• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 1298 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.

ADS1299: Bias Lead Off Detection At Power-up

S. Gaudreau
Intellectual 455 points
Part Number: ADS1299

Hi, 

I want to use this feature on the ADS1299 evaluation kit.

I can't get it to work constantly.

What are the steps and the register involved to activate this feature correctly ?

Also, what is the impendance levels to be considered not connected ?

Thanks !

  • 0
    TI__Guru 66446 points
    Hi Stephane,

    This feature only works if another source is used to define the common-mode voltage of the body. This would allow for the BIAS amplifier to remain powered-down while the current sink and comparator remain active. If the voltage at the BIASOUT node dips below the negative input threshold, the comparator should output a '1' in the CONFIG3 register.

    Best regards,
  • 0 in reply to Ryan Andrews
    Intellectual 455 points

    Hi,

    I want to use the function at power up only, as stated in the datasheet at page 32.  Then activate the bias circuit normally.

    What do you mean by "only works if another source is used to define the common-mode voltage of the body"

    We need to provide an external voltage source ? Can't we get it to work with the current source of the channels or other internal features ?

    Do you have a wiring example and register settings for this function to work properly ?

    Can we test this function with the evaluation board ?

    On my final product, I will have only 5 electrodes connected to the patient, channel 1 differential, channel 2 differential, and the bias feedback reference.

    Thanks

  • 0 in reply to S. Gaudreau
    TI__Guru 66446 points
    Hi Stephane,

    If the voltage on the BIAS electrode is floating, there's no way to ensure that the voltage will remain above the comparator threshold when connected. Admittedly, this was an oversight in the design.

    If you enable lead-off detection on the other connected input electrodes (IN1P/N and IN2P/N), the current should be enough to bias the body on it's own close enough to mid-supply such that the BIASOUT voltage remains above the comparator threshold. I can try a couple experiements on the bench tomorrow using the ADS1299EEGFE-PDK to confirm.


    Best regards,
  • 0 in reply to Ryan Andrews
    Intellectual 455 points

    Hi, 

    In my case, I only have 4 electrodes connected to the body, and 1 reference electrode.  Would this we enough ?


    I was thinking to add pull-up / pull-down resistors to the channels #1 and #2 electrodes ?


    Also, as you understand, I do not want to add a supplementary electrode to check for the bias electrode.


    Thanks 

    Stephane



  • 0 in reply to S. Gaudreau
    TI__Guru 66446 points

    Hi Stephane,

    I believe I found a more reliable way to do this in your application. However, as you'll read below, I had a couple issues with the EVM/GUI itself that may prevent you from testing this on the ADS1299EEGFE-PDK.

    What you can do is purposefully create an imbalance in the current sources by enabling only one of the two positive lead-off current sources (IN1P or IN2P). Leave the negative current sources on IN1N and IN2N disabled. Then, short the four input electrodes together.  Without BIAS connected, the 4 input pins will all be pulled high to AVDD. The comparators on IN1P and IN2P will trigger lead-off (i.e. LOFF_STATx = 1).

    Connect the BIAS electrode WITHOUT enabling the amplifier or BIAS lead-off detection. This should have no effect, except now the BIASOUT pin will also get pulled high to AVDD. As soon as you write to CONFIG3[1] to enable BIAS lead-off, the added current sink should pull the voltages down below the positive lead-off thresholds and the comparators on IN1P and IN2P will output 0. You can then disconnect the BIAS electrode, and now IN1P, IN2P, and BIAS will all trigger lead-off (BIASOUT gets pulled to AVSS).

    The only tricky thing here is that the aboslute voltages are a bit undetermined without a driving source. In my setup, I used channels 3 and 4. With all 5 electrodes shorted and only LOFF3P = BIAS_LOFF_SENS = 1, the common-mode voltage drifted to about -1.35 V with respect to ground (I'm using bipolar +/- 2.5 V supplies). This is far enough away from the negative lead-off theshold to not give a false-positive.

    The EVM does appear to have a couple drawbacks for evaluating the lead-off funtionality of the ADS1299:

    1. There seems to be some leakage path on the EVM such that only the larger current source magnitudes (6 uA and 24 uA) worked reliably. Even with all channel inputs floating, I could not get a single pin to be pulled to AVDD or AVSS for the 6 nA or 24 nA current options. The larger currents had no issue and the pins would be pulled well beyond the comparator threshold settings.
    2. The GUI does not appear to read the BIAS_STAT bit or the LOFF_STATx bits at all. After refreshing the register map tab, the lead-off status bits always read 0, even though the measured voltage on the pin is well beyond the comparator thresholds.

    You can load the following register configuration text files to copy the settings I ultimately tested with:

    Fullscreen ADS1299_BIAS_LOFF_SENS=0.txt Download 
    conf|config1	0x01	0x96
conf|config2	0x02	0xC0
conf|config3	0x03	0xE8
conf|loff	0x04	0x08
conf|ch1set	0x05	0xE1
conf|ch2set	0x06	0xE1
conf|ch3set	0x07	0x60
conf|ch4set	0x08	0x60
conf|ch5set	0x09	0xE1
conf|ch6set	0x0A	0xE1
conf|ch7set	0x0B	0xE1
conf|ch8set	0x0C	0xE1
conf|biassensp	0x0D	0x00
conf|biassensn	0x0E	0x00
conf|loffsensp	0x0F	0x04
conf|loffsensn	0x10	0x00
conf|loffflip	0x11	0x00
conf|gpio	0x14	0x00
conf|misc1	0x15	0x00
conf|misc2	0x16	0x00
conf|config4	0x17	0x00

    Fullscreen ADS1299_BIAS_LOFF_SENS=1.txt Download 
    conf|config1	0x01	0x96
conf|config2	0x02	0xC0
conf|config3	0x03	0xEA
conf|loff	0x04	0x08
conf|ch1set	0x05	0xE1
conf|ch2set	0x06	0xE1
conf|ch3set	0x07	0x60
conf|ch4set	0x08	0x60
conf|ch5set	0x09	0xE1
conf|ch6set	0x0A	0xE1
conf|ch7set	0x0B	0xE1
conf|ch8set	0x0C	0xE1
conf|biassensp	0x0D	0x00
conf|biassensn	0x0E	0x00
conf|loffsensp	0x0F	0x04
conf|loffsensn	0x10	0x00
conf|loffflip	0x11	0x00
conf|gpio	0x14	0x00
conf|misc1	0x15	0x00
conf|misc2	0x16	0x00
conf|config4	0x17	0x00

    Best regards,