Post Go back to editing

Noise caused by short circuit at the input end of AD7134

Category: Hardware
Product Number: CS-305674-W3X6V5

 When the input end of the AD7134 is connected to the ground, the output data is noisy with a fixed frequency,And the larger the input amplitude, the larger the noise amplitude。The AD7134 output data is as follows

                                                   

  The noise is all in multiples of 285hz,There is no such noise in my circuit or in my lab. Where is it coming from?

  • Hi  ,

    We will look into this. I will contact the product owner and get back to you.

     Regards,

    JC

  • Hi,

    Until you get the answer from  , just a comment.

    Looking at your graphs the harmonics of 285Hz seem to have an amplitude below 100nV, while the broadband noise have a peak-to-peak amplitude of about 50uV. Is it right? If so, it is 500 times higher, therefore I don't think the marked peaks have any relevance.

    Zoltan

  • Hi  ,

    As per datasheet, the noise measured is still within the noise performance defined in tables 9 to 16. Please let us know if you have other questions.

    Thanks,

    Janine

  • I have read the datasheet thoroughly.After reading the datasheet,i think if there is noise,the frequency should be 1kHz.But practical experience tells me that the frequency of noise is 285hz and its multiples

    .where this particular frequency of noise is coming from

  • Thank you for your answers,do you know where is the harmonics of 285Hz coming frown.

  • Hi

    the 1 kHz tone is not noise but a FFT of 1 kHz signal thats applied to its inputs, the caption mentions that.

    thanks 

  • Thank you for your answers,I want to know where is the harmonics of 285Hz coming frown.

  • Hi  ,

    I would like to ask what do you mean by your statement below, and how you are connecting the input?

    the larger the input amplitude, the larger the noise amplitude。

    What is your output data rate, performance mode, and what is the peak value of the noise that you measured?



    Thanks,

    Janine

  • I am very happy to receive your reply! Thank you very much!

    My output rate is 45kSPS, the AD7134 is in low power mode,SPI configurable operation,external CMOS clock of 48 MHz,master mode operation, DCLK is output, Wideband filter has a bandwidth of 0.433 Hz x ODR.The LDO takes power from a lead-acid battery, converts the voltage and then powers the circuit board, without using a switching power supply

    First,  answer your question "how are you connecting the input?" .Here's the thing:

    As shown in the figure, I took off R6,R12,C26,C33 and C46, and connected the 1,2 pad of C33 to the 2 pad of C26 and the 1 pad of C46 with a wire to realize short circuit at the input end.

    Then, after collecting the output data of AD7134, import it into matlab for FFT transformation. The results obtained are shown in the figure below. The frequency domain contains 285hz and its harmonic noise.

    Second, with regard to my statement of "the larger the input amplitude, the larger the noise amplitude", I am sorry that I did not say it in detail. Here I would like to elaborate:

    When I used PGA281 as the front-end driver module of AD7134, the output end of the signal generator was connected to the 14th pin of PGA281 (Filter1). The larger the amplitude of the input signal, the larger the amplitude of 285HZ and its harmonics. The circuit diagram is as follows: