• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 327 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.

LM5175: No boot-strap gate drive to right side top FET during Boost mode!

Ram Chandru
Prodigy 170 points
Part Number: LM5175
Other Parts Discussed in Thread: LM5176

Using LM5175 for buck-boost converter. Switching frequency ~250KHz. 

Only notice recently that when running boost mode: Input 12V, output 15V @ 20A, the top right FET (supposed to be syn FET) has no boot-strap gate drive: ie. the voltage of gate drive (pin19) is the same as the source voltage (pin18).  So only the body diode conducts, FET does NOT turn-on and efficiency dropped.  Checked boot-strap voltage on pin 20 and is normal, but no boot-strap at gate drive on pin19.  Checked there is NO short-circuit gate-source on the FET in question.  

When operating in buck-mode, boot-strap gate drives are normal on the relevant FETs.

Any experience/suggestion on this and can anyone help?

  • 0
    TI__Guru 68495 points

    Hi Ram,

    Thank you for using the Lm5175. For debugging purpose, can you change the MODE pin resistor to 93.5k (the resistor to GND) or simply pull up MODE pin to VCC, and see if the top right FET is switching when your VIN=12V?

    Also, would like to review your schematic.

    Thanks,

    Youhao Xi, Applications Engineering

  • 0 in reply to Youhao Xi
    Prodigy 170 points

    Hi Youhao, thank you for your advice. 

    Our MODE pin was connected to 0V directly, ie. DCM with Hiccup disable.  We tried your suggestion and connect this pin to Vcc, ie. CCM with Hiccup disable. The boot-strap gate drive is now present at the right-side top FET and the converter is running as expected efficiency.  Am I correct to say there is a bug in the internal logic of the chip that seems to disable the boot-strap drive to the right-top FET during BOOST operation, if the MODE pin is strap to 0V?

    Ram

  • 0 in reply to Ram Chandru
    TI__Guru 68495 points

    Hi Ram,

    Thank you for doing the test that I asked for.  No, it is not because of a bug in the internal logic of the chip.  It means your PCB layout may have distorted the signal for zero current crossing detection circuit, and the IC turns off the sync FET.  

    As the first step, we would like to review the schematic to see if there is something for quick fix.

    Thanks,

    Youhao

  • 0 in reply to Youhao Xi
    Prodigy 170 points

    Hi Youhao,

    Thanks for the reply.  Our circuit is almost identical to tidru46.  With the exact PCB of our circuit, we substitute the LM5175 with LM5176. Rt adjusted for the same frequency.  Rmode 200K to 0V (hiccup disable), we do NOT have the same problem of absent boot-strap.  LM5176 operates CCM only I believe.

    Am I correct to understand the 'Zero-crossing' is detected by CS and CSG pins?  In our circuit we have 1.5mR sense resistor (chip type), and 100R+100p+100R filter to the CS/CSG pins with dedicated tracks (short tracks) to the sense resistor.  The RC filters situated immediate at the pins of the IC above signal ground plane (4 layer PCB).

    In any case, CCM is acceptable to us and changing from DCM to CCM for LM5175 is of no issue at all.  I am trying to understand the 'why'.  Thanks for your time.

    Ram

  • 0 in reply to Ram Chandru
    TI__Guru 68495 points

    Hi Ram,

    There is zero cross detection circuit monitors the CS signal.  If your sense resistor has inductance, it can distort the signal and cause the IC to "think" the current is negative, hence it turns off the SYNC FET gate.  The CS signal RC can compensate the indutance effect.  Ideal, if the RC and L/R time constant matches, the effects are eliminated, making a pure resistor sensing results. 

    For CCM, the zero crossing circuit signal is ignored.

    Thanks,

    Youhao

  • 0 in reply to Youhao Xi
    Prodigy 170 points

    Thank you Youhao.

  • 0 in reply to Ram Chandru
    Prodigy 170 points

    By the way Youhao, we are using the ULR series from TT Electronics for the current sense resistor (ULR2 1.5mohm).  The datasheet mentioned 'low TCR and inductance'.  Would you have experience of better current sense resistors (SM type and lower inductance) and can you suggest something better........... for our future reference please?

  • 0 in reply to Ram Chandru
    TI__Guru 68495 points

    Hi Ram,

    We usually do not comment on non-TI product here.  One reference point is our LM5176 EVM circuit which uses KRL6432E-M-R004-F-T1 from Susumu, and that series should work fine.  Note that even a very small inductance may cause a big issue because the low resistance value:  the time constant is L/R.

    Thanks,

    Youhao