• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 46 subscribers
  • Views 794 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.

ADS1112 conversion problem

Swathi Reddy
Prodigy 225 points

Other Parts Discussed in Thread: ADS1112

Hi..

I am using ADS1112 in 16bit, 15sps, continous mode to find out the output code with respect to change in temperature. I am trying to find output code across Ain0 and Ain3 with Ain0 reading from a fixed voltage source and Ain3 grounded. I see that , any input volage at Ain0 above reference voltage of 2.048v is giving me a standard output code of 7FFF. As per my understanding, output code corresponds to the difference at Ain0 and Ain3. Am i missing something..? Please help me through this.

Thanks,

Swathi 

  • 0
    TI__Guru** 113765 points

    Hi Swathi,

    Welcome to the forum! The measurement range is based on the reference value used.  For the ADS1112 the reference is 2.048V as you already mentioned.  As given in the electrical specifications table, the full-scale input range is +/-2.048/PGA.  This is the same as +/- Voltage Reference/PGA.  This corresponds to the difference of (VIN+) - (VIN-).  If AIN0 (VIN+) is 2.048V and AIN3 (VIN-) is ground you will get the full-scale positive output code of 0x7FFF.  If you exceed 2.048V you will just saturate the input at that point and always read positive full-scale.

    Given the same input voltages but you assign AIN0 to VIN- and AIN3 to VIN+, VIN- will be a greater value than VIN+ resulting in a negative difference and negative full-scale output code of 0x8000.

    So the output code does correspond to the difference of the two inputs up to the full-scale measurement capability of the device.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 225 points

    Hi Bob,

    Oh ok.. I get it now. Thanks a lot for clearing my doubt.

    Regards,

    Swathi

  • 0 in reply to Swathi Reddy
    Prodigy 225 points
    Hi!
    I have a query again. With ADS1112 ADC being used in 16 bit continuous mode, Whatever output digital code I receive has an offset error value of about 0.05 volts compared to the input voltage/analog value. 
    Is this because of some kind of tolerance that ADS has by itself or some error calculations ? 
    I am attaching a PDF file which plots a graph of the fixed input volatge being provided by potentiometer across Ain1 and Ain3 given to ground and Vout recorded from ADC (vout calculated using the formula (output_code * 2.048)/32768). 
    Please help m clarify this.
  • 0 in reply to Swathi Reddy
    TI__Guru** 113765 points

    Hi Swathi,

    The ADS1112 can have up to 8 mV of offset.  What you are seeing is linear, but larger than what I would expect.  If it is truly an offset, you can subtract out the equivalent number of codes for the result.  Before making this type of correction, make sure that the voltage source you are using to measure the pot voltage is a precision, calibrated meter.

    You will also need to account for bias currents of the ADS1112 relative to the value of the pot.  What this means is there are two current paths.  One through the pot and another through the ADC.  This is quite likely the problem.  How are you measuring the pot voltage?  Is it in circuit with the ADS1112 runing?  What is the value of the pot?  What kind of meter are you using to measure the pot voltage?

    Best regards.

  • 0 in reply to Bob Benjamin
    Prodigy 225 points

    Hi Bob..

    Pot voltage is measured using the multimeter. Yes we are using this setup when our ADS1112 is running. Pot  is of 20k ohms.

    Coming to the bias current issue we saw that   for the POT setup we have ( attached to this post), current measured across points A and B is around 0.7Micro amps. So when ADS is connected to this setup ie at input pin A1, its input impedance being very high of the order 10M ohms, the drop in the voltage across this should be very negligible ( of around 0.1p Amps). Hence I assume I shouldn't be seeing much of a change in my output code. Please correct me if I'm wrong.

    Thanks,

    Swathi1754.POT_setup_diagram.vsd

  • 0 in reply to Swathi Reddy
    TI__Guru** 113765 points

    Hi Swathi,

    I'm not sure how you are measuring your current.  You should see minimum of 3.3V/(27k+20k ohms) or about 70uA.  Also, I'm not sure where you are coming up with 10M ohms.  Please read the section on page 7 of the ADS1112 datasheet titled Input Impedance.  Any parallel impedance will lower the branch resistance.  The difference in resistance will change the expected outcome.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 225 points

    Hey Bob,

    Firstly thanks a lot for your timely response. Its helping me a lot in my project. Yes you were right. I had made some typo error. Also seems like our circuit design has some problem. Circuit has a thermistor in series with a 2k ohm resister connected to the ADC. We observe that the 2k resistor is pulling down the resistance value of the thermistor and hence my output code differs. I also came across a TI application note which suggests me the use of current source to overcome this( Attached along with this post).  Attached also is our design circuit used in our application. Can you please go through it and tell me if my design is right and also suggest an appropriate application/design note that can be used for this particular ADC ?

    Thanks and Regards,

    Swathi

    8686.design_circuit.docx

    0167.TI_Current_source_info.pdf

  • 0 in reply to Swathi Reddy
    TI__Guru** 113765 points

    Hi Swathi,

    Unfortunately I don't have an app note specific to this part.  The constant current source is the best way to go when reading the voltage directly from the thermistor.

    If R18 is a stable precision resistor, you could connect the inputs you are not using to measure the voltage across R18.  Based on this voltage you should be able to calculate the value of the thermistor.

    Best regards,

    Bob B