• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 64 subscribers
  • Views 458 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.

LM5146: Design help

Marvn Arcangel
Genius 13980 points
Part Number: LM5146
Other Parts Discussed in Thread: TPS25982, LM5117, LMP8646

Hi Team, 

A customer is asking for input protection for boards using LM5146.

Their application has a 200W power limit, but voltage can vary from 10V to 60V due to using different battery setups.

The problem with fuses is that the low current rating is too limiting on 12V batteries, but a high current limiting is dangerous for higher, 24V, and 48V batteries because the power rating is way too high. Please see more information below:

200W @ 12V = ~16A

200W @ 48V = ~4A

48V @ 16A = >700W

Will a TPS25982 or similar efuse help? can you recommend other solution?

Thank you in advance.

Regards,

Marvin

  • 0
    TI__Guru 74315 points

     Hello

    I think that the TPS25982 might be made to work.  Maybe some kind of FET circuit that would sense the input voltage and then adjust 

    the current limit of the TPS25982.

    I will assign this to a more experienced colleague to see if we can help further.

    Frank D

  • 0 in reply to Frank De Stasi
    TI__Guru* 76022 points

    Marvin,

    Note the TPS part is only rated to 24V, so it's not sufficient for your 48V input.

    For a constant power (CP) limit circuit, take a look at this article that uses the LM5117 (Vin-max = 65V) with its CM output. A feedforward resistor from VIN could be used to achieve a constant power characteristic. 

    https://e2e.ti.com/blogs_/b/powerhouse/posts/how-to-design-a-simple-constant-current-constant-power-constant-voltage-regulating-buck-converter

    If you prefer to use the LM5146 (Vin-max = 100V), then a CC circuit with a shunt and current-limiter IC can be used in place of the LM5117's CM pin functionality. Take a look at the app notes in the LMP8646 product folder for this function.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    TI__Genius 13980 points

    Hi Tim,

    Thank you. I have a follow-up question. According to the article, in the new circuit arrangement, rather than just having two feedback resistors, it adds voltage reference for constant voltage, and in the customer's case, they'll have to use the LMP8646 for constant current.

    Would you kindly explain why the voltage reference is required if the feedback resistor arrangement should be enough to set the output voltage? Is that because the Rff will affect output voltage too much otherwise?

    Also the Rff goes from Vout to output of LMP8646 to get it similar to the LM5117 CP circuit?

    Thanks again,

    Regards,

    Marvin

  • 0 in reply to Marvn Arcangel
    TI__Guru* 76022 points

    Hi Marvin,

    The Rff can come from Vin to get the CP characteristic here (to effectively reduce the CC setpoint as Vin increases). Take a look as well at app note SNVA829. You need FB resistors for constant Vout and loop compensation components. The LMP8646 is just a handy way to inject current into FB and reduce VOUT when CC is reached (it has a series diode so is effectively disconnected in CV mode).

    --

    Tim

  • 0 in reply to Timothy Hegarty
    TI__Genius 13980 points

    Hi Tim,

    Good day. 

    Just to clarify, the "Vin" mentioned means the +IN of the LMP8646 device right? So one side goes to feedback of LM5146 and other to +IN of LMP8646? Then it would be in parallel with the Rsense resistor of the LMP8646.

    Regards,

    Marvin

  • 0 in reply to Marvn Arcangel
    TI__Guru* 76022 points

    Hi Marvin,

    If you need constant power, then feedforward from the input supply rail (VIN) of the regulator is required (to reduce the CC setpoint as VIN increases).

    --

    Tim