I have two converter in a new design using the TPS54233. One is 5V out and the other is 3.3Vout. The initial designs were made by the WebBench SW. The 5V starts reliable and operates without problems. The 3.3V design starts unreliably. When it starts the 3.3 output is good. It runs fine until power cycled. Vin for the converter is 24VDC. The maximum output current designed for initially was 1.5A. This is actually 5 times my real maximum continuous current of 300mA. I though that I might be bumping up against then minimum duty cycle of the converter so I had WebBench design for 700mA max. I made the changes involved (increased inductor from 22uH to 47uH, increased soft start cap from 6.8nF to 8.2nF and changed the compensation circuit). Still starts intermittently. It seems to start the after being powered down a while better than if the power is cycled off and back on again. Once if doesn't start its done until I leave power of for while.
I thought that the restart behavior might be due to input power when cycled so I tried lengthening the SS time by changing to a 20nF soft Start cap. Still does not start reliably. When it does not start the output voltage sits at 0.1 to 0.2V.
I have also changed the TPA54233 IC, changed the catch diode, tried decreasing the output capacitor from 200uF, to 100uF and then to 47uF (the WebBench value) in case the problems was capacitor in rush current at start. still have problem.
My compensation circuit should be 12nF in series with 9.52kOhms, all in parallel with 270pF. I don't have those exact values on hand so I am using 10.4nF, 10K, and 200pF. Note that for my initial 1.5A circuit I purchased the exact values given by WebBench.
