• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 274 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LM5145: Switch off high side FET

Silvano Ghilardi
Intellectual 330 points
Part Number: LM5145

Hi

I'm using the LM5145 for my buck converter. It looks like that at light load the switch off timing of the high side FET is too slow.

Please check the attached pictures.

CH1: Vds, CH2: Gate low side, CH4: Gate high side

As you can see the behavior is ok during the switch on of the high side FET. Low side gate goes low and after the dead time, the high side gate goes high. Everything ok.

BUT during the switch off of the high side FET, it looks like that the high side gate takes to much time and than the low side gate switches on. There is no dead time.

If I increase the load, it looks better.

Can you explain me that?

BR

Silvano

  • 0
    TI__Guru 73072 points

    Hi Silvano,

    This is likely expected behavior (however, your waveform was not attached).

    When the high-side FET turns off (check by measuring its Vgs), the switch-node voltage decreases slowly when operating at light load as it is the peak inductor current that discharges the SW node effective capacitance. Eventually, the low-side FET turns on and pulls the SW voltage to GND. If the SW voltage ramps all the way to GND before the low-side FET turns on, this is actually a lossless transition.

    As the load current increase, the peak inductor current rapidly discharges the SW node capacitance, and current flows in the low-side FET body diode before the low-side FET actually turns on.

    If you need further assistance, please forward the LM5145 schematic. You can use the LM5145 quickstart calculator to assist with component selection, etc.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Intellectual 330 points

    CH1: SW GND

    CH2: Ugate low side FET (to GND)

    CH4: Ugate high side FET (to GND)

    Hi Tim

    Thanks for the fast answer. I understand your point. I have 2 questions:

    1. The high side FET is turned off by the LM5145 and the voltage at SW decrease slowly. The low side FET turns on before SW reaches GND. Is this a problem? Can this be critical for the circuit? Please check the attached pictures. From my point of view it looks like that the Gate voltage to GND of the high side FET goes slowly low as well.

    2. The result of this "fast" discharge with the low side FET is a swinging below 0V. How can I filter that?

    Thanks for the help.

    BR

    Silvano 

  • 0 in reply to Silvano Ghilardi
    TI__Guru 73072 points

    Hi Silvano,

    See answers below.

    Regards,

    Tim

    Silvano Ghilardi said:

    Thanks for the fast answer. I understand your point. I have 2 questions:

    1. The high side FET is turned off by the LM5145 and the voltage at SW decrease slowly. The low side FET turns on before SW reaches GND. Is this a problem? 

    [TH] No, this is expected behavior. No issue.

    Can this be critical for the circuit? Please check the attached pictures. From my point of view it looks like that the Gate voltage to GND of the high side FET goes slowly low as well.

    [TH] The gate voltage to GND is determined by SW. What really matters is HO minus SW as this is the high-side gate drive.

    2. The result of this "fast" discharge with the low side FET is a swinging below 0V. How can I filter that?

    [TH] No need to filter this, it's just the result of parasitic loop inductance (that is minimized by keeping the input caps really close to the FETs). It will not cause an issue with the controller as it's short duration and relatively low amplitude.

  • 0 in reply to Timothy Hegarty
    Intellectual 330 points

    Hi Tim

    Thanks for the answer.

    1. question

    I have checked the Gate Source voltage of the high side FET and yes it is switched off before the low side FET switches on. So no issue.

    2. question

    This swinging is below -5V and this means out of maximum ratings of the LM5145. But I could solve it with a small gate resister.

    Anyway, to reduce over-/underswinging I will add gate resistors for each FET.

    Thanks for the feedback.

    BR

    Silvano