• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 6 answers
  • Subscribers 64 subscribers
  • Views 941 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-Q1: DC/DC with CC/CV and Input of 20V...90V

user5990144
Prodigy 40 points
Part Number: LM5146-Q1
Other Parts Discussed in Thread: LM5117, , LM5117-Q1

Hello

I have designed a DC/DC with a LM5146-Q1 (20V...90V Input, 13.6V Output, 10A Output), which is working well. But what I'm missing is a constant current (CC) behaviour for a time of at least 100ms. I'have seen the application SNVA829 with a LM5117 device, but the LM5117 allows a maximum of max 65V (recommended). Is there a alternative with a maximum input voltage of 90V?

Kind regards
Daniel

  • 0
    TI__Mastermind 35210 points

    Hello Daniel,

    Can you please elaborate on this?

    "But what I'm missing is a constant current (CC) behaviour for a time of at least 100ms."

    Are you open to fixing current device issue? This device should be capable of handling your requirements.

  • 0 in reply to Marshall_Beck
    Prodigy 40 points

    Hello Marshall Beck,

    That would be great, if it could be working with this device (LM5146-Q1).

    As far as I understand, with the LM5146-Q1 I can set a maximum current, which is hold for 128 clock cycles. I set a frequency of 250kHz -> 128 cycles -> 512us After this, there is a break with 0V output of 8192 clock cycles -> 32.8ms

    I've got this information from page "If the overcurrent condition exists for 128 continuous clock cycles, a hiccup event is triggered and SS is pulled
    low for 8192 clock cycles before a soft-start sequence is initiated."

    How can I configure it, to fix it?

    Daniel

  • 0 in reply to user5990144
    TI__Guru 73402 points

    Hi Daniel,

    Can you confirm if constant current (CC) operation is needed on the output? If so, you'll need a shunt on the output and a current sense amp that ultimately controls FB as required.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Prodigy 40 points

    Hi Tim

    Thanks for your input with the current sense amp.

    Yes, I need it for around 100ms, to start up loads like halogen lamps or loads with a high capacitance. The loads are connected during operation of the dc/dc. In the following Image there is the situation as it is at the moment (blue: Vin, red: Vout, green: Iout). After connecting the halogen lamp, the voltage drops from 13.6V to 11.0V and the current is at 13.7A (which I can set). After heating up the halogen lamp, the voltage would rise again after a few dozen ms. But unfortunately the output voltage is set to 0V after 0.5ms.

    So, the solution with the current sense amp and the shunt would work, but I would prefer a solution without a shunt. Also because the current solution is very close to what I need, except for the time.

  • 0 in reply to user5990144
    TI__Mastermind 35210 points

    If both CV and CC is need, I do not believe we will find an offering.

    Best best would be CV + current sensing ckt, as Tim alluded too.

    In terms of CC, all I see if Led driver.

  • 0 in reply to Marshall_Beck
    TI__Mastermind 35210 points

    I did find this, but unfortunately the voltage range isn't appropriate.

  • 0 in reply to Marshall_Beck
    TI__Guru 73402 points

    Hi Daniel,

    You may use inductor DCR current sensing to avoid the shunt. Just use a larger RC time constant relative to the L/R to avoid the ripple component.

    Here is an app note for CC-CV operation: www.ti.com/lit/snva829

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Prodigy 40 points

    Thank you for your answers.

    Inductor DCR current sensing sounds fascinating, but I think it is not that easy for me to get it to a stable solution.

    Now I have some solutions:

    • The halogen lamp is connected through a MOSFET, so we can use PWM with an on time of < 0.5ms in the first few ms to preheat it. (Which is not so nice because of EMI)
    • Or we limit the input voltage range to 60V and use the LM5117-Q1 (SNVA829)
    • Or we go the hard path and use DCR current sensing or with a shunt and amplifier

    Thank you again!

    Daniel

  • 0 in reply to user5990144
    TI__Guru 73402 points

    Sounds good, Daniel.

    Regards,

    Tim