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 practice for single-ended input

Part Number: ADS1291
Other Parts Discussed in Thread: ADS1292ECG-FE

Hello,

I have identified the ADS1291 24-bit ADC and it fulfills all the requirements in terms of resolution, data rate, power-consumption and interface. I don't plan to use for ECG application, but to measure a single-ended sensor output (0V-1V) (very slow changes).

However, I am surprised by the definition of single-ended input in the datasheet (SBAS502C, page 22, section 8.3.3). I expected the INN input to be tied to Ground, not to mid-supply. If I tie INN to ground, the common mode voltage is 0V according to Figure 21, but then, it is not within the range defined by equation 5 (page 24) (gain=1, VMAX_Diff = 1V).

According to the definition, INN input should be tied to a voltage (mid-supply). Can I simply connect a 1.5V voltage reference to INN and connect my sensor output (0-1V) to INP ? Would that work?

I have some experience with single-ended ADC where the negative input is tied to ground and fully differential ADC, but never come across this specific case. I'd like to have your opinion on this and know the best practices.

Thanks for your support,

Mathieu

  • Hi,

    Note there is a line of smaller font at the Figure 21 -  "Common-Mode Voltage (Single-Ended Mode) = INN"

    It means, when you setup the device to Single-Ended Mode, whatever input at the INN input pin is the common-mode voltage. Yes, I know the common voltage concept is vague in single-ended mode. And, such common-mode voltage could be GND.

    Note, for differential inputs or differential signals,  common mode voltage equation is  (V+ + V-)/2   and the differential amplifier doesn't know/care what V+ are V- are individually, differential amplifier only cares/sees  the difference between V+ and V-,  i.e. V+ - V-.   With this being said,  if the V+ and V-    are both riding with respect to GND, then, the differential amplifier sees&amplifies   (V+ + GND) - (V- + GND) =  V+  -  V-.   and if your V- is actually a GND as well, the differential amplifier sees&amplifies  V+  -  GND = still V+.

    In the case if your inputs are differential signals, i.e. V'+ and V'- are opposite polarity(or 180 degree phase shifted) and are riding(i.e. symmetric/mirrored) about a common mode voltage, and such common mode voltage has to be (V'+ + V'-)/2 and within the spec, the differential amplifier will still sees/amplifies the difference i.e. when V'+ = V+ + VCM and V'- = V- + VCM, the differential amplifier sees/amplifies V'+ - V'- = still V+ - V-  and if they are inverted(180 degree out of phase), it's 2*V+.

    Regarding to your sensor question, is your sensor output reference with respect to 1.5V?  I.e. what are the difference Vsensor - Vref(in your case 1.5V)?

     Also, the entire system should be a common analog GND. Otherwise, your 1.5V with respect to GND on one side may not be 1.5V with respect to GND on the other side.

    And, please refer to datasheet 6.1 Absolute Maximum Ratings, 6.3 Recommended Operating Conditions, 6.5 Electrical Characteristics for the analog input ranges, limits and constraints as it depends on other settings e.g. AVDD, AVSS , VREF, gain, etc. So, yes, your VMAX_DIFF = maximum differential signal at the input of the PGA, CM = common-mode range, Gain, AVSS and AVDD have to meet/satisfy the page 24, equation (5); otherwise, the results&performance cannot be guaranteed. 

    Thanks

  • Hi ChienChun,

    Thank you for your reply.

    My sensor output is referenced to GND, and this output varies from 0V to 1V, with respect to GND. If I understand correctly your second sentence, the sensor output can be connected to INP (V+), and INN (V-) can be connected to GND ?

    I have verified the specifications/characteristics 6.1, 6.3 & 6.5. However, it doesn't satisfy equation (5) from page 24 in these circumstances:

               VMAX_DIFF=1V (1V max - GND = 1V)

               GAIN=1; AVSS=0V (GND); AVDD=3.3V

               Common-mode = 0V; but equation (5) says 2.6V > CM > 0.7V

    That's why I planned to connect a 1.5V voltage reference to INN (V-) and have the sensor output connected to INP (V+), so that equation (5) is satisified. But is it a good idea ?

    Could you please comment on that ?

    Thanks

  • Hi,

    If reads and/or amplifies Vsensor - Vref(1.5V)  is what you want, then you can try.

    Just make sure your input at INN should meet the 6.1 Absolute Maximum Ratings e.g. 

    Analog input voltage between AVSS and AVDD

    Input current to any pin except supply pins ±10 mA

    6.5 Electrical Characteristics

    Full-scale differential input voltage (AINP – AINN) should be bound between ±VREF / gain V

    May I ask do you happen to have an ADS1291 evaluation board/kit EVM?

    Thanks

  • Hi,

    What I ultimately want to read it Vsensor-VGND(0), but as mentionned previously, it does not satisfy equation (5).

    That's why I came up with this idea of connecting a Vref=1.5V to INN, and compensate for it in software/firmware.

    I don't have the ADS1291 kit yet, is it the ADS1292ECG-FE kit ? It does not seem to be in stock on TI.com or any major electronic component resellers.

  • Hi,

    Let me reach out or other department and get back to you regarding to ADS1292ECG-FE evm's availability. 

    Thanks

  • Hi,

       Please consider buying 

    https://www.ti.com/tool/ADS1292RECG-FE

    ADS1291, 1292 and 1292R are the same family that shares the same EVM and EVM user guide.

    The letter R is just for respiration function/feature for ECG application, it's optional to use or not use.

    Thanks

  • Hi,
    Since I did not hear back from you, I believe my suggestions answered your questions.
    I will close this post and if you have any pending questions, feel free to post them here or open a new thread.
    Thanks and have a great day!