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.

ADS1298: Amplifying above noise floor of ADS1298 and ADC input range are far apart

Part Number: ADS1298

I'm using this for EEG signals but tracing back to ECG signals to make my question clear.

I understand that the ADS1298 can be used without any pre amplifier and that ECG signals are in the mv Range. 

Noise floor for the ADS1298 is 7uV and the input range is defined based on the Vref and internal gain.

Say supply was single ended i.e 3.3V and 0V and Vref was 1.65V.  that would require the ECG signal to be amplified so it's centered around 1.65V which is like >1000 gain.

Doesn't that defeat the point of using a 24bit adc, it should be able to work with a low gain right?

Can anyone please help elucidate?

  • Hi,

    The app engineer was on a business travel. Shall get back to you around 7/28.

  • Hi,

    If using single end 3.3V for AVDD, the VREF is 2.5V according to datasheet page 12.

    ECG signal is measured differential, after differential measurement, the common mode voltage is cancelled out, and the differential signal doesn't need to sit at Vref.    Vref is the full/max scale that it can take the differential input signal.  So, not need to amplify to fill or saturate the full scale. in other words, roughly speaking, as long as the differential signal * gain is within the Vref, this is fine.

    Thanks  

  • Based on your explanation that should mean that I should be able to obtain an accurate eeg signal only using ads1298 with  no gain. As EEG is between 10 µV to 100 µV amplitude. Why do people apply gain or even use an instrumentation amplifier before hand?

  • Hi,

    I don't quite understand your question?

    Without INA gain, it's difficult for ADC to sample/quantize correctly&precisely as the signal will be so low.

    After the ADC quantizes/samples, the ADC codes will then need to be converted to voltage and divide by the gain to reflect the real input signal voltage, i.e. this is called input inferred/referred.

    Thanks.