• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 4 answers
  • Subscribers 64 subscribers
  • Views 166 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.

LMG2100R044: GaN driver usable at 10MHz

Kyohei
Expert 1420 points
Part Number: LMG2100R044
Other Parts Discussed in Thread: LMG2100EVM-078, LMG3100EVM-089,

Tool/software:

Hi team,

My customer is considering using the LMG series in an ultrasonic cleaner (related to semiconductor manufacturing).

The required specifications are as follows:

  • Vds: 100V (650V is also acceptable)
  • Current: 5A
  • Frequency: 10MHz or higher
  • GaN + driver IC
  • Full bridge usage

Based on these conditions, I am proposing the LMG2100R44 and LMG3100R44.

However, although the device itself can switch at 10MHz, the EVM is specified for use at a switching frequency of 100kHz to 500kHz. What is the reason for this?

Also, are there any evaluation boards that can be used to evaluate devices at a frequency of 10MHz under the above conditions?

Best regards,

Kyohei

  • 0
    TI__Expert 4855 points

    Hi Kyohei, 

    The LMG2100 and LMG3100 can be used within many different applications - the EVMs simply use these devices in the context of a buck or boost converter. The user can provide the PWM inputs at their desired switching frequency and the LMG devices will not run into any issues in the MHz range. However, depending on conditions such as load current, voltage, and dead-time setting, there may be a practical limitation to high the switching frequency can be set to. This is due to the limitation of other components on the board such as the inductor. 

    Additionally, if the Vds of the application is 100V, LMG2100 and LMG3100 will not be suitable here as there needs to be at least an 80% margin of operation. The higher voltage devices such as LMG3650 would have to be considered here. 

    Thanks,

    John

  • 0 in reply to John Gomez
    Expert 1420 points

    Hi John,

    Thank you for your response.

    The specifications for the LMG2100-EVM are as follows.

    Regarding Vds, 100V is sufficient, and it is planned to be used in a design where it will not exceed 80V.

    Due to time constraints, the customer does not have the capacity to create a board and wishes to evaluate using the EVM.

    However, there are limited EVMs available for evaluation at 10MHz.

    Do you have any good suggestions?

    Best regards,

    Kyohei

  • 0 in reply to Kyohei
    TI__Expert 4855 points

    Customer can swap out inductors and capacitors as needed to be able to operate at 10MHz.

    Thanks,

    John

  • 0 in reply to John Gomez
    Expert 1420 points

    Hi John,

    Thank you for your reply.

    I understand that it is possible to support 10MHz by adjusting the inductor and capacitor.

    Could you please provide the values of L and C that need to be changed to support 10MHz for both the LMG2100EVM-078 and LMG3100EVM-089 evaluation boards?

    Best regards,

    Kyohei

  • 0 in reply to Kyohei
    TI__Expert 4510 points

    Hi Kyohei,

    John is out of office now, I can help support here.

    A big issue with inductors at high frequency operation is AC core losses. There is not one recommendation or calculation that can determine the best inductor to use, since there are many factors that determine an inductor's performance at high FSW. The value inductance is fine, what needs to change is the type of inductor.
    For example: the LMG2100R044 EVM uses a coilcraft inductor. There are other inductors from coilcraft which will have much lower AC losses and may be better suited for these higher switching frequencies like the XGL3014 Series.
    A second example: Using an air core inductor will also eliminate AC losses and may be very well suited for these high frequencies, multiple manufacturers make them.

    In summary, although inductance value will change the ripple current, the main issue is AC losses which can be mitigated by using different inductors with low AC losses, or using an air core inductor.

    Thanks,
    Zach Soviero

  • 0 in reply to Zachary Soviero
    Expert 1420 points

    Hi Soviero,

    Thank you for your response.

    I understand that an air core inductor is recommended. Thank you.

    There was also a comment (from John) about the need to change the capacitor initially. Do you have any recommendations regarding this?

    Best regards,

    Kyohei

  • 0 in reply to Kyohei
    TI__Expert 4855 points

    Hi Kyohei, 

    The caps should actually be okay, since the ripple will be smaller at higher frequency.