• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 4 answers
  • Subscribers 41 subscribers
  • Views 939 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.

TIDA-00778: Suppy Current for gate driver

user4008740
Intellectual 915 points
Part Number: TIDA-00778
Other Parts Discussed in Thread: TIDA-00472, UCC27714, TIDA-00364, TIDA-00173

Dear TI Experts:

When we design an external power supply for the gate driver, we need to determine the maximum current output of the power supply.

For TIDA-00778, 15-V supply current for gate driver is 50 mA (max), as shown in Table 2 on Page 11.

But in Section 1.4.3.1, we can find I_HO_DR=1.47A and I_LO_DR=1.53A.

So, could you kindly show me how to determine the 50 mA (max), shown in Table 2 on Page 11.

Look forward to hearing from you~

Best Wishes,

Tang Han

Ps.

For TIDA-00472, 15-V supply current for gate driver is 50 mA (max), too, as shown in Table 1 on Page 2.

But in Section 4.3.1, we can find I_HO_DR=1.05A and I_LO_DR=1.09A.

  • 0
    TI__Prodigy 425 points
    Dear Tang,
    Many thanks for your question. Unfortunately the designer will only be back in office on 16th April 2019. We will come back to you asap.
    Michael
  • 0 in reply to Michael Seidl
    Intellectual 915 points

    Dear Michael:

    Thanks for your kind and instant reply.

    It's OK~ I will wait for it patiently.

    Best Wishes,

    Tang Han

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Hi Tang,

    Thanks for your patience.

    The section 1.4.3.1 of TIDA-00778 explains the peak current requirement. The 1.47A, 1.53A are all the peak source current required to turn on the external MOSFET/IGBT faster and the time to turn on the MOSFET/IGBT is typically <1us. Once the IGBT turns on, then the gate current requirement reduces to a very small (typically a few micro amperes) to supply the leakage. Refer figures 33- 35 of TIDA-00778 design document to see the gate current waveforms. This means the average current requirement from the gate driver is very small compared to peak current 1.53A.

    If the IGBT with Qg = 270 nC is switching at fsw = 20kHz, then the approximate average current required to turn on the IGBT = Qg x fsw = 270nC x 20 kHz = 5.4mA.

    Since we have six IGBTs, the average current for driving all the six IGBT = 5.4 mA x 6 = 32.4 mA.

    In the design, 50 mA max is assumed. The peak currents of 1.47A, 1.53A are to be supplied with the decoupling capacitor of the gate driver and the boot strap capacitor, and those capacitors should be properly sized as explained in the design document.

    With trapezoidal control, the IGBTs are switching only during 120deg (1/3rd of total period) and the average current will be even lower than the above calculated value.

    Hope this clarifies your question. 

    Thanks,

    Manu

  • 0 in reply to Manu Balakrishnan
    Intellectual 915 points

    Dear Manu:

    I am so grateful to get your kind, experienced, and professional reply.

    You really have taught me a lot!   ^_^

    "The 1.47A, 1.53A are all the peak source current required to turn on the external MOSFET/IGBT faster and the time to turn on the MOSFET/IGBT is typically <1us."

    I am curious to know how to determine the values 1.47A and 1.53A. Is there any formula?  Knowing these values, we can determine these values of resistors in the gate driver ciircuit, right?

    In TIDA-00472, the corresponding values are 1.05A and 1.09A. 

    Could you kindly show us how to determine the peak source current?

    Wish you the best!

    Tang Han

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Hi Tang,

    The design to calculate the peak gate current depends on the MOSFET/IGBT total gate charge and how fast we need to turn on the MOSFET/IGBT. Typically, in order to minimize the switching losses, we need to switch the MOSFET/IGBT fast by providing the maximum peak gate current. However, there are some practical concerns with faster switching like voltage over shoot at the phase node due to PCB parasitics, higher reverse recovery of the diode, EMI issues etc.

