more severe than that, as data in my original post demonstrates. It's on the order of a 6m+ drop.
For simply centering sticks I do consider that an unacceptable drop.
ianwood
Taco Wrangler
So again, what is your vertical gain, did you test with increased value?
Another thing to think at:
I also was very unsatisfied with position and altitude hold acurracy when I got my Phantom 2. Was used to much more precision from my other copter.
Some day I touched the Phandom landing skids while hovering and was frightened about the amount of vibration to feel. So I tried balancing the props statically but with my methods there was no inbalance to mention about.
Then I tried it with dynamic balancing and afterwards I didn't recognise my P2 again. Much more stable and precise compared to the previous conditions!
To do dynamic balancing:
Place the Phantom on a table, just mount a single prop, start the engines, hold the Phantom with one hand at a skid and then increase power until about takeoff rpm.
If you can feel almost no vibration the prop is ok. Otherwise next step is looking into the rotating prop from the side. Normally you should see just a single line made by the prop blades. If there are two lines or the line is not sharp, then this prop is warped and needs to be replaced. If the line is sharp, attach a piece of tape to the lower side of one blade and try again. If vibration goes worse, then move the tape to the other blade. Optimise the tape size until vibration is at minimum.
Remove the prop, mark the motor numer on it, so it will always be mounted onto the same motor and repeat these steps for all other props. Finally test with all 4 props in place and vibration should be significantly lower than before.
In my case I found just a warped prop which got replaced by a spare prop.
Regards, Gerd
Another thing to think at:
I also was very unsatisfied with position and altitude hold acurracy when I got my Phantom 2. Was used to much more precision from my other copter.
Some day I touched the Phandom landing skids while hovering and was frightened about the amount of vibration to feel. So I tried balancing the props statically but with my methods there was no inbalance to mention about.
Then I tried it with dynamic balancing and afterwards I didn't recognise my P2 again. Much more stable and precise compared to the previous conditions!
To do dynamic balancing:
Place the Phantom on a table, just mount a single prop, start the engines, hold the Phantom with one hand at a skid and then increase power until about takeoff rpm.
If you can feel almost no vibration the prop is ok. Otherwise next step is looking into the rotating prop from the side. Normally you should see just a single line made by the prop blades. If there are two lines or the line is not sharp, then this prop is warped and needs to be replaced. If the line is sharp, attach a piece of tape to the lower side of one blade and try again. If vibration goes worse, then move the tape to the other blade. Optimise the tape size until vibration is at minimum.
Remove the prop, mark the motor numer on it, so it will always be mounted onto the same motor and repeat these steps for all other props. Finally test with all 4 props in place and vibration should be significantly lower than before.
In my case I found just a warped prop which got replaced by a spare prop.
Regards, Gerd
GerdS said:
This was observed by testers in the Beta group with a variety of gain values.
I'm currently running
Basic Gains 120, 120, 120, 200
ATTI Gains 220, 220
200 may be a bit too high, ideally it's between 170 and 190 from my feeling for a fully equipped P2. I have added more info to my previous posting while you have already answered, please read again.
Regards, Gerd
Regards, Gerd
GerdS said:
Thanks. Playing with gains is further down my todo list, i just never seem to find the time.
I think I'm good in the vibration department, as I only hand catch and I've got no vibration to speak of. I have statically balanced props. But I appreciate the dynamic balancing/matching tip.
Duly noted on watching for warped props.
