• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 382 views
  • Users 0 members are here
Support feedback
Options
Options

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.

ADS114S08: problem operating the ADC114S08 with 4 ports according to the datasheet

Moshe Levy
Prodigy 230 points
Part Number: ADS114S08
Other Parts Discussed in Thread: ADS114S06

Hello,

We followed the configuration of Multiple Two-Wire RTDs (Figure 15) in ours design.

Also, as explained in the datasheet chapter 9, it is written:

"As described in the Functional Block Diagram section, these devices provide 13 (ADS114S08) or 7
(ADS114S06) analog inputs that are configurable as either single-ended inputs, differential inputs, or any
combination of the two. Many of the analog inputs have additional features as programmed by the user. The
analog inputs can be programmed to enable the following extended features:
• Two sensor excitation current sources: all analog input pins (and REFP1 and REFN1 on the ADS114S06)
• Sensor biasing voltage (VBIAS): pins AIN0, AIN1, AIN2, AIN3, AIN4, AIN5, AINCOM
• Four GPIO pins: AIN8, AIN9, AIN10, AIN11 (ADS114S08 only, the ADS114S06 has dedicated GPIOs)
• Sensor burn-out current sources: analog input pins selected for ADC input"

it is also written on page 28 that:

"The AVDD analog supply operates with bipolar supplies from ±1.5 V to ±2.625 V or with a unipolar supply from
2.7 V to 5.25 V. For unipolar-supply operation, use the VBIAS voltage to bias isolated (floating) sensors. The
digital supplies operate with unipolar supplies only. The DVDD digital power supply operates from 2.7 V to 3.6 V
and the IOVDD supply operates from DVDD to 5.25 V."

later on, on page 87 it is also written that:

"Signals where both the positive and negative inputs are always ≥ 0 V are called unipolar signals. These signals
can in general be measured with the ADS114S0x using a unipolar analog supply (AVSS = 0 V). As mentioned previously, the PGA must be bypassed in order to measure single-ended, unipolar signals when using a unipolar supply."

for some reason, we failed to operate 4 channels as per figure 15. Only after we did the change as attached KLA_ADC_4C_CHANGE it worked.

our questions are:

- why doesn't Figure 15 diagram work for 4 channels? 

- is there option to have 4 channels work per our attached KLA_ORIGINAL_DESIGN diagram and not having to do a rework as in the colored attached scheme?KLA_ADC_ORIGINAL DESIGN.pdfKLA_ADC_4CH_CHANGE.pdf

  • +1
    TI__Guru** 113765 points

    Hi Moshe,

    The only differences that I can see in your two schematics is the mux channel assignments.  The ADS114S08 allows for any combination of mux input channels to be used for measurement.  The only thing that I can think of is the input channels were not configured properly for the original schematic.  Can you provide the configuration settings used in the original circuit?

    Also remember that the excitation source is from the IDACs (and not VBIAS) which require that the internal reference be turned on even though it is not used for the ADC conversion.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 230 points

    Hello Bob,

    Thank you very much for the reply. 

    our configurations per channel are listed below:

    for channel #1:

    1. reset command - 0x06

    2. reset_n toggle '0' --> '1'

    3. WAKEUP command - 0x02

    4. SYOCAL command - 0x16

    5. SYGCAL command - 0x17

    6. SFOCAL command - 0x19

    7. Read ADS114 ID - 0x20

    8. INPMUX register(0x42) - 0x12

    9. PGA register(0x43) - 0x08

    10. DATARATE register(0x44) - 0x14

    11. REF register(0x45) - 0x10

    12. IDACMAG register(0x46) - 0x02

    13. IDACMUX register(0x47) - 0xD0

    14. VBIAS register(0x48) - 0x82

    15. SYS register(0x49) - 0x04

    for channel #2:

    same as Ch-1 above just with following changes:


         INPMUX register(0x42) - 0x45

         IDACMUX register(0x47) - 0xD3

         VBIAS register(0x48) - 0x90

    for channel #3:

    same as Ch-1 above just with following changes:

         INPMUX register(0x42) - 0x78

         IDACMUX register(0x47) - 0xD6

    for channel #4:

    same as Ch-1 above just with following changes:

         INPMUX register(0x42) - 0xAB

         IDACMUX register(0x47) - 0xD9


    Moshe

  • 0 in reply to Moshe Levy
    TI__Guru** 113765 points

    Hi Moshe,

    There appears to be some misunderstanding on the operation of the ADS114S08.  I will detail some specific settings, but first of all do not use both VBIAS at the same time as the IDACs as this will create an error.  Second, the internal reference must be turned on even when not used as the ADC reference to provide the correct output current from the IDACs.

    1. You show sending both the RESET command and the RESET pin control.  You should do one or the other and not both.  Also, following the RESET you need to wait for the device to complete the reset process before issuing any other commands.  The required delay is 4096 tCLK periods which is 1ms when using the internal oscillator clock (4.096MHz).
    2. There is no need to issue the WAKEUP command as the POWERDOWN has not been issued so the ADS114S08 is already awake.
    3. The SYOCAL and SYGCAL require specific input voltages to the ADC inputs when issued.  For example, the SYOCAL requires a 0 volt input to the ADC inputs and the SYGCAL requires a full-scale input.  I would highly recommend only issuing the SFOCAL (if used at all) which removes the internal offset to the ADC conversion.  The SFOCAL in the order as shown will erase the settings of the SYOCAL.  Also any calibration should be done after the PGA, data rate and reference selections have been made.  When changing the input channels I would recommend not using the calibration modes as this will not help that much with accuracy of the measurement given that the calibration is not being correctly done anyway.  I would suggest keeping the process very simple and not issuing the calibration commands for the 16-bit version (ADS114S08) of the device.
    4. You really have no reference as the internal reference is not enabled so that there can be an IDAC output.  The only current flowing is from the VBIAS output which will create an error in the measurement by current flowing through the input filters.  VBIAS is meant to bias floating voltage output sensors, such as a thermocouple, to have a correctly biased input voltage relative to the ADC input range.

    So the register configuration is incorrect primarily for two reasons.  The first is the use of VBIAS which should not be used for this type of circuit.  The second is the internal reference is not turned on which means there is no IDAC output current to create an excitation path for the sensor and the reference.

    I would recommend using the following procedure to see if you get the expected results:

    1. Toggle the RESET pin and wait at least 1ms before proceeding.
    2. INPMUX register (0x42) - 0x12
    3. PGA register (0x43) - 0x08
    4. DATARATE register (0x44) - 0x14
    5. REF register (0x45) - 0x12
    6. IDACMAG register (0x46) - 0x02
    7. IDACMUX register (0x47) - 0xD0
    8. VBIAS register (0x48) - 0x00
    9. SYS register (0x49) - 0x04

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 230 points

    Hi Bob,

    Thank you very much for the prompt reply. 
    I forward your answer to my employee who did the logic design. Hopefully, we will make some progress soon.

    one thing i might forgot and is not clear from the schematics. we do use Rref = 30K. not sure if it changes the answer, but still, worth mentioning. 

  • 0 in reply to Moshe Levy
    TI__Guru** 113765 points

    Hi Moshe,

    Thanks for the additional information.  An Rref of 30k Ohms and IDAC current of 50uA creates a reference voltage of 1.5V which safely allows the sensor to be in a valid input range for the ADC.  This does not change my earlier suggestions/response.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 230 points

    Hi Bob,

    I would like to inform you that the configuration you proposed works excellently. all ports are functional, and work as expected.

    I really appreciate it.
    BR

    ML