ADAS1000 TESTTONE

Dear sir or madam,

I configured my ADAS1000 according to Example 4 in datasheet (activated bit to send header only when data is busy).I have no skipped frames nor any faults exist in HEADER register.

As a first question, I would like to kindly ask you if you could help me resolve the fact why do I have such a deformed amplitude (enclosed snapshot) of 150Hz  sinusoidal frequency (it applies to all 3 lead words). The problem is not in the process of deriving the voltage value as the raw ECGData value is also fluctuating in the same fashion.

Moreover, would you be so kind and explain me why the offset is shifted to 1.3V in TESTTONE measurement? From datasheet, I see that min value of GAIN0 is 0V and maximum is 2.3V, but typical is 1.3V. Does that mean that the signal I am measuring must be shifted to offset somewhere between 0-2.3V or this is done automatically? I know that in Example 4, we have the SINGLE ENDED INPUT set up, therefore (if I understand it correctly) we have 150Hz sinusoidal frequency with 1mV amplitude connected to + sign of the differential amplifier and 1.3V to - sign branch of the differential amplifier therefore I would expect that these two signals are subtracted and it should be fluctuating around -1.3V not 1.3V

I somehow guessed that in this ELECTRODE mode, values could be only positive but I do not understand how we managed to get 1.3V with differential amplifier with 1.3V on - and 1mV on +, I can get only values between:

Minimum value (000...) = 0 V
Maximum value (1111....) = 2 × VREF/GAIN
LSB = (2 × VREF/GAIN)/(2 N – 1)
ECG (voltage) = ECG Data × (2 × VREF/GAIN)/(2 N – 1)

Therefore it must be possitive, but how and why?

But I do not understand how the differential amplifier works this way, should I undestand it as not a differential amplifier but a summation of VCM signal (like WCT) creating a sort of a virtual lead and the sinusoidal signal is superimposed to VCM signal, expecting maximum voltage of 1.301V? This also relates to my second question.

My second question is the fact that I do not understand why my signal is fluctuating around 1.26V randing from 1.262V to 1.264V which is moreover exceeding the expected amplitude peak-to-peak 1mV, in ideal case I would expect it to be varying between 1.301V to 1.300V or 1.300V to 1,299V. I have enclosed snapshot.

Thank you very much and wish you all the best to new year!

Ivo



correction
[edited by: Ivo Hora at 11:45 AM (GMT -5) on 31 Dec 2020]
  • Hi,

    We will look into this, I'll contact the product owner and get back to you.

  • Hi, apologies for the delay answering this post.

    The gain is not 0V (min) and 2.3V (max). That range is the valid single-end input voltage range for an electrode when GAIN0 is used (GAIN0=1.4V/V)

    The four available gain settings are
    GAIN0: x1.4 V/V
    GAIN1: x2.1 V/V
    GAIN2: x2.8 V/V
    GAIN3: x4.2 V/V

    The associated maximum differential input voltage is
    +-1.00V at GAIN0 (the electrodes´ voltage can be anywhere between 0.30V and 2.30V)
    +-0.67V at GAIN1 (the electrodes´ voltage can be anywhere between 0.63V and 1.97V)
    +-0.50V at GAIN2 (the electrodes´ voltage can be anywhere between 0.80V and 1.80V)
    +-0.30V at GAIN3 (the electrodes´ voltage can be anywhere between 0.97V and 1.63V)

    The input signal must be centered at 1.3V. In a real application with a RLD electrode, the RLD amplifier takes care of biasing the body close to 1.3V. In a lab test, the incoming signals must be centered at 1.3V. The ground of your signal generator must be connected to the ADA1000´s ground . Set the function generator to drive a) a high impedance load (Zout=High but not 50 Ohm) b) DC offset=1.3V and c) whatever AC signal you want to drive

    Your assumption is correct. The positive input is the ECG signal and the (internal) negative input is DC voltage equal to 1.3V. At GAIN0 the positive input can range between 0.3V and 2.3V (1.3V +- 1.0V) . The common mode of both signals is subtracted, the difference amplified and centered again at 1.3V for the ADC to make use of its full range .

    1) a 0V differential input (1.3V at IN+) will be digitized as half scale (b10….00),

    2) a differential voltage of -1V (VIN+=0.3V) will be digitized as zeroscale (b00…00)

    3) a differential voltage of +1V (VIN+=2.3V) will be digitized as fullscale (b11…11)

    I hope this clarifies your questions but let us know otherwise

    Best regards

      Roberto

  • Dear sir,

    Sorry for late reply.

    Thank you very much for your detailed info. I would like to kindly ask you If you could help me also clarify my issue with TESTTONE signal. I have set up ADAS1000 according to example 4. In this particular case we have set up 150Hz sinusoidal signal on all leads with amplitude of +-1mV. Since all bits leads of VCM signal are set to 0, no lead is contributing to the VCM signal and the VCM signal is equal to VCM_REF = 1.3V and is connected to minus input of all differential amplifier before ADC. On each plus input of a differential amplifier there is a 150Hz signal of amplitude 1mV (according to datasheet with tolerance of +-0.1mV) common-mode voltage, VCM_REF = 1.3 V.

    If we look at WORD1, we have LA-VCM that means we have 1.3V - 1.3V + TESTTONE. So if the common-mode is subtracted then the remaining signal is TESTTONE = 1mV and how is the difference amplified and with what value? I would expect the resulting signal to be 1.3+-1mV, but my signal is fluctuating around 1.26V ranging from 1.262V to 1.264V. I have no RLD drive connected, but I assume that it anyway makes no sense for this example since leads are internally connected to TESTTONE generator.

    But how come we get signal fluctuating around 1.26V and what should be the desired output in this example? Should we expect signal ranging from 1.299 to 1.301?

  • Please see the customers follow up response below.

    Thank you

  • Hi Ivo:

    Yes, ideally  the values should be fluctuating around 1.3V as you point out but what you see is also as expected, (as long as you get a peak to peak value of 1mV +-10%).

    The ECG channel has an offset which depends on the GAIN setting. It ranges from -7mV (GAIN3)  to -22mV (GAIN0), typical values. The DAC used to drive the test-tone also has an offset too which can add a few extra mV  to the channel´s offset while monitoring the TESTTONE mode.

    As you probably know monitoring the low frequency in an accurate noise-free way is of great importance in an ECG measurement but the DC accuracy is not that meaningful, hence the relaxed specifications for the ECG channel´s offset.

    Keep in mind the peak to peak value of the tone may also be attenuated by the low pass filter (mostly the 150Hz tone if a 150Hz or 250Hz low pass filter is used, as set at register FILTCTL).

    Best regards

      Roberto