[Kettentiger]

Well my Lipo is too big for the bottom plate......

[Mateusz]

I have another Idea: I put it away from the Powerplate und put it under the Topplate turned 180 degres. In the GCS i can tune it. 

If you want to use external Mag in that GPS unit, it must be mounted firmly and you should never change it's position after calibration. In that case it should not be wobbling or be mounted on a battery.

You may want to start flying in Complementary and then try step by step with GPS unit that has external MAG and is on a stand far from electronics and currents. If you have crappy MAG readouts, then GPS is not gonna give you benefit. It's not like GPS from mobile phone which does a lot of extra stuff people are not aware of such as using cell towers, filtering and smoothing and even WiFi signal to get better accuracy.

Adding GPS and Mag wont make it fly better if it's receiving consistent bad read-outs, in that case you are better off flying Complementary.

By the way, I have the same frame (F450) and I mount my battery bellow bottom plate, though I use CC3D on it

[f5soh]

By the way, I have the same frame (F450) and I mount my battery bellow bottom plate, though I use CC3D on it

Yes, i think there is all space necessary for battery bellow power plate.

[Kettentiger]

This is my Lipo and it does not fit in the middle of the Frame.

http://www.hobbyking.com/hobbyking/store/__56840__Multistar_High_Capacity_4S_5200mAh_Multi_Rotor_Lipo_Pack.html.

[Kettentiger]

So all sources of interference is minimized , but the FC still spins : As soon as I move the Throttlestick to 2mm , the motors start ( this is normal) but then give the motors full stuff and the copter is uncontrollable .



But the guys from Rcmaster have come forward and that I should send them a video , where you can observe the behavior .... Let's see what they Write to......

[f5soh]

How it works using the basic complementary algorithm ?

You need to remove all "gps" modes in flightmode tab.

[TheOtherCliff]

It goes straight up when you jump up, that is a good sign.  It is normal for it to tumble a bit when falling if you stop the motors.

In that regard, I didn't really see anything wrong in the video.  Jump up to just 3 feet.  A slow lift off is difficult, and you must NOT correct roll/pitch until it is up off the ground.

If you understand and accept the risk, I would put on a heavy coat for arm protection, and hold it level with my hand while starting a hover at lowest power.  That will tell you if it is close to correct.  It is probably best if you set yaw PIDs all to zero before doing this test, else when you hold it, you must hold it in exact yaw position it had when first giving it power.

[Kettentiger]

How it works using the basic complementary algorithm ?

You need to remove all "gps" modes in flightmode tab.

Without the GPS Mode, the complementary mode the copter flys correct.

[Kettentiger]

It goes straight up when you jump up, that is a good sign.  It is normal for it to tumble a bit when falling if you stop the motors.

In that regard, I didn't really see anything wrong in the video.  Jump up to just 3 feet.  A slow lift off is difficult, and you must NOT correct roll/pitch until it is up off the ground.

If you understand and accept the risk, I would put on a heavy coat for arm protection, and hold it level with my hand while starting a hover at lowest power.  That will tell you if it is close to correct.  It is probably best if you set yaw PIDs all to zero before doing this test, else when you hold it, you must hold it in exact yaw position it had when first giving it power.

Ok......
That I have to test where I have a lot of space.
He wants directly to get up high.

[cato]

About vibes, you can change the Accel filter value. Put something like 0.05 or more.
Double sided foam or more layers should works also.

Just see you previously stick board on bottom plate, where the strong current are.
Bottom plate is used to power the four controllers : Not the best place for Mag.

That is a very good tip. I was wondering why my clone is still a bit worse than the original: the original resides in one of mthreads cases, which are a bit like a damper, while the clone is screwed down to the frame with no dampening. I will change that and see what happens (still need to wait for some spare parts for my cinetank).

edit: just did a quick test with my Nighthawk, that even improved the original Revo's performance.

[f5soh]

@KettenTiger Another strange thing i see from your log file is Gyrobias, around 7 in one axis.

Can you redo Gyro calibration without move frame at all and see how are values after calibration ?

[Kettentiger]

Which tool do you read from my log?

I will try it when the weather is better here.......Only rain and storm....

[f5soh]

Which tool do you read from my log?

You can replay log using GCS, on bottom/right select Log file

I will try it when the weather is better here.......Only rain and storm....

Check current Gyrobias in UavBrowser and redo Gyro calibration, no need good weather, only a stable desk