• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 416 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.

LM74912-Q1: Adding additional ideal diode stage

Joe Miller1
Intellectual 325 points
Part Number: LM74912-Q1
Other Parts Discussed in Thread: LM74930-Q1

Tool/software:

Dear Sir/Madam,
Sorry for the similar questions I have been asking based on
LM74810-Q1, LM74930-Q1 and now LM74912-Q1. Just trying to avoid a re-spin.
Look like the LM74912-Q1 would be a better fit than the LM74930-Q1 for my application as I save a little space
and still get output short and over current protection.

Application description:
* Protect against output short and over current when EN is 5V.
* VBATT is 24V.
* When EN is 5V, Q1, Q2 and Q3 are on and VOUT is 24V.
* When EN is 0V, Q1, Q2 and Q3 are off. Q3 protects LM74912-Q1 and Q2 via body diode up to 150V.
* VOUT has an induced voltage which can reach 50 to 80V. A TVS will clamp transients above this range.

I dont have the LM74912-Q1 to test. I have however verified option 1 on the LM74810-Q1 EVM. 

Not sure which option will work best.

Option 1. Q3 controlled by HGATE.
Option 2. Q3 controlled by DGATE.

Regards Joe

  • 0
    TI__Mastermind 37910 points

    Hi Joe,

    Q3 controlled by HGATE is the only option.

    Next, it is important to note that LM74912-Q1 only offers short circuit protection and no over current protection.

    Also, Once EN is pulled low, HGATE may be low in around 10uS and during this time interval, if the transients is seen, damage can happen.

    Regards,

    Shiven Dhir

  • 0 in reply to Shiven Dhir
    Intellectual 325 points

    Hi Shiven,
    Thank you for the reply.

    Noted about no over current. Short circuit is more than enough.

    Please let me know if you see any potential issue with the following.
    I still thing DGATE will also work.

    In the application one LM74912-Q1 (Q1, Q2, Q3) is connected to the motor speed1 wire and
    another LM74912-Q1 (Q1, Q2) is connected to motor speed2 wire.
    speed1 and speed2 wires are mechanically coupled.
    That is when you apply a voltage to the speed1 wire a voltage is induced into the speed2 wire.
    Same when you apply a voltage to the speed2 wire a voltage is induced into speed1 wire.
    Applying a voltage to speed2 wire induces a much bigger voltage into the speed1 wire.
    This is the reason why I want to add an additional ideal diode stage to the LM74912-Q1 output.

    The process currently working with LM74810-Q is as follows. Just want to add output short protection.

    initial:
    * LM74912-Q1 for speed1 wire is shut-down.
    * LM74912-Q1 for speed2 wire is shut-down.

    (1) Speed1 switch is closed.
    * LM74912-Q1 is enabled and provides 24V to speed1 wire.
    * Around 20V is induced into speed2 wire. Q1 and Q2 with 60V VDS is fine. TVS provides clamp at 58V.
    * Speed1 switch is open.
    * LM74912-Q1 is shut-down and active load is enabled to discharge speed1 wire.

    (2) Speed2 switch is closed
    * LM74912-Q1 is enabled and provides 24V to speed2 wire.
    * Around 60V is induced into speed1 wire. Q1 and Q2 with 60V VDS is fine. Q3 provides 150V VDS. TVS provides clamp at 126V.
    * Speed2 switch is open.
    * LM74912-Q1 is shut-down and active load is enabled to discharge speed2 wire.

    Regards Joe

  • 0 in reply to Joe Miller1
    TI__Mastermind 37910 points

    Hi Joe,

    I would still recommend using HGATE for reason explained earlier. Next, if using DGATE, during reverse connection, Q3 is bound to get damaged due to VGS violation.

    Regards,

    Shiven Dhir 

  • 0 in reply to Shiven Dhir
    Intellectual 325 points

    Hi Shiven,
    If VBATT is 24V and is reversed wont DGATE and HGATE both be at -24V and Q3 VGS will be -24V?
    NMOS VGS is specified as +/-20V so Q3 will get damaged.
    Regards Joe

  • 0 in reply to Joe Miller1
    TI__Mastermind 37910 points

    Hi Joe,

    Only DGATE will be at -24V because "A" will be negative. OUT will be protected because it is after the ideal diode stage and HGATE will be 0V.

    Hence it is safe to drive with HGATE.

    Regards,

    Shiven Dhir

  • 0 in reply to Shiven Dhir
    Intellectual 325 points

    Hi Shiven,
    I dont have the LM74912-Q1 only LM74810-Q1 EVM.
    I tested reverse 24V on the LM74810-Q1 EVM without Q3 and marked up the voltages on the LM74912-Q1 diagram.
    I expect the output of Q3 to be at around -22.9V.
    All specs for Q3 look fine to me.
    Does this seem Ok to you?

    Regards Joe

  • 0 in reply to Joe Miller1
    TI__Mastermind 37910 points

    Hi Joe,

    Looks like voltage at pin C is -23.2V which is not expected at all. Can you share the schematic with me?

    Regards,

    Shiven Dhir

  • 0 in reply to Shiven Dhir
    Intellectual 325 points

    Hi Shiven,
    The voltages measured were on the LM74810-Q1 EVM.
    I assume that the LM74912-Q1 work in a similar way.

    I checked the measurements on another LM74810-Q1 EVM.
    There is something wrong with the first LM74810-Q1 EVM.

    The results now look like what I would expect.

    Regards Joe

  • +1 in reply to Joe Miller1
    TI__Mastermind 37910 points

    Hi Joe,

    The measurements look fine now.

    Regards,

    Shiven Dhir

  • 0 in reply to Shiven Dhir
    Intellectual 325 points

    Hi Shiven,
    Thank you.
    I will go ahead and use HGATE to control Q3.

    Regards Joe