• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 1686 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.

LM5050-1: Using an ideal diode controller as high side driver.

Adrian Bolden
Prodigy 170 points
Part Number: LM5050-1
Other Parts Discussed in Thread: LM5050-2, TPS2490, LM5069, LM5060

The LM5050-1 is a really compact ideal diode driver.

I'm wondering if it can be used as a high side NFET switch as well by fooling the chip to boost all the time, then turning it on/off by controlling the GND.

This would make one of the smallest high side boost circuits that I can find...

Is there any reason this is a terrible idea? Maybe I've missed a different IC that does this better?

Please  note that the symbol below is for the LM5050-2 while I'd like to use the LM5050-1 since that part can power the boost circuit from either side of the NFET. The LM5050-2 only powers from IN pin whereas the LM5050-1 powers from the VS pin.

Thanks!

  • 0
    TI__Guru 57737 points

    Hi Adrian,

    Yes, You can use a Schottky diode from IN to OUT to fool the system thinking that the voltage drop across IN to OUT is positive to force the Gate high. 

    Hope you are aware that LM5050 can drive only drive NFET's which can block reverse current but cannot drive NFET in an orientation to block forward current. 

  • 0 in reply to Praveen GD
    Prodigy 170 points

    Thanks Praveen,

    Yes, In the image I am only passing current from Right to Left, so the back-diode of the high side NFET will not conduct. This is not very clear in my drawing since the "IN" pin is on what I would consider the output  (load side) of the circuit.

    My main concern is verifying that if I have 0V on the Left side, and source voltage on the Right side when I power up the LM5050-1 that it will bootstrap into operating mode. It appears from the datasheet that the 12V boost circuit will run from VS pin even if the IN pin is at 0V.

    The LM5050-2 will not bootstrap itself in that situation as it doesn't have VS pin. I have verified this.

    I have a LM5050-1 arriving today to verify this circuit.

  • 0 in reply to Adrian Bolden
    TI__Guru 57737 points

    Hi Adrian,

    LM5050-1 has not been used in this configuration before. Please let us know the results obtained from your testing. 

    Instead of using LM5050-1 with reverse current protection disabled for blocking the forward current from IN to OUT, why don't you use a Hotswap / eFuse for this task with the IN and OUT connected in reverse polarity compared to your circuit ?

    If you can let us know the Max input voltage and Max load current, I can suggest you a suitable Hotswap / eFuse for your application. 

  • 0 in reply to Praveen GD
    Prodigy 170 points

    Praveen,

    The application is 40V max. with 15A nominal, 25A peak pulse current. Any part suggestions would be welcome!

    My main reason for attempting to use the LM5050-1 is the tiny size and low pin count.

  • 0 in reply to Adrian Bolden
    TI__Guru 57737 points

    Hi Adrian,

    You can consider using LM5060, LM5069, TPS2490 for your application. They all come in a 3mm x 3mm package and are designed to drive NFET to block forward current. 

    You can also refer to Hot Swap Selection Tool - Texas Instruments document which helps you in selecting a right part for your application. 

  • 0 in reply to Praveen GD
    Prodigy 170 points

    Thanks Praveen, I am also going to try LM5060, it looks good.

    The only reason I could see for the LM5050-1 option is that it has wider spacing on the pins and can be used on 2 or 3oz copper.

    The LM5060 is the VSSOP-10 package and if you were making a design that needed 2 or 3oz copper, I'm not sure the trace/space would work.

    This type of component is usually next to larger FETs on thicker copper, and having tiny pin spacing does constrain a design.

    Thanks for your help.