Other Parts Discussed in Thread: LM5050-1
Hello,
I'm working on a power supply board which uses the TPS61178 to boost the voltage from ~5V to ~9V.
In some case, the input is ~12V and the TPS61178 is just skipped with an ideal diode (LM5050-1 + NMOS).
I have now two boards, but only board #1 behaves as expected: when the Vin 5V I have 9V as Vout, while when I have Vin 12V I measure ~11.4V at Vout, and the ideal diode skips the TPS61178 providing 12V at board output.
Board #2 behaves differently: when Vin is 12V, I have 13.1V at Vout, so it is higher than the input and the ideal diode is not used. If I look at the SW pin with a scope I see it switching at ~ 92% duty cycle, even if Vout is >> than 9V (at FB pin I measure ~1.7V, >> than 1.2V which is the reference).
From the Functional Block Diagram on the datasheet at page 10 I really can't understand this behavior: with 12V input and 1.7V at FB the output of Gm ampifier is low, and PWM comp ouput must be high, disabling the CTL/CRT block.
With board #1 I observe what I would expect: the 12V input flows through the body diode of the HS MOSFET and I measure ~11.4V at Vout, so ~0.6V drop over the body diode as expected. This allows the ideal diode to turn on and the whole TPS71168 block is skipped.
With board #2 I can't understand why SW switches, as FB is >> than 1.2V. I know that there could be issues when Vin ~= Vout, but this is not the case. If I swap the two chips from board #1 to board #2, #2 starts to work while #1 shows the behavior that #2 had, so it has to be an issue related to the particulare instance of TPS71168 (same production lot I guess - they were ordered together as samples from TI website).
Do you have any explanation for this?
