Other Parts Discussed in Thread: TPS543C20
I am currently troubleshooting an issue with the TPS543B20. I have another buck converter which provides a clock signal to other buck converters on a design in order to keep all converters at the same synchronized switching frequency. This clock signal is provided to the TPS543B20 through a connection on its sync pin. The TPS543B20 has been configured to operate in standalone mode according to the instructions in the datasheet. The RT resistor is set to keep the internal oscillator frequency below the synchronization clock frequency. The SYNC pin is directly connected to the aforementioned clock output pin of another buck converter. The MODE pin is left floating (but we have included a DNP resistor option to GND in case we want to change the operating mode). The RAMP pin is shorted to GND through a 0 Ω resistor.
When the converters are enabled (the TPS543B20 enables well after the buck converter providing the clock output signal is in regulation), instead of the TPS543B20 synchronizing to the higher frequency clock signal from the other buck, the SYNC pin appears to override this input clock and provides a clock output signal on the synchronization net at the frequency set by the RT resistor of the TPS543B20. This causes the buck converters to synchronize to the frequency of the TPS543B20 instead of the intended clock signal, and the board does not operate in a synchronized manner.
Is there something wrong with the configuration as I have described it? My understanding is that this configuration should allow the TPS543B20 to synchronize to the clock signal applied at its SYNC pin, and that it should not be driving any clock signal from its SYNC pin. My current plan is to make use of the DNP resistor option on the MODE pin and change the TPS543B20 to a master sync in mode by installing a 34.8 kΩ resistor, which should ensure the device will not provide a clock signal on its SYNC pin.
Thanks in advance,
-Paul Rankin
