• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 62 subscribers
  • Views 379 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.

Original question:

LMG1210: Half-Bridge Capability

LMG1210: Half-Bridge Capability with positive/negative supply

mas
Prodigy 75 points
Part Number: LMG1210

I'm working on a design with a half-bridge consisting of two GaN HEMTs using an LMG1210. As a starting point I took the LMG1210 TINA-TI Reference Design (Rev. C) and let it run. It all worked as expected. However, this design is with a single-ended supply of 50VDC referencing to GND. In my design I need to have a symmetric supply (I chose +/- 25V). Unfortunately, with this in place the design doesn't run anymore, i.e. the gate signals HO and LO remain 0V all the time.

I have tried with a bootstrap circuit and also with a seperate 5V supply for HB, which is referencing to HS. Both didn't work. Is there an issue with the model? Based on the discussion in the E2E forum (LMG1210: Half-Bridge Capability), this mode of operation should be supported if a seperate supply for HB-HS is provided.

Please find attached the modified reference design.LMG1210_ReferenceDesign_modified.TSC

  • 0
    TI__Genius 13445 points

    Hello Mas,

    Thanks for reaching out.

    The reference design is not compatible with a symmetric supply due to the models (EPC and LMG1210) internal circuitry likely referenced to a true GND 0V.

    Additionally, the model is limited to 50% duty cycle operation during which the model relies on the VDD supply to provide bias to the HB-HS capacitor through the BST pin.

    In reality, the driver IC will in fact support an external floating bias to charge the HB-HS capacitor and support longer/shorter duration duty cycles.

    I may suggest the unencrypted PSpice model for a more accurate representation of the driver's actual performance.

    Regards,

    -Mamadou  

  • 0 in reply to Mamadou Diallo
    Prodigy 75 points

    Hi Mamadou,

    Thanks for your reply. I'm a bit surprised that the delivered model comes with such poor functionality, especially as the design is called "reference design" and is already in revision C. With all the mentioned restrictions this reference design doesn't provide much more information than a simple plot in the datasheet along with some numbers. I'm aware that all the models come on a "as is" basis. However, some short release notes regarding functional restrictions would help a lot, especially as time is money in R&D.

    Now to the technical part:

    Unfortunately, the suggested usage of the unencrypted PSpice model doesn't work either. I tried to import the model into Tina, but get an error:

    Followed by:

    So, I went ahead and installed PSpice for TI. There, the LMG1210 is already available in the library. Having the whole circuit built up again, the design with symmetric supply shows the same result as described in my original post regarding the "reference design" in Tina. To complete the round dance, I also tested the unencrypted PSpice model after importing it into LTspice. This also showed the same results. Hence, it seems that there is a general issue with the simulation model. Can you please check?

    I zipped the PSpice design and attached it to this post: LMG1210_PSpice.zip

    Thank you,

    mas

  • 0 in reply to mas
    TI__Genius 13445 points

    Hello Mas,

    Thanks for your feedback.

    I will review the PSpice file and reach out to you via email before COB Tuesday.

    Meanwhile, I have sent you a friend request where I will follow up with you separately.

    Regards,

    -Mamadou