What are some best practices after initial take off?

[umanbean]

When you see the very top of the tree in you view finder, is upper tip of the tree in dead center in your viewfinder, it that accurate?

Best practice, and safest, would be for the tip top of the tree, and all the trees in the vicinity, to be below the horizon (in the distance). I keep the center crosshair in Go4 turned on.

 

[EsaT]

In other words, if I marked it at 80 feet, and I ascended 5 more feet would I clear the top of the tree?

5 feet is way too small for safety margin.
20-30 feet would be more like it.

 

[okw]

Best practice, and safest, would be for the tip top of the tree, and all the trees in the vicinity, to be below the horizon (in the distance). I keep the center crosshair in Go4 turned on.

That works in flat areas. But if you have mountains in the background, that method doesn't work.

Making sure that the camera is horizontal and you have the center marker turned on, as you mentioned, if an object is below the center marker, then you're higher than that object. Giving yourself 20-30 feet, as @EsaT suggested, should be plenty accurate.

 

[Jamesz]

Checking that the drone's physical orientation matches that of the map icon is perhaps the most vital of all pre-flight checks after physical checks, like prop and battery security, clean sensors etc. Then assuring you have a correct home point update, briefly test all 8 stick movements when above head height. Any compass error warnings flag up problems with the TO, move rather than calibrate.
These steps eliminate most common drone problems.

 

[LakeTravis]

I have always flown up about 10 feet, hovered, checked satellites, checked ALL inputs (fly left, right, up, down, circle, just a bit), make sure the drone is stable, make sure the home point is recorded, make sure all the readings on the screen look correct, then go for it.

 

[Chaosrider]

I read somewhere that when you first fire up the props you should hover in place until home point has been updated, but what then? Someone on Youtube also pointed out you should rise up 25 feet and let the drone snap a pic for its' return home location assuming it's an Air 2s or higher with percision landing. Others say test out the Yaw in both directions before taking off. Any other suggestions? What happens if you dont go up to 25 feet, is the downward home pic stored in memory still being taken?

I do some testing like this when I first get a new drone, but with my proven Minis, I don't worry about it.

With the Minis, I don't take off until Ms Fly tells me that the homepoint has been updated.

Curiously, the FPV doesn't seem to set the homepoint at all until after takeoff. But then, I just flew it for the first time yesterday, so I may be missing something.

TCS

 

[Chaosrider]

Take off: both sticks inwards, propellers start spinning, left stick up, shoot it into the sky 50-60 meters or so to get rid of the noise as soon as possible and start flying to wherever I need to be.

By the way, I am always impressed on how fast the Air 2S is ascending. It is like it is catapulted into the sky.

That was exactly my response to my first FPV flights yesterday...and I'm still in Normal mode!



TCS

 

[Xitor]

When you see the very top of the tree in you view finder, is upper tip of the tree in dead center in your viewfinder, it that accurate? In other words, if I marked it at 80 feet, and I ascended 5 more feet would I clear the top of the tree? Or does that thing need to be totally out of the frame?

Interested in what other folks do here, but personally I'll line up in my view finder at 0% elevation dead on with the tallest tree, power line, building whatever in my flight area then usually bump that by 20 feet at least for a comfort margin. Some folks may just pick a high altitude like 250, but I still fly to the highest item because I'm not the best best judge of distance and height and like to know the safe height not just for RTH. Also, some of the areas I fly in locally have100 or 200 feet ceiling and require LAANC. Only problem with setting max RTH like this every time is if you forget and it has the height of your last flight and there are taller trees nearby today that could be bad. But also one of the reasons it really helps to have a pre-flight checklist and I always at least review current RTH and max height.

 

[fgurnee01]

I always flight up and look horizontally to determine the height of the trees etc. Making sure my RTH height is suitable.

 

[Rudiger]

I am always impressed on how fast the Air 2S is ascending

I am impressed by how fast it descends - a huge improvement from my original mavic pro

 

[MAvic_South_Oz]

I am impressed by how fast it descends - a huge improvement from my original mavic pro

Hmm, can't say I've had an opportunity to compare to other drones, but thought my M1P wasn't too bad in speed of descent.
Just checked specs though and the Air2s descent speed is supposed to be up to 6m/s, double that of the M1P 3m/s.

 

[skycam509]

Hi Ken and Yaroslav,
"Always turn on the controller, then the drone, to turn off use reverse order." I do exactly the contrary.
I always turn on the drone on its pad, then the controller, to turn off, I use reverse order.
The main reason being that when I turn on the drone, there is less possibility to have the propellers starting while I have my hands on the drone, the controller being off. The same for the way back. I shot the controller first avoiding any signal from the controller to the drone.
Do I have it all wrong?

My feeling is that if the controller is on first, the drone is going to connect to it immediately on startup. If there is no controller signal present, the drone could theoretically (but not likely) hunt for a signal and lock onto some other source it mistakes for the controller. This signal could be transmitting data that would make the drone do something unexpected. I heard a story about an old phantom that took off on its own after it landed and the controller was turned off before the drone.

