• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 30 subscribers
  • Views 290 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.

PGA970: sampling electrical zero of LVDT probe

Xavier
Prodigy 100 points
Part Number: PGA970

HI,

Using PGA970EVM board for 2.5mm LVDT probe.

* 5 kHZ differential mode

* sampling 24bit ADC data and filterd using FIR filter.

* using PH1_PHASE1 and PHASE2 registers resolved the probe polarity.

* ADC data which get is  positive electrical - 0 to 0.3v and  negative electrical - 0v to 0.27V

* the problem is the near zero volt is not filtered properly and data is fluctuate at zero.

* when doing calibration +/- 1 microns oscillating at electrical zero

Thanks and Regards,

Xavier.

  • 0
    TI__Mastermind 31655 points

    Hi Xavier,

    Even with filtering there will be some amount of noise. How large are the changes in the ADC codes near zero in your application?

    Regards,

    Scott

  • 0 in reply to Scott Cummins
    Prodigy 100 points

    Hi Scott,

    * The filter  digital value ranges from  +90000 to -75000 approx,

    * The resolution for 1 microns is  20 digital count.

    * The near fluctuate digital values around +/-10 digital count.

    * After fir filter i averaged the value so that value is stable.

    * but near zero value the digital data is not linear.

  • 0 in reply to Xavier
    TI__Mastermind 31655 points

    Hi Xavier,

    Does this fluctuation only happen near zero? Magnetic or mechanical interference could also impact the measurement, especially when the signals are quite low. Additionally if you are using analog lowpass filters on the inputs, any mismatch in the cutoff frequencies due to component variations could potentially cause noise.

    Regards,

    Scott

  • 0 in reply to Scott Cummins
    Prodigy 100 points

    Hi Scott,

         * No, full cycle its happening.

        * yes, i gone through it and finally the oscillation is surpassed ,

        * but problem now is OFFSET ZERO,

       * In observation  the digital value  ranges +90000 to +21 to +75000.

        * after phase validation +90000 to +21 to -75000.

        * since the cycle  at electrical zero is +21 need to offset then only the microns calibration will be linear.

       *  In this case by subtracting the offset zero the problem is solved .

        * but different LVDT probe the electrical zero will change then  how to solve it...

    Thanks & Regards,

    Xavier

  • 0 in reply to Xavier
    TI__Mastermind 31655 points

    Hello Xavier,

    Typically some amount of calibration will be necessary to get the maximum performance from a specific sensor. The zero point will most likely need to be calibrated for each sensor unit to account for the offset you have described.


    Regards,

    Scott

  • 0 in reply to Scott Cummins
    Prodigy 100 points

    Hi Scott,

    Thanks for your suggestion and ideas.

    * Now facing one more problem

    * usually while using LVDT probe based on resolution requirement will adjust the output gain.

    * example when using +/-2.5mm if needed 1 micron resolution will adjust  to +30000 to -30000 so that       scale will measure +/-2.5mm.

    * when resolution need to 0.1 micron will adjust to above limit but the scale will be  +/-0.5mm.

    * how do to do above condition in PGA970EVM.

    Regards,

    Xavier

  • 0 in reply to Xavier
    TI__Mastermind 31655 points

    Hi Xavier,

    I'm not quite sure I understand the question. Is this used with 2 different LVDT sensors in different applications? if it is with a single sensor that could be used in 2 different applications with different maximum ranges, you may need to adjust the secondary input gains to use as much of the ADC input range as possible to maximize the resolution. There is no automatic gain calculation integrated into the PGA970 however. This will have to be manually changed for each application.

    Regards,

    Scott