• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 64 subscribers
  • Views 233 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.

LM5143: Constant Current Power Supply for Laser Diode

Michele Faini
Prodigy 30 points
Part Number: LM5143
Other Parts Discussed in Thread: LM5190, LM5137, LM5116, LM5117

Tool/software:

Hi everyone,
I'm Michele Faini and I work in Bios srl like HW Designer.
For our new project I have to design a Constant Current Power Supply (current regulator) for a High Power Laser Diode.


The Energy is taken from a 24Volt Capacitor Bank, to obtain the right current peak. The Current regulator is connected to the Capacitor bank and supply directly the Laser Diode.
The working voltage for Laser Diode is 16 Volts and the regulated current will be about 60/70 Amperes.

It is mandatory using a switching mode configuration to keep the efficiency high.
I thought about using a standard (low cost) switching controller to drive an external bank of MOSFET or GaN (unfortunately more expensive!) to reach the current we need, but the controller MUST have the Constant Current Mode Control, with a sensing series resistor.

Furthermore, the controller must have an high gate current capability to drive multiple parallelized MOSFETs at the switching frequency (about 1MHz to keep the Inductor value low).

The output current must reach the nominal value in about 150/200us, so the BandWidth of the control loop must be at least 10KHz.

I need help from Texas Instruments Field Application Engineer to choose the right controller part number to reach my design target: I found the LM5143 that is designed for constant current, but I know there are other controllers with the current sensing integrated. What Do you suggest?

Since the device will be commercialized for ad least 15 years, we need a newer controller to avoid future component shortage problems.

Waiting for your feedback, I thank you,
Best Regards,
Michele Faini

  • 0
    TI__Guru 68757 points

    Hi Michele,

    Take a look at the LM5190 CC/CV controller.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Prodigy 30 points

    Hi Timothy,

    thank you for your reply.

    I think LM5190 could be a good candidate, but I have a question.

    To manage more current (High current!), I have two ways:

    1) Connect in parallel multiple DC-DC converter every one with an independent controller (LM5190).
    2) Use a single controller (LM5190) and adopt multiple MOSFET connected in parallel to manage more current.

    I prefer the second one method because it will be a cheaper BOM.
    In your opinion, do you think it is possible connect in parallel multiple MOSFETs? 

    Will be the LM5190 capable to drive multiple MOSFET Gates?

    Are there any other restrictions using paralleled MOSFETs?

    Waiting for your answer, I send you my best regards,

    Michele Faini

  • 0 in reply to Michele Faini
    Prodigy 30 points

    Hello again,

    reading the LM5190 datasheet, I found there is a fixed Soft-Start of about 3ms I can't disable!
    I need to be very fast during the startup: about 200/300us
    So i think the LM5190 is not good for my application   

    Any other ideas?

    Are there other chips I can use to have a CC/CV control?

    Thanks in advance,

    Michele 

  • 0 in reply to Michele Faini
    TI__Mastermind 38160 points

    Hello Michele,

    You can add parallel FETs to a single gate drive.  note that you will only be spreading the heat over two MOSFET.  Also make sure the Qg (total) x FSW gate drive currents do not exceed the VCC current limit level.  Looking at your current levels above, I think using a stacked 4 phase design is a more suitable approach.  Consider using the LM5137 (2 devices for 4 phase) for such design.

    Will need to implement external circuit as shown in this application note.

    https://www.ti.com/lit/an/snvaa85/snvaa85.pdf?ts=1745849877875&ref_url=https%253A%252F%252Fwww.google.com%252F

    David.

  • 0 in reply to David Baba
    Prodigy 30 points

    Thank you David, tomorrow I'll read the App Note.

    I see the LM5137 is difficult to find in online catalog like DigiKey or Mouser... I'm afraid using a IC with no market. I prefer something of more "diffused and commercial", to avoid component shortage problems in the future.

    The last days I found a lot of Texas Instruments Controller IC with the sensing resistor inputs but in the datasheets is not specified for CC/CV applications.

    So, now I'm a bit confused because I don't understand the difference between a CV/CC controller (not so diffused) from a normal CV controller with sensing resistor: can I use a standard CV controller (with Rsense feature) and set the maximum limit for the Iout current? Isn't it the same like a CC/CV mode?

    For example the LM5143 is diffused, is cheaper and has two phases I can parallelize or LM5116 has only one phase but is cheaper and I can use more chips.

    Please, could you explain me the main differences?
    thanks
    Michele 

  • 0 in reply to Michele Faini
    TI__Mastermind 38160 points

    Hello Michele,

    CV regulates the voltage; CC regulate the current by translating the current back to a voltage signal fed back to the error amplifier.  You can use any controller; the challenge is clock timing for each phase you implement.  Can just sync them to the same clock but ideally, they should be staggered evenly over the 360 degrees, for example 2 phases will be staggered 180 degrees and 4 phases will be staggered 90 degrees.  tying the comp pins is another challenge there is a circuit online that you can copy to ensure sharing, these will be extra components.

    see below.

    https://www.ti.com/tool/PMP7282

    Hope this helps.

    David.

  • 0 in reply to David Baba
    Prodigy 30 points

    Uhhh your solution is a bit complicate!

    I have 8 Laser Diode, each one of about 500W and I should replicate this circuit 8 times: the BOM will became heavy.

    Instead, I found these two AppNote from TI very interesting. They show it is possible reach 1KW and more with a single controller, simply using paralleled MOSFETs:    
    https://www.ti.com/lit/pdf/slpa020

    In the following AppNote, LM5117 with Current Monitor output is used for current regulation: it seems simpler, don't you think?
    Maybe there is some newer controller with this feature (current monitor output): do you know any newer controller to suggest me?

    https://www.ti.com/lit/pdf/snva829

    THanks,

    Michele

  • 0 in reply to Michele Faini
    TI__Mastermind 38160 points

    Hello Michele,

    Yes, you can use the LM5117 as a very simple solution for regulating output current. 

    Thanks.

    David.

  • +1 in reply to David Baba
    Prodigy 30 points

    I think I'll buy the evaluation board and I'll add my power MOSFET board to it.
    Thank you David!