• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 850 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM5146-Q1: How to achieve DEMB function with external driver

Wei-Hao Chen (TW-FAE)
Expert 3260 points
Part Number: LM5146-Q1
Other Parts Discussed in Thread: LM5146, , UCC27712, LM5143-Q1, LM5145, LM5117-Q1

Hi team,

My customer has an application question regarding DEMB function of the LM5146-Q1 when external driver is used. My customer wishes to use external driver with LM5146 due to the SW & BST rating is not enough for their application.

In our previous experience, when external driver is used the SW pin tends to be grounded, however the DEMB function will not work because SW detects the zero-crossing when hitting the zero current. But now it is grounded which cannot detect the zero crossing anymore

Would you please suggest a circuit to achieve DEMB function while external driver is used? They tends to use UCC27712 for the external driver.

Thank you!

Best Regards,

Wei-Hao

  • 0
    TI__Guru 66522 points

    Hi Wei-Hao,

    As you know, the LM5146-Q1 has integrated gate drivers. It is not designed to operate with external gate drivers. If SW is grounded, the current information for zero-cross detection is lost - this is required for pre-bias startup, diode emulation mode operation, as well as adaptive deadtime optimization for MOSFET switching.

    Maybe we can recommend another part. What are the input and output specs for this application?

    Regards,
    Tim

  • 0 in reply to Timothy Hegarty
    TI__Expert 3260 points

    Hi Tim,

    No long no see, hope all is well. I will convince my customer not to use the external driver. They have some other questions wants to clarify:

    1. The VILIM-TH spec has very large range, is this the offset of the ILIM comparator or the ILIM threshold?  if yes, the OCP range will varies a lot. Could you please check the OCP accuracy? 

    2. How to select the proper value for Rs and RLIM to set the OCP. The equation 6 provide how to calculate RLIM, but it did not include how to select the appropriate Rshunt. May I know how should customer design the Rshunt first?

    3. Figure shown on page 16 (function block diagram) and Figure 34 on page has swapped input pins of the internal ILIM comparator, can you please check which is correct?

    Thanks a lot.

    Best Regards,

    Wei-Hao

  • 0 in reply to Wei-Hao Chen (TW-FAE)
    TI__Expert 3260 points

    Hi Tim,

    Any update on this?

    Thanks a lot.

    Best Regards,

    Wei-Hao

  • 0 in reply to Wei-Hao Chen (TW-FAE)
    TI__Guru 66522 points
    Hi Wei-Hao,

    1. That spec is the current limit comparator threshold voltage variaition. Nominally it's -2mV with some variation around that based on process and temperature. Note that these limits are guardbanded so the variation may not be as wide in practice.

    2. A current sense signal between 50mV and 100mV at current limit inception seems appropriate. Such a current sense voltage amplitude balances shunt power loss with achievable signal-to-noise ratio. This allows calculation of the shunt resistance based on the desired current limit setpoint.

    3. The output of the ILIM comparator should go high in a current limit condition, so the schematic on the left is technically correct.

    Regards,
    Tim
  • 0 in reply to Timothy Hegarty
    TI__Expert 3260 points
    Hi Tim,

    One more question, how to interleave for multiple LM5146 devices? How to connect the FB, SS and COMP pin? Do we have any reference circuit that our customer can follow?

    Thanks a lot.
    Best Regards,

    Wei-Hao
  • 0 in reply to Wei-Hao Chen (TW-FAE)
    TI__Guru 66522 points
    Hi Wei-Hao,

    There is a blog written by David Baba that describes paralleling two LM5145 controllers. However, I recommend looking at the LM5143-Q1 as that is a true multi-phase controller.

    Regards,
    Tim
  • 0 in reply to Timothy Hegarty
    TI__Expert 3260 points

    Hi Tim,

    Thanks for your suggestion, but the voltage rating for LM5143-Q1 is enough for my customer's application.

    Also they have two follow-up questions regarding LM5146-Q1.

    1. The design example (9.2.2 Design 2 – High Density, 12-V, 8-A Rail From 48-V Automotive Battery Power & ) that is in the datasheet has relative large Rlim setting than expected. Based on my customer's calculation the RLIM should be around 300~400ohms, but the value shown in the design is 619ohm. Could you please check how to derive this 619ohm resistor? 

    Here's how my customer calculate the expected RLIM value, is this correct?'

    2. How to achieve CC mode on LM5146? As we know pulling the COMP pin might have some problems. Would you suggest a reference circuit for CC/CV with LM5146?

    Thanks lot Tim, really appreciate you support.

    Best Regards,

    Wei-Hao

  • 0 in reply to Wei-Hao Chen (TW-FAE)
    TI__Guru 66522 points
    Hi Wei-Hao,

    Take a look at these blogs for CC/CV operation:
    e2e.ti.com/.../how-to-design-a-simple-constant-current-constant-voltage-buck-converter
    e2e.ti.com/.../how-to-design-a-simple-constant-current-constant-power-constant-voltage-regulating-buck-converter

    Note that the LM5117-Q1 has a CM output that is useful for CC operation. However, it seems that you need more than 65V input rating.

    Please use the LM5146-Q1 quickstart calculator available by download from the product folder for guidance with component selection. The current limit is based on the low-side MOSFET Rds(on) at Vgs = 7.5V (use a value between the specified 4.5V and 10V ratings in the MOSFET datasheet or use the Rds(on) vs. Vgs curve if provided).

    Regards,
    Tim