CGO3+ problem

[cabinfourus]

I don't get on here much these days but I thought I would pop in and see if the few Yuneec owners left might know what to do.
I have a ton of Yuneec products. Recently I started digging it all out of my building and working with some of it. I have found myself spending way too much time trying to figure out a couple of cgo3+ cameras.
The first one's problem is that it has two long beeps and one short beep. I replaced both encoders (not Yaw) on it but it still has this issue. I completely disassembled it, checked all wiring with a multi-meter, and still have the beeps. I had extra yaw sitting around so I thought I would give it a try but it sent the gimbal into a death spin. I put the old yaw encoder back on and it wants to do the death spin again. By removing the yaw encoder it shouldn't change the settings but I am not sure.
The second one's problem was a ferrule on the mainboard had come off. I stole one from an old cgo2 board and it power up nicely. The only problem it has is a constant jolt from left to right. No beeps. I have not checked the diagnostic LEDs yet but plan to do this tonight.
Now, two of the H's I own have no cameras. I am using an old cgo3 from a q500 on one but would like to repair the + to put back on it.
I own 9 of these H's and 4 of the 500s and when I heard Yuneec stopped support and service I quickly purchased a ton of parts. I actually bought a guy's stock out and have a lot I need to fix. I thought of selling it all but then what would I do to kill the extra time I have.
One last thing, I saw a video of a guy who used a 3d printed part with his yaw encoder to get around the calibration. I wanted to print one of these things but can't find the stl files. Did any of you try it to see if it worked?
I can't see it working with all the work that goes into calibrating these encoders to the magnetic fields.

 

[WTFDproject]

I don't get on here much these days but I thought I would pop in and see if the few Yuneec owners left might know what to do.
I have a ton of Yuneec products. Recently I started digging it all out of my building and working with some of it. I have found myself spending way too much time trying to figure out a couple of cgo3+ cameras.
The first one's problem is that it has two long beeps and one short beep. I replaced both encoders (not Yaw) on it but it still has this issue. I completely disassembled it, checked all wiring with a multi-meter, and still have the beeps. I had extra yaw sitting around so I thought I would give it a try but it sent the gimbal into a death spin. I put the old yaw encoder back on and it wants to do the death spin again. By removing the yaw encoder it shouldn't change the settings but I am not sure.
The second one's problem was a ferrule on the mainboard had come off. I stole one from an old cgo2 board and it power up nicely. The only problem it has is a constant jolt from left to right. No beeps. I have not checked the diagnostic LEDs yet but plan to do this tonight.
Now, two of the H's I own have no cameras. I am using an old cgo3 from a q500 on one but would like to repair the + to put back on it.
I own 9 of these H's and 4 of the 500s and when I heard Yuneec stopped support and service I quickly purchased a ton of parts. I actually bought a guy's stock out and have a lot I need to fix. I thought of selling it all but then what would I do to kill the extra time I have.
One last thing, I saw a video of a guy who used a 3d printed part with his yaw encoder to get around the calibration. I wanted to print one of these things but can't find the stl files. Did any of you try it to see if it worked?
I can't see it working with all the work that goes into calibrating these encoders to the magnetic fields.

"The first one's problem is that it has two long beeps and one short beep. "
Two long beeps and one short beep is the pitch encoder. The encoders themselves seldom go bad. The two most common problems are the various wiring problems, and the encoder magnet on the end of the motor shaft. The magnets are a problem in a couple of ways. They loose their magnetism, and they spontaneously disintegrate. Check the magnet. If it seems weak, or if there are any flakes of loose chrome (or a fully disintegrated pile of rubble), change the magnet with one from a scrap motor. Changing the magnets on pitch and roll often do not require a new calibration on CGo3 and CGo3+. The added confusion in your description is the death spin. Not sure what is causing that, but if it has lost the Yaw calibration values, it will be a problem.


"One last thing, I saw a video of a guy who used a 3d printed part with his yaw encoder to get around the calibration. I wanted to print one of these things but can't find the stl files. Did any of you try it to see if it worked?"
Never tried one. But the function was to realign the Yaw encoder with the magnetic field of the Yaw Encoder sensor magnet. Seems like you could accomplish the same thing by rotating the magnet. Not sure, since the magnetic field of the motor is also an influence on the encoder. I doubt it will work for anything other minor "front" alignment issues. If the calibration is wiped out, there is no position the system will accept. It may also create issues with the roll position or the pitch(tilt). They work as a team. Just a thought. Like I said, I never tried it.

