Other Parts Discussed in Thread: SN65DSI84
Tool/software:
Hi,
I’ve designed a custom board featuring the LP8866 (LED driver) and SN65DSI84 (DSI-to-LVDS bridge), both controlled by a Raspberry Pi via I²C. For voltage level translation to 1.8V, I’ve included a TCA9406DCTR.
However, when I assert a HIGH signal on the enable (EN) pin of the LP8866, the boost converter appears to introduce noise/voltage spikes onto the I²C bus.
This disrupts communication with the SN65DSI84, resulting in repeated mistaken values/read errors when attempting to access its registers. Interestingly, most of times the readings from the LP8866 itself are complete without errors, it does not seems to be the result of a random interfering spike. Of course I am aware that the TCA9406DCTR might "amplify" these peaks but there are few mistakes in the communication that should be many more times there if the reason were noise. I've added an RC filter to the I²C lines with no improvement which confirms, apparently my last statement.
SDA (green) and SCLK (yellow) without peaks
SDA (green) and SCLK (yellow) with peaks
SCLK 3V3 (yellow) and SCLK 1V8 (blue)
Any suggestions?
Thanks in advance
Andrés