• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 62 subscribers
  • Views 414 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.

LM5175: OCP for buck / boost / buck-boost

Austin Do
Expert 3275 points
Part Number: LM5175

Hello team,

My customer has tested OCP for their system.

- Vin : 13~19V

- Vout : 12V

- 4SW Buck-Boost Converter

As you can see below table, 12.8A is not expected value if Vin =13V and Rsense = 2m Ohms.

Rsense

Vin = 13V

Vin = 16V

Vin = 19V

2m

12.8A

23.7A

28.6A

2.5m

20A

21.7A

24.7A

2.73m

19A

20.5A

23.5A

Is it related to buck-boost operation? According to the datasheet, there is no equation about OCP for buck-boost mode.

Or is there any limitation about Rsense?

Best Regards,

Austin

  • 0
    TI__Guru 65280 points

    Hello Austin,

    There is nothing obvious that explains this behavior as far as I can see. Is it possible that the input voltage on the converter itself is no longer at the level it should be?

    If this is not the case, measuring the input voltage, output voltage, sense voltage and if possible inductor current would be necessary for starting to debug.

    On the other hand, it would be helpful to know as well, at which input voltage the OCP gets too low.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Expert 3275 points

    Hi Brigitte,

    Thanks for reply.

    In case of Vin = 13V, could it possible be affected by load transient which can affect Vin?

    The customer is wondering why OCP gets too low even though OCP gets normally in case of Rsense = 2.5m, 2.73m.

    Can you explain how to get OCP in buck-boost operation? Let me check how the customer measured this result also.

    Best Regards,

    Austin

  • 0 in reply to Austin Do
    TI__Expert 3275 points

    Hi Brigitte,

    More updates from the customer.

    The customer has a test by increasing electronic load. There was no load step.

    As you can see below table. Every 3 board shows same result.

    Regarding Vsense, that voltage was too small and noisy to get a meaningful value. 

    Rsense

    S/N

    Vin = 12.5V

    Vin = 13V

    Vin = 14.5V

    Vin = 16V

    Vin = 19V

    2m
    (4m x2)

    #1

    -

    12.8A

    19.8A

    23.7A

    28.6A

    #2

    -

    12.6A

    22.8A

    22.9A

    28.4A

    #3

    -

    12.2A

    19.2A

    22.7A

    27.7A

    2.5m
    (5m x2)

    #1

    18.7A

    20A

    22.4A

    21.7A

    24.7A

    #2

    -

    19.5A

    22.3A

    21.5A

    25.5A

    #3

    -

    17.1A

    21.2A

    20.3A

    23.5A

    2.73m
    (5m x1, 6m x1)

    #1

    17.9A

    19A

    21.3A

    20.5A

    23.5A

    #2

    -

    18.5A

    21.4A

    20.6A

    24.5A

    #3

    -

    16.2A

    20.3A

    19.5A

    22.3A

    Best Regards,

    Austin

  • 0 in reply to Austin Do
    TI__Guru 65280 points

    Hello Austin,

    Please ask the customer to measure VIN, VOUT and if possible the inductor current with an oscilloscope.

    Then please check the current limit with 2m at more different votlages to see where the issue starts. At 14.5V it seems to be still ok. What is with 14V? 13.5V?

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Expert 3275 points

    Hi Brigitte,

    Please find the attached file for measured data. 

    Can you review the waveform? 

    LM5175_ocp_230331.pptx

    Best Regards,

    Austin

  • 0 in reply to Austin Do
    TI__Guru 65280 points

    Hello Austin,

    Thank you for the measurements. The device is going into hiccup operation, so it seems to be in current limit. Could you please share the schematic?

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    TI__Expert 3275 points

    Hi Brigitte,

    Sorry for late response. This thread is public forum so that I send the schematic file to your personal email.

    Best Regards,

    Austin

  • 0 in reply to Austin Do
    TI__Guru 65280 points

    Hello Austin,

    The customer seem to use just one ground connection on the whole board. It seems everything is connected to AGND.

    The current through the sense resistor is highest with lowest resistance and as well the noise is highest, so it is possible that the noise filter on CS/CSG is no longer effective against the current spikes at turn on of the MOSFET when reducing the sense resistor to 2mohm.

    The customer can check if a bigger RC filter on CS/CSG can help, e.g. increasing C31 to 150pF or R1 and R2 to 150 ohm.

    Nevertheless, long term a change in layout might be the right way to go.

    Best regards,
    Brigitte