This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM2733 true shutdown question

Genius 3870 points

Hi there, 

We are trying to implement true shutdown by adding PMOS on Vout referring to TI docs that written by @ , and I have some questions.

1.If PMOS is inserted before feedback resistors as following sch’s second circuit, when the IC is just enabled(at very beginning time), the PMOS breaks the feedback loop, and the PMOS will keep OFF until its source pin voltage is higher that Vin. So I am not sure whether the startup process will be affected,because the PMOS is involved in regulation control loop.  I prefer the PMOS be added after feedback resistors as SCH in PPT "Boost Converter Design Tips".

2. Is there any restriction to feedback resistor value selection? I noticed that in datasheet example(section 8.2.26), there is a "92uA current through divider resistors" , how should we interpret this "92uA"? If we use larger resistors, as following SCH, to boost 12V to 24V, will these resistor values be ok? Normal load current is small, about 20mA.

Thanks.

  • With regard to your first question; You are correct that there are some turn-on dynamics that should be evaluated to make sure there is no risk of overshoot. By connecting the divider in front of the PMOS this risk is eliminated but now the divider is always present as a static load on Vin when the switcher is disabled. This would only be a concern if Vin is a battery source. Also, when selecting the PMOS make sure there is no situation where VGSmax rating is exceeded. This could occur at Vout = 24V and if Vin was momentarily shorted to zero volts. This is not a concern with the PNP approach of the top diagram. Also a feature in the PNP approach is that the VCEsat drop is accounted for by having the feedback divider following the transistor.

    In reference to divider current; From page 4 of the data sheet, the input bias current of the feedback input in 60nA typical. If you run a divider string current of 50 times higher than this it will make that negligible. This would lead to a recommended divider current greater than 3uA. You have a divider current of 4.1uA; so that looks fine. I also notice that your design has two series connected upper feedback resistors; one of them zero ohms. If this is just for trimming the output voltage slightly then the configuration is okay. But if this is as an injection point for loop response measurement (Bode plot with FRA) then the feed forward cap should then be attached to the midpoint of the two upper resistors. Alan