• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 46 subscribers
  • Views 104 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS8548: Application issue.

Reed
Expert 2250 points

Part Number: ADS8548

Hi team,

I want to increase the sampling rate of ADS8548, which is mainly used for the analog signal sampling of sine-cosine encoders, to achieve a sampling frequency of over 300K. Currently, I am encountering some issues, please help me with the answers:
1. When considering Shannon's theorem, what is the maximum sampling frequency that this ADC can support? Can it achieve 600K?
2. Can the ADS8548 read the values of each channel during the conversion process?
3. Currently, when the frequency is increased to around 188K, there is a certain abnormality in the waveform input to the ADC, as shown in the figure below:

Yellow represents signal input, and green represents the two ends of C347; there will be a section of values without sine wave,
After removing R386 and disconnecting it from the ADC, no such abnormal waveform appeared in the test signal input and operational amplifier output (pin 7 of U31).

 

  • 0
    TI__Expert 7390 points

    Hello Reed, 

    The ADS8548 has a maximum sample rate of 600kSPS, and it should be able to achieve that in parallel mode. In serial mode its max sample rate is only 450kSPS. 

    The ADC is able to sample both channels in each pair simultaneously during the acquisition phase of its cycle, ideal for sin/cos signal pairs. 

    The ADC's sampling frequency defines how often the ADC measures the input signal, not the input signal it is able to measure. As a rule of thumb (Nyquist Theorem) the ADC is only able to sample up to fs/2, so at 600kSPS the the theoretical max limit is 300kHz.  For this ideal the front-end bandwidth must be much larger than the input frequency.  It must be able to drive the ADC's sample and hold circuit to the point of the signal settling within 1/2LSB within the ADC's acquisition time. 

    I would suggest confirming the current configuration of the driver is able to drive the ADS sample and hold circuit at the 600kSPS. Our TIPL series could be helpful here if needed 

    Without knowing the frequency/div in you oscilloscope or the bandwidth capabilities of your driver it is difficult to say with certainty, but that abnormal waveform pause appears like it is lining up with the ADC's sample and hold phases and there might be some limitation in the ADC driver. 

    For input signals with frequency over 300kHz I would suggest an ADC with a higher sampling rate. The ADS9327 is a recently released device that can sample up to 5MSPS and could be a good candidate here. 

    Best regards, 

    Yolanda