• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 401 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.

ADS1282: Review of ADC selection

Marcel H
Prodigy 10 points
Part Number: ADS1282
Other Parts Discussed in Thread: ADS1256,

Dear TI team,

I'm working on a new ADC board for a geomagnetic sensor with an pseudo-differential output of +/- 5 V at 20 Hz bandwidth. In my application I have a very high noise floor above 20 Hz, so proper noise filtering is a big issue for me. Right now I already have a board working with an instrumentation amplifier for differential to single ended conversion, a 5th order Sallen Key lowpass with fc = 20Hz, an ADS1256 and a FIR lowpass on the microcontroller. With this setup I'm achieving a sufficient noise performance and accuracy, but I'm having issues with RFI rectification and temperature drift.

Therefore, I'm aiming towards a solution with minimum analog signal processing. Because I have limited processing capacity on the microcontroller, I'd like most of the digital filtering to happen on the ADC, so that I can acquire the data with a low output data rate and can minimize FIR filtering on the microcontroller.

So far, I believe that ADS1282 would be a good choice for me, but I would appreciate your feedback on my decision, as I might have overseen something. My main reason for selection ADS1282 is that it has the FIR filter with -140 dB above the decimation filter's Nyquist frequency. Therefore I believe that I don't need to care about decimation aliasing, but only about modulation aliasing. To cope with the latter one I can achieve -130 dB at 500 kHz with a 2nd order RC low pass filter easily (with f_C = 4 MHz).

For achieving full ADC input range scaling in respect of the limited PGA output range, I'm planning to attentuate the sensor signal with a voltage divider to +/- 1 V, use PGA = 2 and a 4.096 V reference voltage.

I appreciate your valuable feedback.

  • 0
    TI__Mastermind 44680 points
    Hi Marcel,

    Welcome to the TI E2E forums!

    I understand the concern to keep the number of analog components to a minimum (for many different reasons including drift, total error, PCB area, and cost), and also to reduce the processing load on your MCU. I agree that the FIR filter of the ADS1282 may help to reduce some of the post-processing required by your MCU.

    You are correct about the ADS1282's input range limitations and your proposed voltage divider implementation would meet the ADC's input range requirements. However, you will want to pay very close attention to the resistor temperature coefficients and avoid using resistors much larger than 5-10 kOhms in the voltage divider, as resistor thermal noise may start to become apparent in your conversion results. However, since your input signal is pseudo differential (and not fully differential), I'm not aware of many alternate implementations... Many of our Precision ADC's allow for +/-5V differential inputs; however, in most cases this would need to be a fully differential signal, and even then not many other ADCs offer this kind of FIR filter (SINC filter type is much more common).

    Your distinction between "decimation" aliasing and "modulation" aliasing is not something that I am familiar with. Any additional details on what you mean by each of those would be much appreciated. However, I will say to look out for the ADS1282's digital filter passbands that repeat at integer multiples of the modulator sampling frequency (fMOD or CLK /4). Therefore, it is important to make sure that your input RC filter provides sufficient attenuation around fMOD.