CGO3+/Typhoon H interface pinout

[WesBur13]

Does anyone know how the typhoons 6 connection interface works for the camera? If not I might try hooking it to an oscilloscope.

 

[Murray Martz]

I know that when it comes in contact with the matching 6 pins.....it works. No clue here, but would also like to know as I have a few different configs I'd like to try.

 

[Tuna]

I'd certainly be interested. I believe there are two pins for power and one pwm for gimbal rotation, documented on a couple of threads on here. My assumption is that at least two more are data between the CGO3+ and Flight controller (as that's how the upgrade is performed, and channel hopping works), but have no technical specs for that.

 

[mcbutler]

You can identify the power pins using a good quality voltmeter, remove the camera, start the H, check each terminal. You could then by a process of elimination diagnose the purpose of the other connections.

 

[Hunter_Michael]

Here is a pic of the board that may provide some answers for you.

 

[porky]

View attachment 5705
Here is a pic of the board that may provide some answers for you.

Can you give few more pics

 

[dylanfm]

Can you give few more pics

I tried to figure out the protocol for the CGO3+ interface for a little while but didnt have much luck. If I remember right it seemed like on the PWM pin there was usually two ~1-3ms pulses that get repeated every 20ms. I kind of of thought that 1 pulse was for the tilt and 1 was for the pan, but I didn't have much luck replicating that to control the camera. But then noticed that during power up (and maybe when changing camera modes?) there is serial data that is sent to the PWM pin also. I know that the serial data that is sent from the TH is different than the serial data sent from hand held gimbal mount (which I think was actually just sent a series of repeated pulses). I wasn't really able to decode the serial data.

 

[Pöllö]

The CGO3+ is a tricky one as its protocol seems to depend what it is connected to. The tilt can be driven with a PWM with an initialization pulse when the camera is used without the Typhoon, but when the camera is connected to the Typhoon, there is two-way serial communication between the camera and the drone. The PWM pin is not used for PWM in this case. I tried to decode that serial communication too but I was unable to get anything out of it with my logic analyzer's decoders with any baudrate, including the few yuneec-special ones (like 250 000).

 

[dylanfm]

The CGO3+ is a tricky one as its protocol seems to depend what it is connected to. The tilt can be driven with a PWM with an initialization pulse when the camera is used without the Typhoon, but when the camera is connected to the Typhoon, there is two-way serial communication between the camera and the drone. The PWM pin is not used for PWM in this case. I tried to decode that serial communication too but I was unable to get anything out of it with my logic analyzer's decoders with any baudrate, including the few yuneec-special ones (like 250 000).

As i recall I captured the the serial data coming out of the PWM pin while the TH was powering up without the camera attached by using the same serial settings as the TH's receiver (115.2Kbps, 8N1, LSB first). Attached is a log file. It looks like it has data and it didn't have framing errors so I think I got the right settings. It seems to be mostly repeating data. It was probably waiting for some reply from the camera. I should try it again with the camera attached. I would really like to figure it out.

EDIT: Now that I think about it, I think this serial capture might have been from my Typhoon G. I don't actually remember.

 

[dylanfm]

The CGO3+ is a tricky one as its protocol seems to depend what it is connected to. The tilt can be driven with a PWM with an initialization pulse when the camera is used without the Typhoon, but when the camera is connected to the Typhoon, there is two-way serial communication between the camera and the drone. The PWM pin is not used for PWM in this case. I tried to decode that serial communication too but I was unable to get anything out of it with my logic analyzer's decoders with any baudrate, including the few yuneec-special ones (like 250 000).

So I captured more of the serial data between the TH and the CGO3. I captured both the data to the CGO3 (the PWM/rx pin on the camera connector) and from the CGO3 to the TH (on the TX pin) during power up and also with the ST16 switches, knob and slider in different positions. The TH doesn't send serial data continuously, so by looking for jumps in time between data it looks like each packet starts with 0xFE. At first glance I believe that byte 2 of each packet indicates (EDIT is related to the size of the packet (probably the data payload size). It seems to correspond to the total size of the packet-10 bytes. So right now I'm assuming the data starts at byte 11. It is also possible that the byte 2 value doesn't count some checksum or values at the end of the packet which could mean that the data starts at byte 9 or 10. I believe that byte 3 is a count of the number of packets sent since it usually increases by one, but there are some weird jumps some times.(/EDIT)

My serial captures are here if you can make any sense out of them:
TyphoonH_CGO3_Serial_capture - Google Drive

 

[Danyboy]

I know this is an old thread...

