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.

TMS320F28374D: adc continuos sampling min window timing multiple channel

Part Number: TMS320F28374D


HI everybody 

According to the datasheet,t the minimum time of Sample window duration is 75ns. I could find example project that uses this configuration (oversampled)  which is adc_soc_continuous_cpu01. The example project uses only one channel to use SOC continuously.

However, in my code, I used multiple channels and tried to use SOC continuously. I had to give at least 150ns of sample window duration (ACQPS) to get the right ADC reading. why  ?

is The min time 75ns can be only applied to a single ADC channel for continuous SOC (oversampled)?

Best regards

Carlo

  • Hi Carlo,

    Have a read through the introduction sections of the following app. note.  When you are scanning through multiple channels with different voltages, this tends to expose any settling issues in the input signal conditioning circuits. 

    Hitting the min S+H time at a high sample rate will generally require pretty good signal conditioning circuitry. You can also use the simulation methods in the app. note to check the settling for your particular signal conditioning circuits.  

  • HI Devin ,

    so the 75nS is OK   IF  you have a proper extranl ciruit as app note ? am I right ?

    thank you 

    BR

    Carlo

  • Hi Carlo,

    Yes, that is correct. 

    If you take a first-pass at the input design using the Analog Engineer's Calculator, you'll end up with an input circuit along the lines of ~40MHz BW, 300pF capacitor, and series R in the 10 to 100 ohms range (and you'd want to simulate to optimize the R value).  A less ideal circuit will result in slower settling and thus require a longer S+H time. 

    Alternately, if you are using a charge-sharing input design (ADC pin capacitor is large, nominally 14.5pF*4096*2 = ~120nF) then you will have a sample-rate limitation (but increasing the S+H will still help here, since it will increase the time between samples, which will then decrease the sample-rate since the samples are back-to-back).