EN CN
setting up new gimbal: sensor auto configure not working
-
I am trying to set up a new Basecam simple BGC on a gimbal. Connection via USB established, 4s power attached. But I am already stopped by trying to calibrate the sensors. Under sensor auto, I am trying ot configure camera and frame IMU, but both don't respond to the instructed rotation. What is going wrong?
-
Somehow I got it working when I unchecked the "swap frame and main sensor" checkbox. But now the next problem arises under IMU calibratin help: it just won't accept my movements.
-
Figured out now, how to calibrate the axis': Din't realize I had to click "calibrate" each time I moved gimbal into requested position. BUT: the -Z (yaw-) axis won't calibrate. Why would that be? Any help?
-
Still desperately trying... Just for clarification, do I get it right:
X = roll-axis
Y = pitch-axis
Z = yaw axis
? -
They are not axis, but installation directions, my advise is to take some timeout and read the users manual, that will save you at the end a lot of time.
-
I've been going up and down the manual. I can't find any info on how to mount the sensor. Thing is, in the visual indications on the right hand side of the gui, everything corresponds fine when I move the gimbal, even though the sensor is not calibrated properly.
Now in a video tutorial I saw that the camera sensor was mounted 90 degrees rotated compared to my setup. I will try this setup next. -
It does not mater how you instal the sensor (as long as it is parallel/90 degrees to axis), you just need to configure it on the GUI as it is installed.
-
Thanks Garug. Good to know. But why then wouldn't the Z-axis calibrate when I rotate the yaw axis?
-
Ok, got it. I mistakenly tried to calibrate the z-axis with rotating the sensor in the yaw axis. Instead I needed to turn the system upside down.
-
A new problem arises: I want to cotroll pitch through the RC, roll should be stable. In the RC-Tab:
roll: no input
pitch: RC-Pitch PWM
Pitch is in angle mode, mix channels is on none and 0%.
But now, roll somehow follows pitch. When I tilt down, roll turns to left, when pitch up, roll moves right. How to prevent that from happening? Thanks. -
You have something wrongly configured. What FW is that?
-
I think it's the latest: 2.56b7
-
Have you changed euler angles? if you have nomal gimbal, better to start with default euler angles, but other than that, most likely something wrong on the basic setup.
-
Are the euler angles the min. and max. angles?
-
No, they are at bottom of the RC settings, so you probably have not touched them, the basic setup then failing.
-
Nope, those I didn't touch. It's Cam - PITCH - ROLL - YAW.
-
Some weeks later, I reworked the Gimbal, set the controller higher, so the frame sensor can be further away from the motor, layed out the cabling better, so less interference can occur. I also deleted the eeprom, calibrated the sensors from scratch (taped to a box).
Still I got the same 2 problems:
1) Pitch and roll (withouth yaw) work well. But when I tilt pitch down via RC, roll tilt about 5 degrees to the left. Am I bringing Pitch back up right, roll comes back to level.
2) Yaw stabilizes, but in nofollow mode, the gimbal turns slowly counter clockwise without ever stopping. In the software the compass-like arrows are also constantly moving. In follow mode, the gimbal stays stable, but forward is always on a different position when I turn on the gimbal.
Can someone help me out here? Thanks.
-
Sounds that at least you need to do better gyro calibration. Disable gyro calibration at startup, and perform it manually. make absolutely sure nothing moves when doing it. easiest just to leave the gimbal on steady floor during the calibration. It does not need to be leveled for gyro calibration.
Pictures/video of the gimbal/settings would be helpful, otherwise it is just guessing. -
Thanks Garug, will try that again tomorrow... What do you mean by "it does not nee to be leveled"? Does this say it is not so important whether the IMU is completley straight during calibration?
-
Is it possible, the sensors are broken? How can I check whether that's the case?
-
During gyro calibration the sensor needs to be only stabile. during ACC calibration leveled and stabile. ACC calibration you need to perform only once. Gyro calibration more often.
See the monitoring tab or the indicators on right of the GUI, that they move logically when you move the IMU.
Overall, you should read carefully the users manual, it explains everything pretty well. -
I am sorry to sound stupid. I now searched the manual for ACC and Gyro calibration. But I cant't find any difference in calibration. The only option I got in my software ist the IMU calibration helper. And there is only one type of calibration necessary. Could you please specyfy, how I am performing a gyro calibration only? Thanks
-
Strange that the ACC_Z is constantly around 550. See screenshot of GUI. Here with no power connected, just usb: http://we.tl/GFCy62jRHl
-
I can not see the pictures, it asks me to agree something I do not understand and will not agree.
what controller are you using, what firmware. I think the ACC and Gyro calibration have always been a separate calibrations. They should be clearly visible on GUI under the helper. See page 17 http://www.basecamelectronics.com/files/v3/SimpleBGC_32bit_manual_2_50_eng.pdf
After calibrated the ACC_Z (if that is the axis pointing up) should be around 512. that means 1 g, rotate the IMU and other axis (the axis pointing up at that time) should show this value and the two other axis around 0) -
I am using SimpleBGC GUI v2.56b7. Hardware is DYS 32-Bit Mini 3 Axis AlexMos BGC. BTW I never had any problems with wetransfer, but I understand if you don't want to agree with any terms. Is there a place on this forum, where I can upload the pictures?
"rotate the IMU and other axis (the axis pointing up at that time)": I am unsure, what direction of rotation you are talking about... -
I calibrated again and now it works more or less. But I now get a lot of I2C Errors. I tried to lay out the IMU cables seperate from all other cables. But I had to pull it through the axis hole of the Y-Motor. could that be a problem? what else? Thanks.
-
""rotate the IMU and other axis (the axis pointing up at that time)": I am unsure, what direction of rotation you are talking about..."
X up, Y up or Z up.
All other efforts are pointless until you get rid of I2C errors. You could try ferrite rings on motor cables, IMU cable, shorter IMU cable etc. -
Ok, great! Got rid of the errors with shielding the IMU cable with a ferrit ring. Now it's just one to three errors on a run. I think that's acceptable. The only thing that is not working to my wishes: Forward of the gimbal is not constant. It's always at another position insteda of forward of the frame sensor. Is there a way to tell the gimbal, what i reguarily forward?
-
http://www.rcgroups.com/forums/showthread.php?t=2243241
You should ged rid of the remaining I2C errors too, they will show on video. -
Thanks Garug. I figured, the errors only occur during startup. After theat the gimbal runs without errors, so no problem during filming.
A next challenge is, the the gimbals forward is always at a different direction, and not tied to the frame IMU. I couldn't find anything in the manual: How can I use the frame sensor (above yaw) to determin what is forward? -
There is no forward.
unless you use compass, encoders or potentiometer -
- So the camera-IMU cannot follow the frame IMU?
- Could I use the compass of the flight control somehow? -
I wonder then, why platforms like Mövi, Ronin, pilolfly, came-tv and other manage to do this? Some of them also got the AlexMos controller.
-
Encoders.
-
I see. Thanks.
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!