• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 46 subscribers
  • Views 366 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.

ADS1292R: ads1292r ecg signal

user4521484
Prodigy 100 points
Part Number: ADS1292R

Problem Description:
Measurement scenario:
1. The back of the left hand (where the watch is worn) contacts one electrode, and the finger of the right hand contacts another electrode;

During the measurement with ads1292r, the following phenomena will appear:
1. When the measurement is just started, the waveform will be pulled up or down, and it can return to the baseline for normal display after a long time;
2. sometimes, a rectangular wave similar to the truncated top appears all the time
3. sometimes, there will be a very rough waveform near the baseline, and no ECG signal can be seen
4. sometimes, a very clean waveform will appear, which is an inverted triangle signal, and no ECG signal can be seen

  • 0
    TI__Mastermind 47625 points

    Hi,

    The first issue is typically caused by a changing contact impedance and as you mentioned, the measurement settles out. If you have lead-off detection enabled, try disabling this feature. As for the rest of the issues, without pictures it is difficult to say where this originates from. Software changes and digital filtering are necessary to account for these different phenomena and create a reliable measurement system. 

  • 0 in reply to Alexander Smith
    Prodigy 100 points

    Hi,

    Thanks for your reply!

    We're trying to turn off the lead-off detection feature and see if this issue will disappear.

    Our application is a smart watch, two electrodes (stainless steel material) are for left wrist and right hand finger.
    We always experienced this kind of phenomena: in the beginning (maybe 10~20seconds long) while we just put on the right hand finger, the ECG signal is always poor,
    but it may suddenly become very good after a while. The tester didn't change the pressing strength. Please refer to the two pictures of ECG signal.

  • 0 in reply to user4521484
    TI__Mastermind 47625 points

    Hi,

    Did turning lead-off detection off help?

    I'm not entirely sure. Is it possible that it is a display issue? It is not clear if the first pictures are simply "zoomed out" versions of the second picture? 

    You can look at the codes being sent from the ADC to the MCU to confirm if their magnitude is increasing after 10-20s? This will help narrow down where the issue is. 

  • 0 in reply to Alexander Smith
    Prodigy 100 points

    Hi，

    We turned off detection, but it didn't seem to help.Turn off the detection configuration as follows. Please help to confirm whether the configuration is correct

    In addition, we confirmed that the MCU was getting larger data from the ads1292r rather than a display magnification

    unsigned char ADS1x9xR_Register_Settings[15] = {
    //Device ID read Ony
    0x00,
    //CONFIG1
    0x02,
    //CONFIG2
    0xA0,
    //LOFF
    0xF0,
    //CH1SET (PGA gain = 6)
    0x60, //0x00,
    //CH2SET (PGA gain = 6)
    0x60, //0x00,
    //RLD_SENS (default)
    0x2C,
    //LOFF_SENS (default)
    0x0F,
    //LOFF_STAT
    0x00,
    //RESP1
    0x02,
    //RESP2
    0x03,
    //GPIO
    0x0C
    };

  • 0 in reply to user4521484
    TI__Mastermind 47625 points

    Hi,

    Just confirming: CHXSET of 0x60 will set the gain to 12, not 6 as indicated. 0x00 will set the gain to 6. 

    Otherwise, the register settings seem fine. 

    More interesting is that the ADS1292R is sending larger data values to account for this. Can you measure at the input pins of the ADS1292R as well to confirm if the data is increasing in value?

  • 0 in reply to Alexander Smith
    Prodigy 100 points

    Hi

    Thanks for your reply.

    During the test,we found that  the lead off function cannot be completely turned off when  CONFIG2 is set to 0xA0 and the LOFF_SENS is set to 0x0F .You need to set CONFIG2 to 0xA0, LOFF_SENS to 0x00 .After this configuration, the signal quality mutation will no longer occur.

    Currently we are still experiencing a poor ECG signals problem with ADS1292R. We are not sure if it is caused by the electrode materials (stainless steel 316L) or the thickness of electrodes. The thickness of left hand electrode is 1.2mm, the thickness of right hand finger is 1.1mm. Sometimes we could see a good ECG signal, sometimes not.

    We did such an experiment: We replaced above mentioned electrodes with two thin metal pieces (thickness 0.2mm, material stainless steel 304), the ECG signals became much better. Does the thickness heavily affects the signal quality of ECG? According to our understanding, material stainless steel 316L is better than 304, how come the signal from thin piece stainless steel 304 is better than the one from thick 316L?

    Looking forward to your great support, thanks!

  • 0 in reply to user4521484
    TI__Mastermind 47625 points

    Hi,

    Very interesting. I am not sure why Lead-off would cause this result. I will have to think on it. 

    Unfortunately I don't have any insight to offer on the thickness of the electrode material. However, it could be that there is noise in the system/environment and it is easier to couple onto the plates with a larger thickness. If you can provide pictures of the comparison - it may help me identify possible causes.