• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 1258 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.

ADS1291: Best signal quality for a 1-lead EKG system - Common mode noise and DRL-electrodes

Erik.
Prodigy 30 points
Part Number: ADS1291
Other Parts Discussed in Thread: ADS1294, ADS1293

We are designing a single lead EKG recorder with sensors about 7.5 cm (3 inches) apart. As it is a small system and wireless (nothing is connected to the ground) it is my understanding that the common mode noise/power-line noise would be quite small.

However, we are contemplating the ADS1294 instead of the ADS1291 as the 1294/6/8 seem to be superior in some ways (except for power consumption) see table:                          

 

ADS1291

ADS1294

Input-Referred Noise

8 μVPP

4 μVPP (150 Hz BW, G = 6)

Data Rate

125 SPS to 8 kSPS

250 SPS to 32 kSPS

CMRR

–105 dB

–115 dB

Power

335 μW/channel

0.75 mW/channel

 But having a closer look at the data rate shows (ADS1291 on top ADS1294 on bottom):

I believe 8kSPS is more than enough for all software analysis/filtering of a single lead EKG, therefore aren´t the data rates actually the same under these conditions?

So the question is: are the Input-Referred Noise and CMRR differences going to be noticeable? Or am I wrong regarding the data rate?

Secondly: Would adding a DRL (driven right leg) electrode help reduce noise even though that electrode would be placed near the other two and not on the right leg, i.e. far away?

Finally: Are there any data converters other than the ADS129x-series I should consider?

Thank you in advance

  • 0
    TI__Guru 57446 points
    Hello Erik,

    Welcome to our forum!

    The ADS1291 probably makes the most sense for a single-lead, battery-powered application. Power line noise can still couple into the system from lighting even if you're using batter power, but the inherent CMRR of the device and the use of the RLD can help mitigate that. Also, depending on the signal bandwidth you need, you might be able to low-pass filter 50 Hz / 60 Hz noise digitally. ECG bandwidth ranges from 0.5 Hz up to 150 Hz for clinical applications, but a lot of consumer applications only require up to 40 Hz.

    It's hard to say how effective the RLD is as you place the electrode closer to the other sensing electrodes. Ideally, you want a high-impedance between the RLD electrode and the others for the most effective common-mode cancellation.

    The ADS1293 would be the only other device to consider that you have not mentioned yet. This is a 3-channel device that is also well suited for low-power ECG applications!

    Best Regards,
  • 0 in reply to Ryan Andrews
    Prodigy 30 points

    Thank you very much for your answer!

    Yes I will work with digital filtering as well and I will try to get the full "clinical" ekg-spectrum.

    Ok is there any advantage of the ADS1293 over the 1291 if I am only using one channel? I couldn't find anything.

    Yes I understand regarding trying to get a high impedance between the RLD and the other electrodes and maybe I am thinking backwards but will the RLD actually be able to "sense" the common mode noise (even in theory) if it is placed exactly in the middle of the other two?  Again, maybe I'm misunderstanding the "sensing".

    Thank you in advance

  • 0 in reply to Erik.
    TI__Guru 57446 points

    Hi Erik,

    The common-mode sensing is done by the measurement electrodes themselves, not the RLD. Please refer to Figure 52. The outputs of the PGA stage are individually selected and connected to a summing junction at the RLD amplifier inverting input. The RLD amplifier is driving an inverted common-mode signal back to the body at the RLDOUT pin.

    Best Regards,

  • 0 in reply to Ryan Andrews
    Prodigy 30 points
    Thank you Ryan,

    I see, I thought the RLD was also involved in the sensing but the figure makes it clear. Given that the RLD will be suboptimal due to low impedance between the electrodes I think that a software solution will be more effective, but would it hurt using both software filtering and RLD? Is there risk of extra noise, severly increased power consumption etc? In theory I guess it would help a little but maybe it's best to keep it simple in practice.

    The pricing is almost the same, so it wouldn't make much economical diffrence but is the ADS1293 superior in any way except for the no of outputs?

    Thank you in advance,
  • 0 in reply to Erik.
    TI__Guru 57446 points
    Hi Erik,

    I don't think using the RLD anyway would hurt and I have seen it done before where the RLD electrode was placed closer to the other electrodes instead of on the right leg itself.

    The ADS1293 can handle more inputs, has more flexibility in common-mode, RLD, and WCT derivation, and also supports higher data rates for pacemaker detection. If you need these features, it does offer an attractive power consumption spec, but the noise performance is about 2x worse than the ADS1294. I think the ADS1291 sounds like the right device for your application.

    Best Regards,