• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 47 subscribers
  • Views 383 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.

ADC128S102: Counts lower than what it reads on power up

Randhir Kalsi
Expert 4545 points

Part Number: ADC128S102
Other Parts Discussed in Thread: LM4040

Hello team,

Hope you are doing well. One of the customer is evaluating this device for their application.

They are noticing that no matter what voltage is applied to the inputs on power up, after 2 minutes it is 10-15 counts lower than what it reads on power up.

It is doing this everywhere that they use the ADC128S102.

Please advise.

  • 0
    TI__Expert 5575 points

    Hi Randhir

    Can you please let me know more details about customer application

    1. Are they using any front end driver to drive ADC input

    2. What are the RC filter values for ADC input filter.

    3. What is the sampling speed ADC is running at. If they reduce sampling speed does this improve measurement accuracy.

    If you can share schematic snapshot of ADC section that should help. I need to check what biasing components/ ADc driver they areusing and at what sampling speed ADC is running

    Thanks & regards

    Abhijeet

  • 0 in reply to Abhijeet Godbole
    TI__Expert 4545 points

    Hey Abhijeet,

    Thank you for the response. See below comments from the customer

    1)      I’m driving it with an op-amp in unity gain configuration.  The output is directly connected to ADC input

    2)      I have a 110kohm/0.01uF on input to the op-amp

    3)      I’m using a 6-12MHz SPI clock (I have tried 6 and 12 and it is the same so I just have been using 6).  I take about 30k reads per second 

    I got a better Vref and the issue seems to have disappeared.  I was trying to use a LM4040, and switched to a different part.  Seems the problem is solved.