• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 297 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.

LM5046: Current sense

Téo Robert
Prodigy 100 points
Part Number: LM5046

Hi,

I am designing a PSFB converter for a POL application (100W), and prototyping with the LM5046 controller. I have a few questions about current sensing to do PCM control with the part :

  1. I am concerned about the inductance the current sense transformer adds in the switching loop. Is there particular requirements to use it (regarding decoupling capacitors for example) ?
  2. The other way to proceed is using a resisor in series with the input. But the resistance required to have the 710 mV drop recommended on the datasheet with my 10 A max peak input (70 mOhms) causes huge power losses (350 mW at my max RMS of 5 A, accounting for a 2 % impact on efficiency). Can I use a lower resistor value, as I can provide an external overcurrent protection and will the PCM control still work (is there a mimimum voltage drop on the Rcs to make it work) ?

Thanks in advance.

Regards,
Téo

  • 0
    TI__Expert 7900 points

    Hello Teo,

    1) TI has a reference design for the LM5046, here.  It delivers 48V/10A and uses a current sense transformer without issue.  You can review the specifications for the current sense transformer in this design here.

    2) If the current sense resistor is very small it increases the gain of the current control loop.  So you have to account for this when setting compensation values.  If you implement an external overcurrent protection can you guarantee it will provide all the built-in protections of the LM5046, such as pulse-by-pulse current limit and hiccup mode?

    I think the best way to proceed would be a current sense transformer because it takes advantage of all the built-in functionality of the LM5046.

    Regards,

    Eric

  • 0 in reply to Eric Reicher
    Prodigy 100 points

    Hi,

    Thanks for your answer. We will use the crurent sense transformer. I have an additionnal question : is it possible to sense the primary transformer current instead of the input current ? It also contains the current slopes needed for the PCM and this could be more convenient as it preserves the input switching loop decoupling, and as more inductance in series with the transformer is not an issue. I have never seen it in PSFB applications …

    Thanks again,

    Teo

  • 0 in reply to Téo Robert
    TI__Expert 7900 points

    Hi Teo,

    There are designs that show current sensing at the source of the low-side MOSFETs.  See below.  This would preserve input switching loop decoupling.

    Regards,

    Eric

  • 0 in reply to Eric Reicher
    Prodigy 100 points

    Hi,

    We excluded using a sense resistor as it causes too much losses. Will the LM5046 current mode work with the current sense transformer in series with the main transformer (primary side) ?

    Thanks.

  • 0 in reply to Téo Robert
    TI__Expert 7900 points

    Yes, I remember ruling out the sense resistor due to power loss.  What I meant was to use the current sense transformer from the source of the lower FETs to GND.

    You CAN place the current sense transformer in series with the main transformer.  This location tends to be the noisiest location to sense current because it sees both HS and LS FET switching.

    Regards,

    Eric