 

[cabinfourus]

"The first one's problem is that it has two long beeps and one short beep. "
Two long beeps and one short beep is the pitch encoder. The encoders themselves seldom go bad. The two most common problems are the various wiring problems, and the encoder magnet on the end of the motor shaft. The magnets are a problem in a couple of ways. They loose their magnetism, and they spontaneously disintegrate. Check the magnet. If it seems weak, or if there are any flakes of loose chrome (or a fully disintegrated pile of rubble), change the magnet with one from a scrap motor. Changing the magnets on pitch and roll often do not require a new calibration on CGo3 and CGo3+. The added confusion in your description is the death spin. Not sure what is causing that, but if it has lost the Yaw calibration values, it will be a problem.
View attachment 26095

"One last thing, I saw a video of a guy who used a 3d printed part with his yaw encoder to get around the calibration. I wanted to print one of these things but can't find the stl files. Did any of you try it to see if it worked?"
Never tried one. But the function was to realign the Yaw encoder with the magnetic field of the Yaw Encoder sensor magnet. Seems like you could accomplish the same thing by rotating the magnet. Not sure, since the magnetic field of the motor is also an influence on the encoder. I doubt it will work for anything other minor "front" alignment issues. If the calibration is wiped out, there is no position the system will accept. It may also create issues with the roll position or the pitch(tilt). They work as a team. Just a thought. Like I said, I never tried it.

I did find an answer for my death spin. When taking the 4 screws on top of the mount out, I didn't realign the mount to the same exact place and it messed up the calibration. I recalibrated the yaw encoder and it seems to be good. I did find the part I had mentioned. I am going to link it on here for anyone else who may want to try it. I have not had time to mess with it much but did print a few out. I can print one and send it to anyone who may want to give it a try.
It wouldn't let me post the link but here is a picture of the mount.

 

[cabinfourus]

When calibrating these encoders you have to use the magnet's 4 poles. Finding the strongest point from all 4 is hard but can be done. I use a small electromagnetic graph tool that really helps. The calibration is not the hard part. Finding all the usable variables can be frustrating. I also use a modified GUI to communicate with the board of the CGO3.

 

[Vaklin]

The calibration is not the hard part. Finding all the usable variables can be frustrating. I also use a modified GUI to communicate with the board of the CGO3.

Can you explain this bit more deeply, please?

 

[cabinfourus]

Can you explain this bit more deeply, please?

It's a very long process. I spent over a year putting together a way to calibrate these things. I had help from a friend who knows more about coding than myself who put together my drivers and software. Your best bet is to start by learning how the encoders work. The Yuneec gimbals use magnetic fields to align themselves to the correct position. You also have to know the motor's step or turn counts. It is not an easy process to teach. I have never seen the software Yuneec uses (Y Ground Control) but I am sure it would make it easier than how I do it. If you can micro solder, you can move the chip from one board to another to keep your current calibration settings but again it's not easy. The Yaw encoder has to allow the camera to turn 180 degrees so it is more difficult to set. I will say this, Yuneec spent a lot of time and money trying to stop their customers from camera/gimbal swapping and are all about making more and more money. Yuneec would rather sell you another drone and camera than keep a five-year-old drone in the air. I am working on a way to adjust the Yaw encoder rather than program it. Once I verify that it works I will share the process. Honestly, I would not waste my time trying to figure this out since Yuneec is a dead brand in the US. You only have a select few who are willing to repair these things and their prices reflect the demand. If you really want to save money, buy another setup with a different name.

 

[Vaklin]

If you can micro solder, you can move the chip from one board to another to keep your current calibration settings but again it's not easy.

What IC you're mentioned? Picture if this is possible, please.

 

[cabinfourus]

What IC you're mentioned? Picture if this is possible, please.

The main chip on the board. Its not too bad but you have to be careful with the solder plates.

 

[cabinfourus]

I just ran into another cgo3+ problem. This one is tilting pretty hard to the left when turned off. The camera is vertical. It too has the roll encoder beeps. I have never seen one of these tilts this way. If you are facing the drone with camera and gimbal installed, the drone is turned off, the camera is tilting completely vertical to the left. Any ideas? I am missing something somewhere.

 

[WTFDproject]

" This one is tilting pretty hard to the left when turned off"
This is typically caused by one of these problems.
1.) Bent shaft on the roll motor.
2.) Camera out of balance. This can be caused by bent arms or by installing the wrong size counterweight on the left arm (the "bearing housing" is actually a counterweight, and they come in different weights).
3.) Wire tension. If the wires are twisted in the vertical arm or at the joint between the vertical and horizontal arms, it is likely the joint has been separated during some previous maintenance, and got rolled during installation.
4.) Short wires in the pitch arm. Rarely, the problem can be caused if someone installed a roll motor or roll encoder on the pitch arm. Same basic components, but the wire is shorter. The same idea applies if the pitch wires are caught on something inside the arms, or if the pitch arm wires have been repaired by cutting out a damaged section.

You can do a preliminary check by holding rear arm horizontal with the camera facing down (i.e., tilt the whole unit forward). If the camera arms remains centered, the problem is likely a bent motor shaft, bent arms, or the wrong counterweight. If the camera still rolls to the left, it is likely a wire tension or wire length problem.


"the camera is tilting completely vertical to the left"
This may be an additional problem instead of the cause of the left roll. If the camera is rolling completely vertical, it means the arms are down so far the camera hubs are not hitting against the gimbal cover. The "arms down" is caused by a bent vertical arm or a bent roll motor shaft, and is often both.
You can do a preliminary check for this one by using the gimbal lock as a gauge. You can always force it in. But if the camera ball is in the right position, the camera hubs will line up perfectly in the center of the gimbal lock openings.