• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 33 subscribers
  • Views 159 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.

TIC12400: A to D reading is off a bit

Diane Eaker
Prodigy 75 points

Part Number: TIC12400

Hello,

First of all, the TIC12400 is a very nice part and has a lot of nice features.  Question about the a to d.

I'm monitoring some switches and then using some of the TIC12400 inputs to measure analog.  I understand the part has a 6V a to d and is 10 bit, so each bit is 6/1024 = 0.005859 Volts.  When I read the raw value in the a to d register and convert it based on 6/1024, my values are off just a bit.  For example:

2V converts to 2.13V and 1.9V converts to 2.03V.  The voltage I am inputting comes from a nice precise power supply and I confirm the value using my Fluke 289 so I believe in the voltage value going in.  BTW, for this test, I was using pin 10 which is IN21

I am trying to think about ways to handle this error.  Thoughts from your side.

  • 0
    TI__Mastermind 32760 points

    Hello Diane,

    What is your set up? Are you using any of the available wetting currents?

    Regards

    Josh

  • 0 in reply to Joshua Salinas
    Prodigy 75 points

    Josh,

    I am not using wetting currents.  I don't understand enough about those to use them at this time unless I should be using them.

  • 0 in reply to Diane Eaker
    TI__Mastermind 32760 points

    Hello Diane,

    If you are not using the wetting currents then by default they will be configured to 0mA. When configured at this state, you will see a small amount of leakage current coming from the MUX circuit (+/- 110uA). This current will mix with any external loading you have and will cause your ADC code to be slightly off (which is happening here)

    There are a couple of ways to solve this

    1) Place a filter (decoupling) capacitor on each input pin to absorb and stabilize some leakage current. Place a discharge resistor to GND in parallel with this cap to discharge the capacitor between polling cycles

    2) Collect a sweep test across 0-6V ADC range to determine the offset voltage between the external measure voltage on the input pin and raw ADC code. Then mathematically subtract to calibrate the ADC results

    Regards,

    Josh

  • 0 in reply to Joshua Salinas
    Prodigy 75 points

    Hi Josh,

    A couple of things.  You said when wetting currents are configured at 0 mA I will see a small amount of leakage current.  Are you saying that if I configure them to something other than 0mA, then perhaps the problem goes away? I can try that, but I just wanted to make sure I understood your comment.

    Next, you said place a decoupling cap on the input pin.  My design does have a 0.01uF cap there already in the same position as you have the 15nF above in your diagram.  So I don't think 10nF vs 15nF should make too much of a difference.  I believe I did upload my schematic in the previous post if you want to check that.

    The second thing you said about collecting a sweep from 0 to 6V and that has me interested.  Let me make sure I understand this.  If I collect this sweep on one board and use this as a correction, will this apply to all boards with this part on it.  If so, then we are assuming there will not variation from part to part.  Is that how you see it?  I just want to make sure I understand.  I can certainly do this if it will then apply to all our boards with this part.

  • 0 in reply to Diane Eaker
    TI__Mastermind 32760 points

    Hello Diane,

    We are looking into this and will respond shortly.

    Regards,

    Josh

  • 0 in reply to Diane Eaker
    TI__Mastermind 32760 points

    Hello Diane,

    1) The leakage current will always be there, regardless in how you configure the wetting current. Since the other wetting currents are 1mA and higher, the 110uA leakage current won't be noticeable. This, however, could be a good way to "eliminate" the leakage current.

    2) I don't see your schematic. Could you upload it again?

    3) If you got all your parts from the same batch, then this would be correct. I would also recommend capturing this offset voltage in the same environment as your actual application. Same board, switches, temperature, etc.

    Regards,

    Josh