Part Number: TXB0104
Other Parts Discussed in Thread: TXB0604, TXB0606
We are currently encountering technical issues with the TXB0104 being used as a level shifter in an SPI topology. We would like to request your technical analysis and guidance on the potential root causes and recommended solutions.
The TXB0104 is implemented in an SPI interface to provide level shifting. During operation, we observed significant waveform oscillation (please refer to the attached waveform images).
To isolate the issue, we performed a troubleshooting step by disconnecting the load on the B-side (right side) and leaving it floating. In this configuration, only the BMC (on the A-side) remains connected to the TXB0104. During the abnormal rising state of the TXB0104, we measured the supply currents and observed that the MOSI signal continues to exhibit abnormal voltage rises. Specifically, the measured voltage levels are 3.021V on the B-side and 1.79V on the A-side. We suspect the abnormal voltage on the B-side is caused by an internal pull-down voltage divider within the BMC. Our measurements show that during this abnormal rising state, there is no significant current injection into VCCA; however, a current of 65µA was detected on VCCB, which correlates with the abnormal voltage rise. (Note: At this stage of testing, only the MOSI channel of the TXB0104 remains active, while all other channels have been disconnected via series resistors.
Can I ask what might be causing this phenomenon, and whether it’s related to TXB’s own chip characteristics?
Thank you for your time and professional support. We look forward to your insights.
Further we changed it into TXB0604. The waveform seems good But there are still abnormal sharp falling edges.
After replacing the TXB0604 for verification:
1. Measuring the waveforms at the front and back of the LevelShifter shows no oscillations;
2. The TXB0604 issue of locking the MOSI level high on the falling edge still exists.
