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.

Design of non contact ECG sensor

Other Parts Discussed in Thread: INA116, LMP7702, LMC6442, LPV521, LMP2012, TL082

I would really appreciated if i get some help on the below topics

I am designing a non contact ECG sensor to display ECG signals in an oscilloscope. The basic design would be having two electrodes to pick up the signals and then some signal processing scheme to filter off noise and probably a DRL circuit as well. Could you suggest suitable componets availabe for this & any  design recommandation  to this work?

  • Hello Eldos,

    I am looking into this and will get back with you.

  • Hello Eldos,

    Non-contact sensors have extremely high impedance.  Therefore you will want to look at devices with extremely low input bias current.  From what I have read in the journal literature the INA116 instrumentation amplifier and the LMP7702 have been used in non-contact sensor applications.  A literature search in the IEEE biomedical engineering field will yield circuits that have been used to buffer & filter signals from non-contact sensors as well as DRL.  I hope this helps!

  • Hi Pete and Eldos,

    Essentially I'm trying to do the same thing as Eldos, except that I am making an EMG sensor.

    I did actually find the usage of INA116 and LMP7702 that you mentioned, previously in literature. Since the focus is more on having a very low input bias current, and more importantly, a very low current noise, I am interested in three op-amps for the buffers before the Instrumentation amplifier stage.

    LMC6442 This seems to have ultralow bias current and low current noise. The only thing I don't like is the SOIC package. If it were anything smaller, even an MSOP, It'd have been perfect.

    LPV521 As this is in SOT package, this seems to be a better option in terms of size. However, in terms of current noise (in) and input bias current, I think it is worse than 6442.

    Finally, LMP2012 seems a bit of a mystery to me. I can't find the spec for input bias current (Though it states an input current of -3pA and I have no idea what to make of that). Also it states that it has no 1/f noise and the current noise is not listed. This seems like a very stupid question to ask, but is the LMP2012 constructed in such a way that there is no current noise? And I am not sure about the input bias current. Else, it seems to be the best possible solution in the industry for this application. Additionally, There is obviously a need for high input impedance. Yet none of the datasheet seem to state that specification.

    I would appreciate your stance on this.

    Thank you.

  • Hai Eldos , Pete and Usama,

    I am Wali, a Doctor  student in Malaysia. I want to make Non-contact ECG sensor for my research.

    From Literature i found that it can be done with following components:

    Start from input side-

    1. De-electrode (plastic for example)should be pasted on antenna.

    2. Aluminium antenna of diameter 5 cm.

    3. TL082 Trans impedance amplifier (if you make by hand) or Integrated Trans impedance amplifier circuit  (can be found in Alibaba.com of china, 5 usd)

    4. Buffer (basically a simple voltage follower)

    5. 3 filters:  first order active low pass filter, 50 Hz notch filter  and an active high pass filter

    then output comes...

    so far, not yet determined parameter level of spec's, that is actually very important. this ECG sensor supposed to sense from 40 cm distance.

    I wish you guys tell me if you know non contact ECG better than that !!!!

    cheers,

    Wali

  • Hi, Eldos,  I just started working on the non contact ECG sensor as well, and saw your post a year ago.  Could you please point me to the right direction, i.e., literature, circuit schematics, etc?  Thanks a lot!  

  • Hello all,

    I have the same problem about LMP2012: no input bias current specified or maybe I can't see it. Does any of you found out the input bias current? 

    Thanks,

    Ion