...but...

...I am looking into converting a Q500 4K with a damaged mainboard into a betaflight controlled unit.
Would use the Diatone Mamba Mk4 H743 for that. A PDB can easily be fabricated. Also a 3d-printed adapter to mount the FC in place of the original mainboard.
But before starting something that might not work, I would like to clarify some points...

I read above that the CGO03 needs to have certain pulses to initialise PWM-control for up and down?
Did anybody go further into that?
PWM like a standard servo?
Is there a way to get the video-signal feed directly from the camera, so I can send it over a regular 5,8Ghz transmitter?

Also I wonder: Are the Yuneec-ESC controlled like a standard-servo? Or is there something more sophisticated in place, like an SPI- or I2C-databus?

If all is based on standard Servo PWM, it would be super-easy to control with nowadays flightcontrollers...
...given the state of technology back in the days between 2010 and 2015, it is likely to be based on servo-signals...
...and I have to admit: having a 13-inch cruiser beside my "normal" 2 to 5 inchers is tempting...

Looking forward to your thoughts!

 

[WTFDproject]

What I read into the description of what you want to do suggests you may not be aware that the gimbal and the camera are two different control systems. The video related camera stuff, including the settings, camera select. camera mode, snapshot, video, etc. are all on 5.8 WIFI.
It is the gimbal part that has the PWM issue. It will stabilize without any external signal. But the manual tilt control comes through the 2.4 drone control WiFi and includes an initialization code. I believe some folks managed to deduce the code and even emulate it with some extra equipment, but I am not sure if that was for the CGo3 or a different Yuneec camera of the same era. @h-elsner may know how to find the website that included the discussion.

 

[Vaklin]

A steady grip costs around 20 dollars and it is a good starting point. Do the rest first, moving the camera is the simplest of all.

No way to use this camera as an analog transmitter but every tablet with 5.8G is capable to show the camera downlink.

 

[Danyboy]

What I read into the description of what you want to do suggests you may not be aware that the gimbal and the camera are two different control systems. The video related camera stuff, including the settings, camera select. camera mode, snapshot, video, etc. are all on 5.8 WIFI.
It is the gimbal part that has the PWM issue. It will stabilize without any external signal. But the manual tilt control comes through the 2.4 drone control WiFi and includes an initialization code. I believe some folks managed to deduce the code and even emulate it with some extra equipment, but I am not sure if that was for the CGo3 or a different Yuneec camera of the same era. @h-elsner may know how to find the website that included the discussion.

...well that was informative...!

Yes, indeed I looked at camera and gimbal as one single unit, not two individuals.
That makes it easy to swap the camera for something from open-source and current technology. What I would call a "off the shelf" solution instead of hacking to tap the video-signal before it goes into the wifi-part of the unit.

I will definitely look into that gimbal control some more! While I have that function, I would definitely like to use it...
It should be easy to feed data to the gimbal with an tiny arduino, which in turn gets fed by the flightcontrol. Instead of just passing the servo-PWM through, the arduino would sit idle to allow for the boot-sequence, then once initialises the gimbal for tilt-control and from then on send the servo-PWM.
Definitely doable and a small exercise - once you did the reverse-engineering to know what to send to the gimbal...

Do you have knowledge about the ESC-controls? Are these standard-servo PWM-signals? Or is there a databus, communicating between ESC and mainboard?

It's a shame Yuneec did not drop any data or specifications on the system and interfaces. The product could live quite long if the firmware was open-sourced and components could be exchanged by new and modern adapted components...

PS: I just replaced the IMU of the specimen in question. Got a new SMD-chip for about 8 bucks from mouser. Bought two to be on the safe side, had to pay 20 bucks for shipping... Diagnostics show everything as ok, test flight will have to wait some days for the weather to improve...

 

[h-elsner]

Source is a tutorial from a user named "brused" in a german forum (which is unfortunately closed for guests):
Anmelden

Also see here:
Using CGO2 with standard PWM receiver - RC Groups

In attach I have translated this tutorial.


Regarding the ESC I don't know what it is but I doubt that ESC control is plain PWM. I never tested it. Honestly, I am afraid to open the felt 100 screws and I have unfortunately no spare board to experiment...

br HE