New Solo, first thoughts, and a crash

Peen · Dec 11, 2017

Decided I wanted to play around with this a bit tonight. Took off the standard solo gimbal, and just covered the circle hole with some tape.

In loiter mode, it still did the wobble. Uploaded a video to see and hear.

Went to Fly:Manual (alt hold) and it was flying perfectly smooth (unlike the GPS wobble) and even hit 55mph with no wind. No video though, was flying one hand and was too difficult.

So whatever it is, it only happens in GPS mode. Sorry the videos are night, but only time I got to test. The high speed one you can hear the wobble in the motors better then seeing.

Did a little hover test again as well. There is no wind in any of these videos or logs.

Log - Dropbox - 2017-12-10 20-08-39.bin

- Hover test
- Loiter wobble

Peen · Dec 11, 2017

Ordered a new motor pod from B&H since it's only $30 shipped no tax.

Did a vibration test, and noticed the pod that received the brunt of the crash and got dirt in it, has a lot more vibes. It doesn't sound too different from the others, but you can feel it. Video of test attached. Bad motor is 3rd test (pod 2)

RichWest · Dec 11, 2017

You can rebuild (bearings) the motors, so don't toss it out with the trash just yet.....$4 and a little time.

Peen said:
Ordered a new motor pod from B&H since it's only $30 shipped no tax.

Peen · Dec 11, 2017

I might do that as well for a spare, but this motor looks pretty bad anyways. It was all dinged up from hitting the ground. $29.99 shipped I can deal with for now

I just found out the Green Cube comes with the Sorbo for vibration isolation. So by the end of the week I will have a new motor, Green Cube and isolated board. Hopefully the motor fixes the wobble, and get the benefits of GC and isolated board for stability.

Peen · Dec 13, 2017

Got the Green Cube today, install went well. Also put in the sorbo for vibrations. Everything else the same, and it flew good! No more wobbling at high speed. Not sure what actually fixed it since technically changed two things. Hovering seems about the same though, but I thought it was fine before.

Was thinking a motor could be bad because I kept hearing almost a grinding sound when doing quicker punches of acceleration. . Also, from the motor tests in mission planner, it just sounded weird. It did this before the Green Cube as well, but sounded much worse after the Green Cube. I noticed on the arm it crashed on, the LED cover was loose feeling so I popped it off and the grinding sound was gone. Bad news is I have a bunch of covers and they are all loose, so the arm is just tweaked ever so slightly. I put little thin pieces of scotch tape on bottom of LED cover to fix it, kind of like fixing a wobbly table by putting something under a leg. Feels good and solid now!

Peen · Dec 14, 2017

Can anyone take a look at this log? Still getting IMU Mismatch = Warn. Tried a couple IMU calibrations, but maybe it's still vibration related.

Dropbox - 2017-12-13 21-20-44.log

This is .bin - Dropbox - 2017-12-13 21-20-44.bin

Peen · Dec 14, 2017

If I only knew how to read the logs. Seems like since Green Cube and vibration isolation it's hover actually got quite a bit worse then stock PH 2.0.

The IMU Mismatch is pretty strange since I never got before Green Cube. Have tried plenty of IMU calibrations using Solex but it still remains.

Link to a log that's just hovering, basically no wind - Dropbox - 2017-12-14 15-33-09.bin

Vehicletype ArduCopter
Firmware Version V3.5.4
Firmware Hash 284349c3
Hardware Type
Free Mem 0
Skipped Lines 0
Test: Autotune = UNKNOWN - No ATUN log data
Test: Brownout = GOOD -
Test: Compass = GOOD - mag_field interference within limits (4.03%)

Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = GOOD -
Test: GPS = FAIL - Min satellites: 0, Max HDop: 99.99
Test: IMU Mismatch = WARN - Check vibration or accelerometer calibration. (Mismatch: 0.79, WARN: 0.75, FAIL: 1.50)
Test: Motor Balance = GOOD - Motor channel averages = [1230, 1251, 1231, 1228]
Average motor output = 1235
Difference between min and max motor averages = 23
Test: NaNs = GOOD -
Test: OpticalFlow = FAIL - FAIL: no optical flow data

Test: Parameters = FAIL - 'THR_MIN' not found
Test: PM = FAIL - 33 slow loop lines found, max 24.34% on line 94799
Test: Pitch/Roll = UNKNOWN - 'BarAlt'
Test: Thrust = GOOD -
Test: VCC = GOOD -

pyrate · Dec 14, 2017

Did you update the parameters, reboot
calibrate level reboot
Calibrate compass reboot?

Peen · Dec 14, 2017

Hey Pyrate,

Yes on all that, many times. I always reboot after any setting change, IMU cal, compass cal, etc.

Charles Lamoureux · Dec 15, 2017

IMU calibrations can only be done via SOLEX...I ‘ve just come back to SOLO and I thought the original 3DR app is where you did the IMU calibration...Can someone provide a refresher for me thanks

Peen · Dec 15, 2017

You can do it in 3DR app as well, it's still there next to compass cal if I remember right.

Charles Lamoureux · Dec 15, 2017

Peen said:
You can do it in 3DR app as well, it's still there next to compass cal if I remember right.

That’s what I thought and its not...Compass and Gimbal that’s it

Peen · Dec 22, 2017

Made some changes to gimbal and unsheathed my gimbal data cable and HDMI cable. Also, I put some moongel underneath my Green Cube just to see what would happen with my IMU Warn. The moongel seemed to help quite a bit.

Changed my mRo with V2 shield to HERE. In my testing area, I used to get 25-28 satellites, 0.5 HDOP, Galileo activated with mRo. With HERE, I got 18-19 and 0.6 HDOP. Need to do more testing, but nothing feels very different. Actually like the look of the HERE, looks purposeful.

Here is a quick YouTube clip just to show how it captures video with the gimbal mods and no editing. Half way through I do some color correction and stabilization just to show the difference. I am not using an ND filter yet, but I do need one ASAP!

Peen · Dec 22, 2017

Still playing with videos, no ND filter or music yet. Put moongel in the gimbal itself, but I think it made it worse. This video is with a little warp stabilization.