We use TPS40422 in our project, and monitor the output voltage through pmbus interface.A problem shown, 0.95V jumped to 0.83V for a moment. But at the same time, the oscilloscope observed nothing wrong.
Then we used TPS40422EVM-091 and TI Digital Fusion Designer to solve the problem. But it also appeared in the EVM board. The situations as follow:
1. Disabling the second phase and logging data, we found 1.2V jumping up to 1.7V or down to 1.0V instantly.
And the second phase voltage, currents and temperatures also jumped.
But there no UV or OV fault.
2. Through the log, it often appeared for successive two times.These instant fault might continued more than 50ms.
3.It might not a noise issue.
First there are no NACK in PMBUS LOG, standing that PMBUS interface is stable.
Then although our product's chip's VSNS2 and GSNS2 connecting to GND, VOUT2 also jumped up to 0.7V, standing for not the noise from GND causing.
We could avoid this problem through reading a certain times and getting the middle. But it takes risks that ignoring actual voltage problem.
And want to localize the source of the problem, design problem or chip problem?
Additionally, We found a function named "voltage transition" in the OPERATION register in the GUI. But it isn't a function of TPS40422. Is there newer revision of TPS40422?
