• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 144 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.

LM5575: 24V In -17V 0.5A Out Design Help

Graham North29
Intellectual 385 points
Part Number: LM5575
Other Parts Discussed in Thread: LM46002, LM5164-Q1, LM5164

I've used Webench for the following design:

24V In

-17V Out @ 0.5A

500kHz switching frequency

The Webench design did not include the snubber R/C that datasheet and eval board shows. Would these still be a good idea for this design? Would I connect them to GND or somewhere else?

I'm presuming I need to connect the EPad to the Output rail? That's not going to be great for sinking heat out of the device.

This part needs a lot of external components. Is there another part that requires less?

Thanks,

G

  • 0
    TI__Mastermind 23815 points

    Hi Graham,

    The RC snubber should be connected to the -Vout rail for LM5575 since that is the lowest potential on the board. 

    The EPad should be connected to output rail as long you have enough copper, heat can still spread regardless the net name

    One other part that may help is LM46002

    https://www.ti.com/lit/an/snva722b/snva722b.pdf

    Thanks

    -Arief

  • 0
    TI__Mastermind 22299 points

    Hey G,

    What is your minimum VIN?

    In this inverting buck boost topology, the average inductor current  IL = IOUT * [1+( |VOUT| / VIN )]

    For a lower VIN, a larger inductor current is needed, and a larger current device has to be used.

    The LM5164-Q1 has an inverting buck-boost model in WEBENCH.

    However for a 17V 0.5A output the minimum VIN is 18V, any lower & WEBENCH will not find this part as a possible solution.

    Otherwise the LM5575 seems to be the best option.

    Also the exposed pad should connect to the PGND pin of the IC, yes in this topology that is the nVOUT node.

    Hope this helps,

    -Orlando

  • 0 in reply to Arief Hernadi
    Intellectual 385 points

    Hi Arief,

    Thanks for the clarifications on the connections.

    Yes, copper is copper, but my ground plane will be much larger than my negative rail :-)

    The LM46002 does look interesting. How would I choose between which is the best between the two for this design?

    Thanks,

    G

  • 0 in reply to Orlando Murray
    Intellectual 385 points

    Hi Orlando,

    The 24V In will be pretty well regulated, it will come from a AC to DC open frame supply, so I'd expect 23V to 25V max. So I should be ok yes?

    I had spotted the LM5164 and have another thread of questions about that part on the forum which I think you are also helping on.

    Thanks,

    G

  • 0 in reply to Graham North29
    TI__Mastermind 23815 points

    Hi Graham,

    I think in general, whatever makes the design easier to implement is one way to go :)

    Other consideration is efficiency, thermal and load transient performance. Some of these parameters you can get a look ahead through Webench or TI Spice Simulation. 

    Thanks

    -Arief