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.

ADS1299EEG-FE PDK problem

Other Parts Discussed in Thread: ADS1299

I'm working with the EVM ADS1299EEG-FE PDK and I have identified in a test conducted something strange.

The test was performed as follows:

The parameters was sets for single-ended inputs, with the pins (5-6) JP25 shorted and with the bit SRB1 set in the MISC1.

Using a signal generator, a (FIXED) tension of 10mVpp was applied and a (VARIABLE) frequency from 10Hz to 100Hz, ranging from 10 to 10 Hz in the positive pin of the channel 1.

The result obtained in this test was that, while the frequency INCREASES, the tension that should remain fixed, DECREASED.

 

How can I solve this considering that frequencies around 10 to 100Hz are useful in an EEG test?

This feature is characteristic of this EVM or I need to do any special configuration to hardware or software?

  • Hello user4310770,

    As with all delta sigma ADCs, the ADS1299 possesses an on-chip decimation filter which acts to low-pass filter the modulator output based on the output datarate. Details on the filter can be found starting on page 21 of the datasheet. The filter has a sinc magnitude response with zeros at multiples of the datarate. What datarate are you using for this test? At the lowest datarate of 250 SPS, you could see some noticeable attenuation at 100 Hz. To widen the band of the filter, you could simply increase the datarate. Do you believe this could account for the attenuation you are seeing?

    Regards,
    Brian Pisani
  • Hello Brian,

    Thank you very much for your help, I was using the datarate of 250SPS, but when I changed it for a larger datarate, according to your explanation, the resulting signal has improved a lot.

    I'm currently trying to assess the functioning of ADS1299EEG-FE in an acquisition with electrodes plugged directly into the scalp.
    I'm using single-ended inputs, with the pins (5-6) JP25 shorted and with the bit SRB1 set in the Misc1 to accomplish a common negative reference.
    The question is where to connect the bias electrode, and what configuration I need to perform on the software on the (Bias Control Register) and (Configuration Register 3 - CONFIG 3)?

    Can you explain me what means the terms BIASIN,  BIASINV, BIASOUT and BIASREF

     

    Regards,

    Roger Gabriel

  • Hello Roger,

    I am legally required to tell you that the ADS1299EEG-FE is to be used for evaluation purposes in a lab environment only. Therefore I cannot give any advice about body testing.

    The Bias Amplifier on the ADS1299 is equivalent to what you might be familiar with as the "right-leg drive" or "patient drive" which is common in ECG systems. Its purpose it to inject a known voltage onto the body to establish a valid common mode for the ADC to convert the desired voltages.

    The Bias Amplifier on the ADS1299 can be modeled at a basic level like any op-amp you may have studied in school. A diagram showing its circuit-level functionality is located on page 54 of the datasheet. The BIASIN pin gives the ability to route the patient bias signal to the channel for measurement. It can be multiplexed to the channel using in the CHnSET register. The BIASINV is the inverting input to the bias amplifier. The BIASOUT pin is connected to the output of the amplifier, and the BIASREF is connected to the noninverting input under some circumstances. The BIASREF provides the reference voltage for the amplifier to drive the bias signal to. The restart default, though, is for the noninverting input of the bias drive amplifier to be connected internally to the midsupply voltage.

    To use the bias drive functionality on the ADS1299, you can configure it using the CONFIG3, BIAS_SENSP, and BIAS_SENSN registers. In the CONFIG3 register, you can route the BIASIN pin to the channel, change how the BIASREF signal is generated, and power the amplifier on and off. The BIAS_SENSP and BIAS_SENSN registers control which inputs are fed to the bias amplifier as feedback for its common-mode canceling functionality, though they are not required for the amplifier to work as a source for the buffered, midsupply voltage. Details on that functionality, can be found in the datasheet for the ADS1299.

    Regards,
    Brian Pisani
  • Hello Brian,

    I'm currently working on a college work analyzing the ADS1299 functionality with de ADS1299EEg-FE PDK , I am intending to carry out a test where I need to synchronize the software with other system and I think that the source code to the LabView software would be helpful. So could you email me the password for the LabView source ( ftp.ti.com/.../ADS1299LV.zip ) and the ADS1299 firmware ( ftp.ti.com/.../ADS1299FW.zip ).

    Thanks for all

    Roger Gabriel
  • Hey Roger,

    The information has been sent as a private message.

    Brian Pisani
  • Just a follow up question,

    Is a bias necessary to get a recognizable signal? Or is a reading from an electrode taken with respect to the reference electrode generally sufficient in just getting a basic signal that's distinguishable from noise?

    I'm having trouble getting anything more than noise, even after filtering out 60Hz

    Thank you!

  • Hey Meagan,

    The signals need to be maintained at a near constant common mode voltage with respect to each other as well as within the power supply range of the device. Even if a reference electrode is used, its voltage will also be floating with respect to the board voltages. One alternative to using a bias electrode is to AC couple the inputs. If that strategy is used, only the signals that you intend to measure actually make it to the device and any near-DC common mode drift gets blocked. The common mode voltage of the inputs will then get set on the board to a known level.

    Regards,
    Brian Pisani
  • Thank you!!!

    Meagan

  • please send differential mode circuit diagram . particularly give the positive and negative electrode input connection details


    bala