Random roll and pitch after enabling gps

[zied923]

This issue is discussed before but i don't see any solution so im opening this topic again maybe somene figured out how to solve it.
Im on inav 7.1. 2
Speedybee f405 v4
Gps M10Q-5883
I have 9/10/11 sat lock (indoors)

Uploading VID_20240830_180104.mp4…

The issue is roll and pitch become not stable after enabling gps, drone drift badly and crash with gps enabled. Without gps all is good.
I have lidar installed the tf luna and it hover almost perfect with surface / angle mode. No any issue.
Wire are isolated and gps mounted in the arm to keep it far of interference.
Campass showing the right north no issues.
What could be wrong? Any ideas?

( i joined pictures and small videos)

[zied923]

VID_20240830_180104.mp4

[zied923]

[zied923]

VID_20240830_183935.mp4

[sensei-hacker]

Your video shows roll deviation of less than 0.1 degrees. Turning upside down would be 180°, so what your seeing is less than 1/1800th of that. Literally within the rounding tolerance.

A teenager driving by with thumping bass can vibrate your craft enough to change the rounding at 0.1°. That's not a problem.

Similarly for pitch - half a degree. That will have absolutely no effect on flight whatsoever. Just compensating for a light breeze is several degrees.

If you are having some flight trouble, please provide information about that. What flight mode, precisely what you saw occur, such as "forward pitch tendency upon quickly increasing the throttle", preferably with video of the problem. Also let us know - is this a blimp? A submarine? A glider? A wall crawler? The reason I had asked you to fill out the template is because we actually need that information in order to help you.

If your submarine has poor yaw authority, that's a completely different thing than if your AR Pro has pitch oscillations. We can't help without having the information.

[zied923]

Why you are having bed attitude with me??.? I told you im new with github and im new with fpv ,and i don't know how to use github properly, and What u think it is? Its a quadcopter, i mentioned all details , doesn't matter i will fix it by myself.

[sensei-hacker]

What u think it is?

We have absolutely no idea what it is, for as long as you refuse to tell us. As long as you refuse to fill out the form or tell us anything about the vehicle or the flight, we have no way of having any idea what problem you might be having.

I asked you before, in the other issue you hijacked, whether it's a submarine, a blimp, a wall crawler, a quadcopter, a tailless plane ...
You chose not to answer. You chose to ignore two requests to please fill out the template so we would be able to help you - so we would know what kind of vehicle you have, and what problem you are trying to solve.

Unfortunately, we can't read your mind, so until you tell us about the problem, we can't fix it for you.

I'm trying really hard to help you, but of course I can't without any information.

[zied923]

The issue is abviously so clear but your making of it a big deal! With your rules or whatever, what template i have to fill? I don't see any.. Im using a phone, i told you im new IDK HOW TO USE GITHUB, and i put all the info, but really doesn't matter, your making a simple question a big deal, i know you know what im talking about, its just i know the kind of person are you, the kind of ppl who complicate everything and making of it a big deal.. Your waisting your time and mine with nonsense, have a nice day and last time i will post here and i will leave github also.

[sensei-hacker]

The issue is abviously so clear but your making of it a big deal!

What you've told us so far is that you crashed. That's the info we have so far. There are about 72,000 different things that can cause different kinds of crashes.

What happened before the crash?
Did it pitch forward? Did the motors cut off? Did it run full speed into the side of a house? Did your video feed cut off?

What flight mode were you in?

What kind of quad are we talking about?

[OptimusTi]

wow...

[Jetrell]

@zied923 There is a possibility this is related to a known issue. As referenced here.
I'm not totally convinced this hasn't crept in as a side effect of work added to 7.1, even though much of that testing in that link, was done with the master build of INAV 8.0.

Regardless, the more information you can add, the better.
Things like a 1) Log 2) Diff settings file 3) Was this flight outside ? 4) How many tests did you conduct ? And as @sensei-hacker said, video footage is always helpful too.. We need to be sure it is related to the above. And not just caused by poor position accuracy (2.5m isn't great). You did mention something about being indoors.
Something I would suggest. Is for you to load a build of 7.0 on your quad. And report back if it has the same issue.. And also providing a log for that test as well would help.