• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 46 subscribers
  • Views 148 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.

ADC081C027: Tri-state address input pins

Flemming Monsrud
Prodigy 10 points
Part Number: ADC081C027

Dear Texas Support,

The ADC081C027 has one address pin, which can be tied to VA, ground or left floating. The data sheet provides a nice simple diagram of the input, showing two 41.5 kΩ providing the mid-voltage when the pin is floating. It also says you can add two resistors yourself to the pin, if you don't like the pin left floating.

I would simply like to add a 100nF capacitor from the pin to ground instead of two resistors. Would that be ok? I guess my question is: Does the IC just detect the voltage on this pin, either VA, ground or in between, or does the pin also output something during operation? If it's solely an input pin, a capacitor should be enough.

I'm sure you understand why I suggest a capacitor. Just leaving the pin floating is almost 'against my religion' as a long-time developer. And two pull-up/down internal resistors of 41.5 kΩ could perhaps struggle keeping the voltage in the middle in noisy environments. Adding two lower resistors, like two 4.7 kΩ, would only add to the power consumption. A simple decoupling capacitor solves everything in my book.

Best regards,

Flemming Monsrud

  • 0
    TI__Guru 62965 points

    Hi Flemming,

    The device pin is only an input from what we understand and detects if the voltage is at VCC, GND, or roughly at VCC/2.  The guidance in the datasheet is to force the voltage to the VCC/2 potential if leaving it floating is undesirable.  We understand the desire to include the capacitor to reduce EMI interference.  A concern with the capacitor is that it may not allow the node to charge to the desired VCC/2 voltage by the time the address is determined for the device on power-up.  

    We don't have the required information to determine what the internal timing of the address setting is on power-up but consider using a smaller ~33-68pF capacitor.  The smaller values should provide a similar effect and should also allow the node to charge quicker to the desired voltage.  Please test in you application and verify things work robustly before going to production.

  • +1 in reply to Collin Wells
    Prodigy 10 points

    Hi Collin,

    Thank you for your swift reply to my question.

    However, I don't believe the slave address is determined at power-up, as if it's 'sampled' by an internal reset signal. This would imply that the slave address cannot be changed 'on the fly'. I'm sure the slave address can be set to anything you like, following the state of the address pin, at any time while under power. There is nothing in the data sheet that indicates the slave address is sampled at power-up. The internal registers as well as the A/D conversion state machine, if you like, are set to initial default values and states by the internal power-on reset, but that's all. Please check with the designers of the chip if this is not in fact so.

    Best regards, Flemming Monsrud

  • 0 in reply to Flemming Monsrud
    TI__Guru 62965 points

    Hello Flemming,

    We've seen devices designed both ways where some determine address on power-up and others allow for on-the-fly changes as you've mentioned.  The designers aren't around to ask for this specific product.

    Coming back to how we got here the ~100pF capacitor on this pin should still provide the desired EMI reduction effects and should also charge up fast enough that any concerns, should they exist, should be mitigated.