    While designing, we start with an approximate peak current value depending on the MOSFET/IGBT gate charge and the required turn on time of the FET/IGBT (for ex: 200ns). Then during the board testing, the peak gate current value is tuned(by adjusting the gate resistor) to optimize between switching loss, phase node voltage overshoot/ringing, diode reverse recovery, EMI etc.

    For more reading you can refer below document and training.

    http://www.ti.com/lit/ml/slua618a/slua618a.pdf

    https://training.ti.com/battery-powered-motor-drives-power-stage-design

    Regards,

    Manu

  • 0 in reply to Manu Balakrishnan
    Intellectual 915 points

    Dear Manu:

    Thank you so much for your detailed explanation and for introducing me valuable material.  You are so kind to us.

    You mentioned, "The peak currents of 1.47A, 1.53A are to be supplied with the decoupling capacitor of the gate driver and the boot strap capacitor". For UCC27714, it has 4A sink and 4A source current. So the peak currents (1.47A, 1.53A) must be less than 4A, right?  Thus the selection of gate driver chip is also important, right?

    On the other hand, if we apply parallel operation of power MOSFETS  (like TIDA-00364).  With the same gate driver circuit in TIDA-00778, if the number of parallel MOSFETS is 3, we should increase the size of boot strap capacitor, right?  It may cause slow switching, which means slower rotation in PMSM motor control, right?

    Best Wish for you~

    Tang Han

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Hi Tang,

    Please find below my comments on your questions

    1. Your understanding is correct on the selection of gate driver. The peak gate current (1.47A/1.53 A) should be less than the peak current gate driver is specified for  (4A for UCC27714)

    2. When you parallel FETs please consider the total gate charge (gate charge of all 3 paralleled FETs) while designing the bootstrap capacitor. Switching speed depends on how much is the designed gate current per FET. It is not directly related to your motor speed. Your motor speed mainly depends on the duty cycle you are operating. When you switch very slowly, the effective duty cycle may reduce a bit and will have very small reduction in speed, which may not be noticeable.

    I hope that clarifies your question. If yes please click "resolved". If you have more question, please go ahead and open a new thread.

    Thanks,

    Manu

  • 0 in reply to Manu Balakrishnan
    Intellectual 915 points

    Dear Manu:

    Your answers really clarify my questions. You are so learned and professional.

    Could I ask for your one more guideline?  

    I just found that power MOSFETS "in series" can improve the limit of Vds of a power MOSFET. But it seems that TI only provides practical reference designs with MOSFETS "in parallel", right?

    Could you kindly show me some practical reference designs with MOSFETS "in series" or related teaching materials?

    If I open a new thread, I am afraid that it will not be you to answer the question. So I have to propose the question here.

    Regardless of your answer, I will click "resolved" and be grateful to you.

    ^_^

    Wish you the best!

    Tang Han

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Hi Tang,

    You can have a look at the reference design TIDA-00173, which shows the series connection of MOSFETs in a flyback power supply.

    For the new thread, if you have a question related to TIDA-00778, I will be able to answer it.

    Thanks,

    Manu

  • 0 in reply to Manu Balakrishnan
    Intellectual 915 points

    Dear Manu:

    Thanks for introducing the reference design TIDA-00173.
    But it seems that its technical documents are not available at www.ti.com/.../TIDA-00173

    I can't help but ask for your help again.

    I am sorry to bother you so much~

    Wish you the best!

    Tang Han

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Hi Tang,

    This could be a glitch in the system. All the documents are available in the TIDA-00173 page now.

    Thanks,

    Manu

  • 0 in reply to user4008740
    Intellectual 915 points

    Dear Manu:

    The technical documents of TIDA-00173 have been available now.
    Thanks for your professional and kind help!
    Wish you the best!
    Tang Han
  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Good to hear that it is solved.

    Thanks,

    Manu

  • 0 in reply to user4008740
    TI__Intellectual 1260 points

    Thanks for confirming.

    All the best,

    Manu