 

[skycam509]

My preflight check was pretty well covered here, but here are a couple other things I do:

1. Check the camera lens for dirt or fingerprints.

2. Set my RTH altitude based on my surroundings. There are a lot of pine trees around here. Most are under 150' tall, but if they are on a nearby hill, they could be higher. I don't want the RTH altitude to be any higher than necessary because winds can be stronger at higher altitudes, and I also fly along the coast occasionally, and want to stay as low as possible to avoid any Coast Guard helicopters or other aircraft that might be in the area.

When setting my RTH altitude, sometimes I'll take off and yaw around to find the highest obstacle in the vicinity. I have the crosshairs turned on in my camera and my Function button set to snap the gimbal to horizontal or straight down. By snapping it to horizontal, you can then climb to a height where the the crosshairs see over the top of the highest obstacle and set your RTH altitude accordingly.

It's my understanding as well that if you pause after takeoff, the drone will take a downward photo of its takeoff point. To make this more effective, I try to take off from a spot with some contrasting shapes or colors so the drone has a better point of reference. A landing pad should work well for this. As for the height to climb to before waiting for this mysterious photo to happen, I feel like the heights mentioned previously (25'-ish) might be a little high. I'm wondering if the height it climbs to when you use auto takeoff might be optimum? It might be interesting to do some RTH testing using different heights.

 

[EpicFlight]

If there is no controller signal present, the drone could theoretically (but not likely) hunt for a signal and lock onto some other source it mistakes for the controller. This signal could be transmitting data that would make the drone do something unexpected. I heard a story about an old phantom that took off on its own after it landed and the controller was turned off before the drone.

No, none of this is possible, at all, in theory or reality.

DJI systems communicate over a digital protocol using RSA encryption.

Without the encryption keys, which are dynamically generated at binding. This is, at the moment, completely secure.

 

[dondupuis57]

I do. I use the following checklist, prior to take off and immediately after takeoff. I open the checklist on my phone, which I don't use for my UAV'S. Open to feedback though.

https://imgur.com/a/kGDMeQq

Now this is something us beginner need. Seem to me these checks can save mistakes that could cost you money or your new drone. Thanks for sharing.

 

[drakkor]

I read somewhere that when you first fire up the props you should hover in place until home point has been updated, but what then? Someone on Youtube also pointed out you should rise up 25 feet and let the drone snap a pic for its' return home location assuming it's an Air 2s or higher with percision landing. Others say test out the Yaw in both directions before taking off. Any other suggestions? What happens if you dont go up to 25 feet, is the downward home pic stored in memory still being taken?

My own pre-flight check procedure before ANY take off, instead of "going for it" right away like some likes to do, is:
To first be sure the bird has good satellites connection. Then record the home point on the remote and be sure I get the vocal confirmation it did. Then, both sticks inwards to have the propellers start spinning. Then while they are spinning, I visually check if the drone & props don't behave weirdly while still on the landing pad. After that when all checks in OK, I take off around 10 meters straight up from the landing pad and hoover there for a few seconds to be sure the drone has a good picture of its landing area (if needed. Some model don't do that) and test all the controls and be sure everything respond & behave the way they should. Only THEN I start my mission with more confidence in my drone.

You don't see any plane or helicopter pilot go seats on their cockpit and just go for it right away. Flying a drone should not be different in my opinion, if you are serious about your gear and the safety of people around you.

 

[edcMavicPilots]

Since the Mavic 3 I turn the drone on first to let it start getting satellites but that is less of an issue now after firmware updates than when I first got it.

 

[makneyse]

Hi Ken and Yaroslav,
"Always turn on the controller, then the drone, to turn off use reverse order." I do exactly the contrary.
I always turn on the drone on its pad, then the controller, to turn off, I use reverse order.
The main reason being that when I turn on the drone, there is less possibility to have the propellers starting while I have my hands on the drone, the controller being off. The same for the way back. I shot the controller first avoiding any signal from the controller to the drone.
Do I have it all wrong?

Yorkshire_Pud has it right. Turn on the controller first because you don't want someone else's controller accidentally connecting to your drone. It shouldn't happen but I've heard of instances where it has. Plus this is the order DJI recommends.

And regarding checklists, it's highly recommended that you have one. It can be a simple one that you create for yourself or find one from elsewhere that you make your own. I created a pretty intricate one that is meant to be used for pre-production at home (with links to various airspace/NOTAM/weather sites) and check boxes for last minute checks in the field. Plus it has lots of troubleshooting tips that hopefully will help newer pilots avoid some of the issues others have come across over the years. Full disclosure: it's for sale but considering it's 5 pages long, there's bound to be something that will benefit most people.

www.groundcontrolresources.com

 

[Meta4]

Yorkshire_Pud has it right. Turn on the controller first because you don't want someone else's controller accidentally connecting to your drone. It shouldn't happen but I've heard of instances where it has. Plus this is the order DJI recommends.

It can't happen and it doesn't happen.
What you turn on first makes no difference.

 

[snowghost]

It can't happen and it doesn't happen.
What you turn on first makes no difference.

A rogue remote couldn't be binded to your craft like that.