Problem with Motor Power Balance

Lynxmotion UAVs / Drones
1

I’m new to the world of UAVs, but not a total noob at working with electronics.

I’m working my way through the setup of MultiWii 2.4 on the Quadrino in incremental steps. At the moment, I have a simple setup, as follows:

PC <–> USB <–> Quadrino <-- (AR610 Rx) <-- (DX6 Tx)

I’ve got my Tx bound to the Rx, and I can ARM/DISARM the Quadrino from the Tx (I know this based on the ARM indicator in the WinGUI turning ON/OFF accordingly).

I’ve calibrated the ACC/GYRO/MAG, and according to the sensor graph all readings look kosher:

ACC : X/Y ~ 0, Z ~ 512
GYRO: PITCH/ROLL/YAW all 0
MAG (heading) : steady

Basic config settings : MIN Command = 1000, MIN Throttle = 1100, MAX Throttle = 2000

Configured my Tx endpoints to range from ~1000 to ~2000.

GPS is disabled (since I’m just testing, and I’m inside anyway).

Set the AGGRESSION factor to minimum (-70).
Set the ANTI-WOBBLE factor to minimum (0).
Disabled in-flight ACC calibration.
Fail-Safe is OFF. Battery monitor is OFF.

I’m just testing the current configuration through WinGUI (no airframe, no props, no ESCs, just the aforementioned sub-systems).

When the system is DISARMED, the motors read 1000 for FRONT L/R and REAR L/R.
When the system is ARMED, the motors read ~1150 for FRONT L/R and REAR L/R. At this point, if I do nothing, and just let the system sit idle (no throttle, no inputs, no movement), I can watch the FRONT R, REAR L motors settle to 1100 (exactly), while the FRONT L, REAR R will slowly climb and settle at ~1350 (+/- 30 units b/w F_L, R_R).

If I move the throttle to maximum, then FRONT L will reach 2000, the REAR R will reach ~1980. But the FRONT R, REAR L motors will only reach ~1600 - 1800. If I leave the throttle at maximum (no inputs, no movement), then I observe the same behavior; the F_R, R_L power drops to ~1500, for both, while the other two motors climb, maxing out at 2000.

I don’t understand the source of this behavior. Based on the behavior, this looks like a PID problem to me (just a guess). Not that I’ve tried every configurable property, but nothing I do seems to change this behavior. Despite the Quadrino being on a flat surface, with additional inputs from the Tx, and no movement, the system deviates from level flight (and migrates into a YAW).

Also, I observed that the “config.h” enables the “MPU6050”, but the FCT defaults to the “MPU9150”, even though it’s not referenced anywhere in the source code. Not sure if this matters, just an observation.

Any guidance on this would be much appreciated.

Thanks in advance.

2

I figured it out. Dumpster diving in the code, I studied the PID control loop a little further. The “conf.angleTrim” parameter is used to adjust the error in the Proportional term of the PID loop. But the default values of the angle trim are off. So I searched around for where the angle trim is set, and it turns out that you have to use a series of magical stick commands to adjust the trim angle (see line 1015 in MultiWii.cpp). I’m still working to balance the motor power, but now I know how to affect the behavior correctly.
With the Quadrino disarmed, and HORIZON flight mode enabled, trim the angle applied for either pitch or roll. Arm the Quadrino to see the results of the trim. Rinse and repeat until the power applied is level.
The following incantations trim the angles:
THR HI, YAW CE, ROL CE, PIT HI -> increase PITCH trim angle
THR HI, YAW CE, ROL CE, PIT LO -> decrease PITCH trim angle
THR HI, YAW CE, ROL HI, PIT CE -> increase ROLL trim angle
THR HI, YAW CE, ROL LO, PIT CE -> decrease ROLL trim angle

3

Hi TheSheriff,

Sorry for the long wait regarding this forum post.

Regarding the values sent to motor, they will be off and it’s normal.
The Flight Controller will try to stabilize the copter but without any reaction (since there is not copter) the values will build-up. You can read more HERE.

Regarding the sensor, it’s normal as the MPU6050 use the same internal but mix the Magnetometer as well.
All the sensors inside the chip are just like external ones and answer to a different i2c address.

:slight_smile:

4

Hello Team,

I am also facing almost the same issue.
(Drone is placed on the flat surface)
When the system is DISARMED, the motors read 1000 for FRONT L/R and REAR L/R.
When the system is ARMED & if I do nothing, and just let the system sit idle (no throttle, no inputs, no movement), I can watch the motors read 1000 for FRONT L/R and REAR L/R.

If I move the throttle to maximum, then FRONT L will reach 1850(stays constant), the REAR R will reach ~1600(& continuously varies in between 1592 to 1600).
But the FRONT R and REAR L motors initially reaches to ~1700 then If I leave the throttle at maximum (no inputs, no movement), then I observe FRONT R and REAR L motors power immediately starts to reduce and drops to ~1200, for both, while the other two motors climb, maxing out at ~1600.

I don’t understand the source of this behavior.
Based on the behavior, this looks like a PID problem to me (just a guess, if my guess is correct please guide me hoe do I modify the PID values?). Not that I’ve tried every configurable property, but nothing I do seems to change this behavior.
And because of this behavior when I tried to fly the drone its tilting on one side and its out of control.

Any guidance on this would be much appreciated.

5

This is normal in MultiWii you can read more about it HERE.

6

Ok.,

But the drone is not stable and I am unable to fly it, so what I need to do to make the drone stable.

Please suggest

7

What do you mean by “not stable” exactly ?
Do you have any flight mode selected in WinGUI ?

8

Can you please suggest what should be the standard PID values if I am using the Quadrino Nano flight controller?

9

Not stable meaning when I am trying to fly the drone its getting lift like for 4 fit and then tilt to one side and going out of control.

No I didnt select any flight mode…

10

The FCT have a PID configuration which should get you on track most of the time.
Frame size / motor / etc will affect the PID settings.

You need a flight mode if you want any kind of stabilization otherwise you are on your own. (no stab)
Is the copter reacting to inputs (transmitter) the right way ?

11

Yup, the copter is reacting the inputs(transmitter) right away.

Which type of flight mode i need to setup?

12

All Flight Modes are listed in the Quadrino Nano user manual HERE.
We suggest putting “HORIZON” on an auxiliary switch of your transmitter and try it, this is or recommendation.