• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 4 answers
  • Subscribers 64 subscribers
  • Views 498 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.

LM5149: Power supply for active EMI

Federico Battaglin
Intellectual 545 points
Part Number: LM5149

Hello to All,
for a new project (Vin=48V (-15% and +10%), Vout=34V, Imax=15A, EN55022, class B), I'm trying to use the active EMI feature, but I don't understand very well (or I don't want accept the real reality), if I need to apply to VCCX the external 5V or can I connect this pin to another pin (Vcc: is it output?). But reading the data sheet, the Vccx-max is 6,5V... Considering the Vin(max) is near 60V, I need a regolator from 60V to 5V...

Please, someone, clarify me...

Federico

  • 0
    TI__Mastermind 22299 points

    Federico,

    VCCX is optional.

    VCC will always be powered from VIN based on the internal LDO.

    VCC can power the Active EMI filter.

    Hope this helps,

    -Orlando

  • 0 in reply to Orlando Murray
    Intellectual 545 points

    Dear Orlando,

    thank You for your reply!

    Yes, it's enough help, even if I don't understand in which case it's preferable to connect VccX to external 5V power supply. In the data sheet it's not so clear or I don't underdstand enough. I think it's preferable whne the Vin is over a threshold to minimize the intenral LDO power dissipation. Ok, but which is the current consumption of circuitry of active EMI filter?

    If I will use the internal 5V LDO voltage regulator, how I connect the VccX?

    Thank You and Best Regards
    Federico

  • 0 in reply to Federico Battaglin
    TI__Mastermind 22299 points

    If it's unused, VCCX pin should be tied to GND.

    In normal operation AEF is an AC-circuit, if AEF is doing alot of AC canceling then the max DC current from VCC should be ~50mA.

    However the short-circuit, DC current limit of AEF is ~100mA, this is if INJ is accidentally tied directly to GND.

    -Orlando

  • 0 in reply to Orlando Murray
    Intellectual 545 points

    Ok, about 50mA. With Vin=48V typ, the 5V internal regulator dissipation is about Pd=(Vin-5)/0,05=2.15W and I think it's too much high for this chip... In he data sheet, in many table Vin=8..18V and in the worst case the Pd=0,65W...

    Federico

  • +1 in reply to Federico Battaglin
    TI__Mastermind 22299 points

    Hi Federico,

    We have a 48VIN application in the datasheet, it is definitely possible to use AEF with LM5149 and at 48V VIN.

    50mA would be a worst case, I have measured ~15-25mA in typical AEF use cases.

    But yes, the efficiency hit would be more significant if not using VCCX.

    See LM5149-Q1EVM-400 EVM.

    Hope this helps,

    -Orlando

  • 0 in reply to Federico Battaglin
    Intellectual 545 points
    Orlando Murray said:

    Hi Federico,

    We have a 48VIN application in the datasheet, it is definitely possible to use AEF with LM5149 and at 48V VIN.

    50mA would be a worst case, I have measured ~15-25mA in typical AEF use cases.

    But yes, the efficiency hit would be more significant if not using VCCX.

    See LM5149-Q1EVM-400 EVM.

    Hope this helps,

    -Orlando

    Dear Orlando,
    sorry but I pressed "this solved my issue" instead "reply".

    1) "50mA would be a worst case": in which situation?

    2) which is the max dissipation on this chip using a good plane for termal disispation?

    Thank You
    Federico

  • 0 in reply to Federico Battaglin
    TI__Mastermind 22299 points

    Federico,

    50mA is worst case when the EMI noise is extremely high at the fundamental frequency. 

    The Rtheta JA is 37.3 degrees.

    So for every 1W the IC temperature should rise 37 degrees.

    Note this is just an estimate and it really depends heavily on board layout design.

    I would say 1W is OK. 2W might not work depending on ambient temperature.

    -Orlando

  • 0 in reply to Orlando Murray
    Intellectual 545 points

    Dear Orlando,

    thank You for your reply and for the table!

    About board layout... it's a bit off-topic, but is it possible two layers, instead 4 or 6 indicate in data sheet and in EVM?

    Thank You
    Federico

  • 0 in reply to Federico Battaglin
    TI__Mastermind 22299 points

    Federico,

    It should be possible, but would be very difficult to route.

    HO, SW, LO are noisy signals, and ISNS+ and VOUT are noise sensitive current sense signals, so it would be difficult to route everything with just 2 layers.

    It would also be harder to pass EMI, as you dont have a GND plane and you cannot route noisy signals internally, they would act as antennae on the outer layers.

    Hope this helps,

    -Orlando