No PH waveform of TPS54610 @Vin=5.6V Vout=3.8V fsw=500kHz in no load

Can you post some waveforms that show your concern?

Hi John,

           I checked it again. Below is a normal power-up operation. CH1 is PH, CH2 is Vin.

What frequently happens is as below shows. Vin=5.6V Vout=3.8V in no load fsw=500kHz.There is no switching at PH, and PH equal Vout=3.8V.

Sometimes, it is as below.

Each below measure can total remove the no switching phenomenon.

1. add some load (50mA);

2. reduce the fsw to 350kHz;(no load)

3. decrease the Vout;(Vout=2.35V/1.2V, no load)

It seems there is a UVLO for BOOT-PH.

Is there a pre-bias voltage on Vout?

No, I thought it might be, but I verified the issue happened without prebias as well.

How many circuits do that?  Just one?  You may have a defective IC.

I checked it on two boards in no load. They are the same exactly. The issue on both circuits can be removed by the measures I mentioned before.

I also found it might relate to the soft start of input DC source. If I turn on Vin by a switch from the 5.6Vin directly, no issue anymore. 

I strongly guess there is an UVLO for BOOT-PH.

I don't think there is a BOOT UVLO on TPS54610.  That is usually only required for converters that support extended on time past 100% duty cycle for low headroom operation.  I have not seen that type of behavior before, although it has been several years since I last worked with TPS54610.  You may want to check all the peripheral signals during start up such as VBIAS, SS/ENA, BOOT and COMP to see if any are inhibiting switching.