AD5933 enlarges the frequency measurement problem

Hello, I am a graduate student. I would like to ask whether it is possible to expand the measurable frequency of AD5933 to 500k or 1000k. I have read the post mentioning Y-mixer. I am not very clear. I still want to Ask about the plan to expand the frequency。 Thank you. Tang

Parents
  • You can use the AD5933 as is to perform measurements up to 1 MHz, albeit at a diminishing accuracy due to:

    1. Built-in LPF with 100 KHz cut-off frequency on the receiving side

    2. Diminishing number of voltage steps in the DAC output at higher frequencies  (only 4 steps per period of the sine wave at 1 MHz on 16 MHz MCLK) on the synthesizer side

    Depending on your application, the former could be tolerable as the built-in LPF rolls over relatively slow with frequency, so there is some signal to work with even at 1 MHz.  The latter can be somewhat alleviated by overclocking the chip - running it at MCLK higher than recommended 16 MHz (I ran it on 36 MHz w/o any signs of functional issues or overheating).

  • Hello, thank you for your reply. I still have some questions about the built-in LPF with a cutoff frequency of 100kHz at the receiving end.

    1. Because the LPF is located inside the AD5933, am I not able to mount a higher performance LPF on the periphery of the AD5933 to improve accuracy.
    2. Can I only rebuild the components of AD5933 or choose another impedance chip to expand the frequency and measure accurately?
    3. Do you know some other impedance measurement chips with higher frequency range? If you know, can you tell me

    Looking forward to your answer, thank you

  • Because the LPF is located inside the AD5933, am I not able to mount a higher performance LPF on the periphery of the AD5933 to improve accuracy.

    Indeed, it is not possible to do anything externally to bypass the internal LPF. From observing it's behavior though, it appears to be a simple RC LPF that has -20 dB/decade slope, which at 1 MHz still passes about 10% of the signal to ADC, so that 10% of the ADC dynamic range can still be utilized for the measurements. Understandably it may or may not be adequate for your particular application.

    I do not know of any impedance-measuring chips with greater frequency range. Hopefully somebody form ADI could recommend one.   

Reply
  • Because the LPF is located inside the AD5933, am I not able to mount a higher performance LPF on the periphery of the AD5933 to improve accuracy.

    Indeed, it is not possible to do anything externally to bypass the internal LPF. From observing it's behavior though, it appears to be a simple RC LPF that has -20 dB/decade slope, which at 1 MHz still passes about 10% of the signal to ADC, so that 10% of the ADC dynamic range can still be utilized for the measurements. Understandably it may or may not be adequate for your particular application.

    I do not know of any impedance-measuring chips with greater frequency range. Hopefully somebody form ADI could recommend one.   

Children
No Data