• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 203 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.

INA381: - Vcm higher then 26V

Sunset
Genius 13395 points
Part Number: INA381

Hello Team,

in the datasheet it is written that the INA381 may withstand higher Vcm in case the current on the inputs is limited by 5mA max.

On the PCB we have already the filter resistors on IN+ and IN- as in the datasheet section 8.1.4, that can be used to limit the current.

30V is not a normal working condition, but a fault condition. If this will happen, the INA381must not get broken as well.

I would assume that if Vcm is 30V the fault current is 30/Rin+ < 5mA thus Rin+ must be > 6K.

  • Does the internal INA381 input stage allow to use the flowing formula (30V - 26V)/R instead? 

6K on the input is quite high, the tolerance on that will create an error (which is fine).

  • What is the input impedance of the INA381?

Thanks,

SunSet

  • 0
    TI__Mastermind 26730 points

    Hey SunSet,

    To calculate the required resistance use (30V-26V)/5mA, so Rin+ = Rin- > 800Ω. The differential input impedance of the device is 2.5kΩ ±15%. Equation 2 of the datasheet accounts for this differential resistance. Use Equation 2 to determine how the gain will be attenuated when using input resistors. For example, if input resistors (RF) are 1kΩ and the gain is INA181A4 (200V/V), then the new gain is 200*GainErrorFactor = 90.91V/V.

    Sincerely,

    Peter

  • 0 in reply to Peter Iliya
    TI__Genius 13395 points

    Hello Peter,

    thanks for the information.

    For the Formula (30V-26V)/5mA, to be valid sounds like that there is a "zener" in the input that will reduce the Vcm.

    otherwise I do not fully understand why to compute the current I should use 30V-26V rather then 30V.

    Can you better explain it?

    Thanks,

    SunSet

  • 0 in reply to Sunset
    Guru 140200 points

    Hi Sunset,

    I guess Peter's formula handles the worst case: 26V is the earliest voltage at which the input can break down.

    Kai

  • 0 in reply to Sunset
    TI__Mastermind 26730 points

    Hey SunSet,

    The input ESD cells are pretty complex structures and are meant to begin conduction once the input voltage exceeds the Absolute Maximum rating of 26V. There is no internal Zener. The part can operate with VCM at 26V, yet you don't need a resistor to limit current because the ESD cell is not conducting.

    By using the equation R > (30-26V)/5mA, you ensure that once the input voltage reaches 30V that the input current is < 5mA. This is a worst-case scenario as it assumes conduction will start at 26V, but there will be some margin and conduction might not start until 28V for example. To make the design more robust I would recommend reducing the conduction current to 3mA instead of 5mA. So calculation would be (30V-26V)/3mA = 1333.333Ω. This also ensures that at worst-case input voltage (30V), the voltage at input pins would be 26V due to voltage drops at 3mA.

    Hope this helps.

    Sincerely,

    Peter

  • 0 in reply to Peter Iliya
    TI__Genius 13395 points

    Hello Peter,

    thanks for the info. The behavior you have just described is wath I meant for the "zener".

    Since Vcm can be higher than Vcc, make sense that the ESD diodes do not actually get in conduction before that voltage.

    Standard ESD are connected to Vcc and they get conducting once the voltage is higher then Vcc or lower than ground of about ~0.5V.

    In this case the ESD diode they looks more zener/TVS rather than standard diodes.

    @Kai,

    As worst case I would have considered 30V. Out of the understanding on how the ESD would work, Vcm-26 works out.

    Now I would need to check the datasheet to see if it is written.

    Thanks,

    SunSet