[coolbuck]

I think I know why it rolls to the right.

I unhook the flight controller from the copter (propellers are not installed).

And I hook
Chan 1 (pitch) of the receiver into motor 1
Chan 2 (roll  ) ---------------------------2
Chan 3 (throttle) ------------------------3
Chan 4 (yaw)  ---------------------------4

Connect the battery.  Without moving the sticks while the right stick is in the middle position,
the motors 1 & 2 running. 

So my guess is the default is the right roll. And that's why it rolls to the right.

How do I fix this?

[hwh]

...However, it flipped to the right or rolled over to the right....

Hard flipping on takeoff is usually the flight controller not mounted pointing in the correct direction or if it isn't then the firmware not told the way it's actually mounted.  https://librepilot.atlassian.net/wiki/display/LPDOC/CC+Attitude+Configuration is the wiki page about that.  The video at https://librepilot.atlassian.net/wiki/display/LPDOC/PFD#PFD-MyPFDlooksreversed shows how to check to make sure the board is mounted pointing correctly (or the virtual rotation has been set correctly).

Or as others have said, a backward motor or wrong prop but you've already checked that.

...One thing I found a litter strange is that the input config shows: Throttle  at Chan3 Roll at Chan4 Pitch at Chan1 Yaw at Chan 2 Should it be different?...

That depends entirely on your transmitter and how you wired the receiver to the cc3d.  As long as when you move a stick the input tab shows the same one changing the order is correct.  And if the transmitter is setup properly without mixing then only the stick you move should change in the input tab.  If two channels move when you only move one stick axis then the transmitter has a mix in it that needs to be removed.

...Without moving the sticks while the right stick is in the middle position,
the motors 1 & 2 running. ...

With a stick or sticks in middle position the corresponding motors should be running.  If yaw is also in the center, springloaded position then it's motor should be running too.  With the motors directly connected to the receiver all four motors should only be off if both sticks are in the lower left position.  With both sticks in the center position all four motors should be running.

[coolbuck]

Thank you for the info.

I calibrated the ECSs and aligned the flight controller the direction you mentioned.   However, it still does not work.  The propeller at NE does not spin until I increase the throttle higher.  Then it flipped to the NE corner.  Maybe I think because the corner NE is the weakest compared with other corners.

Should I calibrate more?  or replace the NE motor?  or something else?

I have a feeling the it's not going to fly.

Thanks

[hwh]

You're right, it probably won't fly right if all the motors don't start at the same throttle setting.  I can't think of how that could happen, you set the minimum setting for each motor individually in the wizard.  Even if one motor took a little more to start spinning that should compensate.

[TheOtherCliff]

The way you should imagine that it works:  If board is mounted unlevel in copter or copter is not sitting perfectly level on ground, then when you start motors, the "top" motor does not need to run at all until it is flying and closer to perfectly level.

With motors running slowly on the ground, it will try harder and harder to make it level and so the "bottom" motor(s) will spin faster and faster.  If you try to take off like this, it will flip before the bottom motors slow down enough.  This is "PID-I term windup".

Knowing that it does fly and so the motor / prop directions are correct, etc. and assuming that you have default options set (not enabling always stabilize when armed, etc) (not even from some uav or cloud config), then the solution is that as soon as you start the motors, you should jump it up to knee-waist high instead of sitting on the ground and winding up.

[TheOtherCliff]

I think I saw that you did ESC calibration, but read this to make sure you did both parts (endpoint calibration where you power the ESCs up and wait for beeps, and neutral setting in the GCS):
https://librepilot.atlassian.net/wiki/display/LPDOC/ESC+Calibration

After that...

For a small and low power quad, if you think you can do this safely, you can use a lipo with less cells (2s instead of 3s), put on a winter coat and glove for protection and grip the quad very very very firmly and very level and give a very small amount of throttle.  It will either try hard to twist out of your hand or or not.  If not, you can relax your hold slightly to let it level itself more and more.  Remember to also move the quad in the yaw direction to help the motors level out.  At the end, it is stabilized as if flying in your hand.  This will prove that it does stabilize.

[coolbuck]

Thanks for the advice.  I'll get a new battery with 2s and make connection and try it.

I bought another similar quad on ebay and the two behaved very similarly, ie rolledl to the right and flipped.

I think it could cause by 3 things:

1. The setting in the firmware. In GCS,  I tried to calibrate and stabilize & level out also.

2. Something is wrong with either Tx or Rx or both.  I use the Tx Flysky FS-T6 and Rx fs-r6b with 6 channels.  Anything might be wrong with this?

3. I have one confession to make:  I have not soldered all the wires and have not used the power board.  So far I use the wire connectors to twist the wire connection together.  Will this create some sort of interference or other problems?

Please advise.
Thanks

[Mateusz]

1. The setting in the firmware. In GCS,  I tried to calibrate and stabilize & level out also.

There is no software problem, it works for many people. You may have other problem.

2. Something is wrong with either Tx or Rx or both.  I use the Tx Flysky FS-T6 and Rx fs-r6b with 6 channels.  Anything might be wrong with this?

Check if you're not using any trims, reset trims. Don't use any curve in the TX. Maybe set completely new model with default settings.

3. I have one confession to make:  I have not soldered all the wires and have not used the power board.  So far I use the wire connectors to twist the wire connection together.  Will this create some sort of interference or other problems?

If the connection is week you might get problem in powering up motors, some might not be getting enough current. I don't quite understand how you connected ESC, just make sure connection is solid.