unstable flights' behavior for a quadcopter

Hello everyone,

I am trying to find a way to diagnose the problem (s) I am having with my drone.
My goal is to build a drone with a long flight range. Then when it flies properly, I will mount a payload such as a camera carried by a gimbal.
Since the first flight, I have observed flight behavior as well as logs (mission planner) that do not appear to be correct:

  • My most frequent observations in flights are:
    *The yaw control is not symmetrical: the drone turns more easily to the left than to the right.
    *In Loiter, sometimes it loses orientation and turn a few dozen degrees on the horizontal plane and fail to resume the initial heading
    *Sometimes motors 3 and 4 start to vibrate (visible from the ground)
    *Once the flight is over, engines 3 and 4 are very hot even burning while engines 1 and 2 are just lukewarm.

  • LOGs: https://drive.google.com/file/d/1Yt2AKft8ECzP6sGxV4lIBeX2o0veaWfa/view?usp=drivesdk

=>This link allows to download a zip named “wetransfer_ok_2021-08-11_1246.zip (136M) “

It contains 2 different logs:

Log - 02-08-2021-morning “the bad logs”

Log-02-08-2021-afternoon “the better logs”

Unloading the Logs makes it possible to observe significant differences between the motors, in particular on the RCOUT parameters, Motor rotation speed, motor current consumption, etc.

There might be other things to observe with the logs, but I’m a beginner and there are a lot of features that I don’t know the meaning of.

I’ll let you observe the logs.

The problem doesn’t end here. Indeed, sometimes without making any changes, the drone works better (or even much better).

The problems listed below then become much weaker. I let you see Log-02-08-2021-afternoon in the zip file.

Hope that you can help me.

Equipment currently mounted::

-Frame name: H4 680

-Radio: Radiolink AT9S

-Motors : TAROT Martin motor 4008 330KV

-Propellers: 15*5.2

  • ESC: VGOOD; 36A; BLHELI_32; 2-6S; telemetry

  • GPS With Compass: M8N

-Flight controller: pixhawk 2.4.8

=> Firmware: Stable 4.0.7

=> Frame selected in Mission Planner: Quadcopter/ configuration: X

-Batterry: Li-ion 6S4P; 14000mah; weight: 1150g

-Weight drone without battery: 1250g

-Weight drone without battery: 2400g

You are flying on Default PID’s and haven’t followed the Initial Tuning Parameters guide for pre-maiden flight. Connect to Mission Planner and hit Alt>A for the tuning plug-in and make those changes. Update Mission Planner to latest Beta (from the Help screen) before doing so. Then you should set the Dynamic Notch Filter parameters. Then, you would run Auto Tune. At that point you can determine if the flight characteristics need help.
Initial Tuning Process Instruction
Dynamic Notch Filter
Auto Tune

Ditch the frame. I had one and it just will not tune properly. It is not an X, or a V, or an H. Its crap pure and simple. Worst money I ever spent.

A much better quad frame is a Tarot 650 Sport. I used stock arms for a Tarot 680 Pro and some aluminum motor mounts and 5008-335Kv motors swinging 1755 Carbon fiber props.

Link to 650 Sport: https://drive.google.com/file/d/1ZhULjdTF1z1yF51NdkLTIg33NZcOvYAP/view?usp=sharing

One really good thing about the motor mounts is the through bolt in the clamp allows you to index the motor mounts to be parallel with the frame and prevent the mounts from twisting and causing motor thrusts to be unequal.

The YouTube video titled “Daya H4 680 Maiden and Crash” is quite comical. 2 seconds of flight. Unless you paid for that frame that is…

He made the rookie mistakes of not testing aileron and elevator before he took off, trying to fly a new quad in high wind conditions, and doing a maiden with every thing installed. I always go with the bare essentials and add “accessories” one at a time…

Hello,
Thanks a lot.

I will follow your advice.
Then l’ll give you the results.
hope it will work.

In the meantime thank you for your answers.

Goodbye,
JB