• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 1039 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Bias ECG inputs for single supply

ccent
Intellectual 340 points
Other Parts Discussed in Thread: ADS1292

I am designing a two electrode ECG using a single supply. Right now, I am using the ADS1292 for the ECG. I noticed that for single supply, most circuits are showing the inputs AC coupled and then biased to mid supply through large resistors. However, I am trying to design the ECG without RLD and without highpass filters on the input. Is this possible with single supply? 

I noticed that these ECG application notes from TI do not AC couple the inputs of the single supply ECG. Will this work:

http://www.ti.com/lit/an/slaa280a/slaa280a.pdf (figure 2)

http://www.ti.com/lit/an/slaa486a/slaa486a.pdf

http://www.ti.com/lit/ds/sbos168d/sbos168d.pdf (figure 9)

Most of these circuits bias the input using a resistor, but omit the ac coupling capacitor. Thank you!

  • 0
    TI__Mastermind 49176 points
    Hello Ccent,

    I am moving the post to the Precision Data converters forum that covers the ECG devices.
  • 0
    TI__Mastermind 38745 points
    Hey ccent,

    While it is possible to bias the input through large resistors without AC coupling the input, it may not be advisable. The body a very well may have an offset that is significantly different from the board mid-supply voltage. If the inputs are biased through resistors, current would flow from the body to the source on the board that is biasing the inputs. This current will cause an Ohmic voltage drop across any low pass filter resistors, contact impedance, or patient protection resistors that exists at the input. If the body's offset were to change for some reason, the amount of current flowing and thus the offset seen at the inputs would also change.

    AC coupling the inputs effectively reduces DC current flow to zero which prevents any Ohmic offset from being introduced at the inputs. If you are concerned about the high-pass filtering effect, you can simply make the cutoff for the filter as small as possible. As long as there is a series capacitor at the input, the filter will produce complete cancellation of any DC voltages. For instance, to achieve a cutoff frequency of roughly 0.03 Hz, you could chose a 1 uF capacitor and a 5 MOhm resistor (or a resistor divider whose parallel combination was equal to 5 MOhm).

    Regards,
    Brian Pisani
  • 0 in reply to Brian Pisani
    Intellectual 340 points

    Thanks for the detailed reply - very helpful. My main concern is that a very low cutoff high pass filter (ie. 0.03Hz) will result in a very long settling time. Therefore, anytime the user moves or adjust the electrodes, the filter will have to settle out for a number of seconds. I require a fast settling solution for my application. 

  • 0 in reply to ccent
    TI__Mastermind 38745 points
    Hey ccent,

    Your concern with the settling time is valid. After consulting with a colleague, another concern was brought to my attention. The size of the resistance used to bias the inputs without a capacitor must be large due to safety concerns. From my understanding, the standard limitation on single fault current on the body is 50 nA. When using a 5 V supply, that standard would require a 100 MOhm resistor. From that stems the issue in which the input leakage on the ADS1292 would cause a significant Ohmic voltage drop across that 100 MOhm resistor.

    This configuration is valid. You'd simply have to account for this large offset when choosing the PGA gain you'd like to use. In addition, a digital filter could remove the DC values with a much smaller settling time if that is desirable.

    Regards,
    Brian Pisani