• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 35 subscribers
  • Views 738 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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:

PCA9306: PCA9306 as switch and voltage translator

PCA9306: PCA9306: 200k pull-up on Vref2

prabhakaran jayabalan
Prodigy 20 points
Part Number: PCA9306

Hi,

I have question on the resolved topic. "PCA9306: PCA9306 as switch and voltage translator".

Below is the circuit used in my application to interface imx28 processor to I2C IO expanders. This circuit without 200k pull on the Vref2 works well for last three years and no issues observed.

* Is 200K pull up is to limit current from Vref 2 to Vref 1 during Failure (short between Vref2 & Vref 1 in IC)? 

* Not using 200k pull-up will  cause short on Vref2 and Vref1?

*  Not using 200k pull-up will  cause any signal integrity issue?

  

Thanks,

Prab

 

  • 0
    TI__Guru 73271 points
    Hey Prab,

    "Below is the circuit used in my application to interface imx28 processor to I2C IO expanders. This circuit without 200k pull on the Vref2 works well for last three years and no issues observed."
    Doesn't look like your image was posted.

    "* Is 200K pull up is to limit current from Vref 2 to Vref 1 during Failure (short between Vref2 & Vref 1 in IC)? "
    This is correct, it limits current between Vref1 and Vref2 but also limits inrush current into the enable pin as it is a FET of a gate (high capacitance).

    "* Not using 200k pull-up will cause short on Vref2 and Vref1?"
    Depends on circuit, but normally yes. (I don't see the circuit you are referencing)

    "* Not using 200k pull-up will cause short on Vref2 and Vref1?"
    Not using a current limiting could cause issues in separating the two voltage rails (when used as a translator). You could get away with using a 10k resistor for instance, the biasing current would just be larger.

    Thanks,
    -Bobby
  • 0 in reply to BOBBY
    Prodigy 20 points

    Thanks for your quick response Bobby!

    I have 10k for EN, that will limit the inrush as you mentioned.

    I have attached the circuit, Vref2 connected to 5V directly. with this circuit we didn't observe any issue or failures with many samples tested for worst cases/ reliability tests for last three years. do you see any chance of failure with this circuit without 200k pull up? 

    Thanks,

    Prab

  • 0 in reply to prabhakaran jayabalan
    TI__Guru 73271 points
    Hey Prab,

    It doesn't look like the enable and Vref2 are shorted together so what we would see when enable is turned on is large current during operation between Vref1 and Vref2. The other thing I see is that the device would not separate the two voltage rails correctly because the enable pin is driven past Vref1+Vth.

    If you short Vref2 and the enable pin then this will fix both issues. (10k would be acceptable as well)

    Thanks,
    -Bobby
  • 0 in reply to prabhakaran jayabalan
    Guru 317020 points

    The PCA9306 architecture is explained in Voltage-Level Translation With the LSF Family (SLVA675).

    The 200 kΩ resistor is needed to force the voltage at gates of all the transistors down to Vref_A + VTH.

    Without a resistor at Vref_B, there is no reasonable limit on the current flowing from Vref_B to Vref_A.