LP5036EVM: LP50 unable to talk to board ("Couldn't write register")

Hi Stephan, 

The hardware setup is correct and verified by the green connection status. For the I2C slave address, the jumper in the picture corresponds to default 0x30. Would you kindly examine the external power supply, swap the USB2ANY 3.3V jumper, and restart the GUI for testing? 

Best Regards,

Aaron

Thanks Aaron for the quick response!
Multimeter says 5.15V:

Can you clarify what you mean with "swap the USB2ANY 3.3V jumper"?
I tested the ribbon cable and it seems fine.

I'm running the software in a VirtualBox on a Linux host but I don't think that can be the problem as the connection status shows green.

Hi Dear E2E users. Since April 3rd to 5th is Chinese "Qingming" Holiday, our expert will reply your question on April 6th.
Thanks for your understand.

Hi Stephan, 

It is the jumper in the picture below which connects to USB2ANY 3.3V for power supply - instead of the 5.15V provided - just for testing purposes. I am sure the ribbon cable and software compatible with Linux host are fine; would you capture the waveforms on SCL/SDA lines and share with us?

Best Regards,

Aaron

Thanks Aaron,

unfortunately I don't have the tools to analyze the signal.

With the jumper closing the first two pins counting from the left I made the following observation:

I can see "Written register..." messages and setting a single LED color works. The "Single Color chasing effect" also works. When I add more than 4 LEDs or use an effect that includes more LEDs, writing the registers fails. At that point the board stops responding. I suppose that this means that the power draw is too much for the USB.
So it seems supplying external power is not working. The power supply I'm using is a simple 30W universal AC/DC adapter: PCHero 30W switching power supply UC05E. Maybe that one doesn't handle peaks well? Can you recommend a power supply?