• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 3 answers
  • Subscribers 28 subscribers
  • Views 529 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:

BOOSTXL-DRV8323RS: TDRIVE Testing and Dead Time

BOOSTXL-DRV8320H: Vgsth temperature drift when using a VGS monitor to determineinsert a deadtime

zhipeng hu
Prodigy 30 points
Part Number: BOOSTXL-DRV8320H
Other Parts Discussed in Thread: DRV832X

Hi Expert 

 I am develop a motor controller with DRV8320HRTVR， In my application I used a DRV8320HRTVR to drive a MOSFET which typical Vgsth= 1.7V  and the minimum is 1.2V.

and The DRV832x family of devices uses VGS voltage=2V monitors to measure the MOSFET gate-to-source voltage  to  determine insert a fixeatime=100ns;

My question is :Does  the DRV832x considered about  temperature drift or variation between  MOSFETs Vgsth when use   VGS voltage=2V monitor to determine insert a fixeatime.

thanks a lot

Zhipeng

  • 0
    TI__Mastermind 18836 points

    Hey Zhipeng,

    Thank you for your question.

    The VGS threshold is based on the driver gate sense and is irrespective of the MOSFET VGS threshold.  The 2V is an approximation and does not account for temp drift. Since your MOSFET VGS TH is below 2V the driver VGS cannot be used to tell if the FET is actually OFF. 

    The driver uses 2 V approx to decide when to start adding deadtime so in your application you may need to add extra deadtime to account for when the FET actually turns off and this can be tested and finalized for end application.

    Best,
    Akshay

  • 0 in reply to Akshay Rajeev Menon
    Prodigy 30 points

    Hi Akshay

    Thanks for your reply.

    But there is a descripment about " temperature drift and variation in the MOSFET parameters ", So what is this feature talk

     about?


    Thank you

    Zhipeng

  • 0 in reply to zhipeng hu
    TI__Mastermind 18836 points

    Hey zhipeng,

    I will look into it and provide feedback.  

    Best,

    Akshay

  • 0 in reply to Akshay Rajeev Menon
    Prodigy 30 points

    Hi Akshay

    Another problem is which is the right deadtime as show in the picture？

    deadtime1：the time between VGS=2V for turn off  MOSFET and VGS=0V for turn on  MOSFET ；

    deadtime2：the time between VGS=2V for turn off  MOSFET and VGS=2V for turn on  MOSFET ；

    Thank you

    Zhipeng

  • 0 in reply to zhipeng hu
    TI__Mastermind 18836 points

    Hey Zhipeng,

    Sure, I will check with my team and provide feedback.

    Best,

    Akshay

  • 0 in reply to Akshay Rajeev Menon
    TI__Mastermind 18836 points

    Hey Zhipeng,

    The temperature drift is accounted in the sense that the mechanism is based on voltage at the gate so temp change affecting MOSFET parameters (causing rise/fall in gate voltage be it faster or slower) would be accounted for as it checks gate voltage at the pin.

    The timer for the deadtime starts at VGS = 2V and needs to run out before the MOSFET is allowed to be turned on.

    Best,
    Akshay

  • 0 in reply to Akshay Rajeev Menon
    Prodigy 30 points

    Hi Akshay

    Thanks for your reply.

    What is the temperature drift rate parameter used for compensating gate voltage drift in the 8320? Is it 7mV/°C or something else?

    In the case of the deadtime issue, what voltage is used for the Another MOSFET to be allowed to turn on（which means when the timer of deadtime stopped）? Is it Vgs=2V or 0V (starting the rise of Vgs)?

  • 0 in reply to zhipeng hu
    TI__Mastermind 18836 points

    Hey Zhipeng,

    An open loop deadtime inserts the time regardless of the gate turnon/off state, however, our DRV device measures the gate voltage to reach the threshold before inserting deadtime. If closed loop deadtime is used then the gate voltage is measured to check if the FET is turning off (based on the threshold) before inserting the deadtime. If you don't measure the gate voltage then the temperature shift might result in the gate not turning off when the deadtime is inserted.

    The timer starts at the falling (2V) then stops at around the start of the rising (around 0V)

    Best,
    Akshay

  • 0 in reply to Akshay Rajeev Menon
    Prodigy 30 points

    Hi Akshay

    Thanks for your reply.，
     Now，I can understand the logic about  deadtime timer.
    However, I still have doubts about the temperature compensation for Vgsth.
    For example, when the temperature is 25℃, the device 8320 uses Vgs=2V to insert a deadtime of 100ns.
    So, when the temperature is -20℃, what is the Vgs value that insert a deadtime of 100ns?
    Also, when the temperature is +100℃, what is the Vgs value that insert a deadtime of 100ns?

  • 0 in reply to zhipeng hu
    TI__Mastermind 18836 points

    Hey Zhipeng,

    The temp compensation is as mentioned below:

    The temperature drift is accounted in the sense that the mechanism is based on voltage at the gate so temp change affecting MOSFET parameters (causing rise/fall in gate voltage be it faster or slower) would be accounted for as it checks gate voltage at the pin.

    There might be slight variation in the vgs threshold voltage but mainly our driver accounts for temp variation by checking the gate voltage at the pin.

    Best,
    Akshay