Arducam driver board on Navio2 via I2C

[MCMH2000]

Hello,
I have some issues with getting the I2C Bus working.
I want to use an Arducam PTZ Camera on my Drone (Redirecting...). The camera lens can zoom and focus with servos and therefore this camera comes with a driverboard which can control all functions of the camera. While the camera is a MIPI camera connected to the Raspberry Pi camera connector, the board ist connected via I2C to the Navio2.
My Plan is to control the camera lens with a software, but the problem is that as soon as i connect the board to the I2C bus of the navio, the barometer loses his connection.
Since I’ve been trying to figure out why, I have read that the barometer is the only sensor that is connected via I2C. But I know that it is possible to have many slaves on one I2C bus so it must be possible to get this working.

I have tried to use i2cdetect. Without the driver board connected this is the output:
pi@navio:~ $ i2cdetect -y 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: – – – – – – – – – – – – –
10: – – – – – – – – – – – – – – – –
20: – – – – – – – – – – – – – – – –
30: – – – – – – – – – – – – – – – –
40: – – – – – – – – – – – – – – – –
50: – – – – – – – – – – – – – – – –
60: – – – – – – – – – – – – – – – –
70: – – – – – – – 77

as soon as i plug in the driver board:
pi@navio:~ $ i2cdetect -y 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: – – – – – – – – – – 0d – –
10: – – – – – – – – – – – – – – – –
20: – – – – – – – – – – – – – – – –
30: – – – – – – – – – – – – – – – –
40: – – – – – – – – – – – – – – – –
50: – – – – – – – – – – – – – – – –
60: – – – – – – – – – – – – – – – –
70: – – – – – – – –

but sometimes the 77 is there for a little time:
pi@navio:~ $ i2cdetect -y 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: – – – – – – – – – – 0d – –
10: – – – – – – – – – – – – – – – –
20: – – – – – – – – – – – – – – – –
30: – – – – – – – – – – – – – – – –
40: – – – – – – – – – – – – – – – –
50: – – – – – – – – – – – – – – – –
60: – – – – – – – – – – – – – – – –
70: – – – – – – – 77

but just as quickly gone.
and sometimes there is a 6e:
pi@navio:~ $ i2cdetect -y 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: – – – – – – – – – – 0d – –
10: – – – – – – – – – – – – – – – –
20: – – – – – – – – – – – – – – – –
30: – – – – – – – – – – – – – – – –
40: – – – – – – – – – – – – – – – –
50: – – – – – – – – – – – – – – – –
60: – – – – – – – – – – – – – – 6e –
70: – – – – – – – –

but I don’t know exactly what this means.

I use the latest emlid-raspbian image 20191128 and arducopter configured to version 3.6.
The setup is shown here: Redirecting...
but i have connected the SDA, SCL, 5V, GND to the Navio2 I2C port and the servo VCC and GND to the Navio2 PWM header.

[mikhail.goryachev]

Hi @MCMH2000 ,

Welcome to our community forum!

We are looking into your issue. I’ll come back to you when we get any results.

[mikhail.goryachev]

Hey @MCMH2000, I’m back.

Could you, please, provide a photo of your setup?

Can you also try to launch a Barometer.py from Navio2’s examples and send a screenshot along with a screenshot of emlidtool tests?

[MCMH2000]

Sorry for the late reply.
These are the Results.

Barometer.py executed first with the I2C connection and then without.

The wiring of the camera.

An overview of the board connection.

An overview of the build.

The tests with the I2C connection.

And the tests without the I2C connection.

[mikhail.goryachev]

Hey @MCMH2000,

It is possible, that your platform control module causes interference that affects the barometer. We haven’t faced such behavior previously and can hardly reproduce it ourselves.

I might advise installing a ferrite bead around your control module wires leading to I2C port. This can suppress high-frequency noise and improve the situation.