impedance measurement using AD5933

Discussion in 'The Projects Forum' started by Kirti Sharma, Jun 16, 2016.

  1. Kirti Sharma

    Thread Starter New Member

    Jun 8, 2016
    10
    0
    I am using IC AD5933 to measure impedance. I communicate with the IC through Arduino Yun using I2C protocol. Using AD5933 I can measure impedances from frequency range 1 Hz to 100kHz. The system works fine for frequencies greater than 1 KHz using the internal clock and for frequencies lower than that, I have to provide an external clock. When I use an external crystal oscillator to provide an external clock of 1MHz, I get good results upto a frequency of 350Hz. Next, I wanted to provide a lower clock to measure frequencies lower than 300 Hz with good accuracy. I used 32.768 KHz external clock http://www.abracon.com/Oscillators/ASVK.pdf and tried to measure impedance at frequency of 100 Hz but then the IC does not produce any output signal for measuring the unknown impedance. Using an oscilloscope I found that AD5933 is not giving out a sine wave. I also tried providing an external clock using NE555P timer at clock frequency 58 kHz , 76kHz, but still no result. Using 1 MHz external clock, IC is able to generate a sine wave of 100 Hz but not when using the external clock at 32.768 KHz or 58 KHz.

    Could anyone please guide me where could the problem be? I do not have much experience with circuits , and I have no idea what could be going wrong.

    Any help is greatly appreciated.

    Thnakyou.
     
  2. dannyf

    Well-Known Member

    Sep 13, 2015
    1,809
    362
    What does the datasheet say?
     
  3. Kirti Sharma

    Thread Starter New Member

    Jun 8, 2016
    10
    0
    The datasheet just says this - "The system clock for the AD5933 can be provided in one of two ways. The user can provide a highly accurate and stable system clock at the external clock pin (MCLK). Alternatively, the AD5933 provides an internal clock with a typical frequency of 16.776 MHz by means of an on-chip oscillator."
     
  4. DickCappels

    Moderator

    Aug 21, 2008
    2,653
    632
    Since the range for MCLK goes all the way down to DC, it might be a problem with the quality of your clock signal. Is your MCLK signal clean with fast rise and fall times and switching nearly all the way from Vdd to Vss?
     
  5. Kirti Sharma

    Thread Starter New Member

    Jun 8, 2016
    10
    0
    yes, the signal from the clock is proper.
     
  6. Kirti Sharma

    Thread Starter New Member

    Jun 8, 2016
    10
    0
    upload_2016-6-16_13-54-53.png

    I also found this in Datasheet. Could this give a hint to anyone about where could be the problem?

    here is also a block diagram of IC
    upload_2016-6-16_13-57-55.png
     
  7. DickCappels

    Moderator

    Aug 21, 2008
    2,653
    632
    Are you telling the multiplexer (via I2C interface) to use the MCLK input?

    Sorry if this sounds like obvious questions but the symptom as described seems very simple. Next question would be about bypass capacitors and grounding...
     
  8. Kirti Sharma

    Thread Starter New Member

    Jun 8, 2016
    10
    0
    upload_2016-6-16_17-3-20.png

    Yes. The information whether to use external clock and the frequency of the external clock used in the arduino program is coded by me in arduino program which writes it in the registers of the IC AD5933 using I2C interface.

    And one update is that, By changing resistors of the oscillator circuit which uses NE555P timer, I generated a clock signal of 300 KHz. This signal when fed to MCLK at IC worked fine. The IC generated a sine wave signal as it is supposed to and I could measure impedances correctly. So, the problem starts when I reduce the external clock signal, for example to 80 KHz. What can be the reason?
    I use bypass capacitors at the power supply and use the same power supply and ground for the complete circuit.
     
Loading...