• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 1385 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

AFE4300: ADC Values of weight measurement inconsistent

SHANKARNARAYAN PS
Intellectual 625 points
Part Number: AFE4300

Hello,

I just got the AFE4300 sesnor up and running with an Arduino DUE. It all works fine from a flow perspective but I see some inconsistent behavior on weigh scale front.

1) The adc values read for same voltage (12mv) ranges from 330 - 550 am not sure why this is happening

2) When I try to measure the voltage using a multimeter, the VLDO goes off!!! if I hold the multimeter before I turn on weight scale front end, then no issues I read 1.7V always but if I try to put the multimeter after the front end is on and try to measure, it goes off!!! and I read 0.0 on ADC (Really strange)

Can you please help me in this regard?

regards,

Shankar

  • 0
    TI__Mastermind 49916 points
    Hi Shankar,

    Can you provide scope waveform plots of the following signals: AVDD, VLDO, VREF and CLK while turning on the weigh scale FE?
    Is your Arduino board supplying 3.3V for the AFE?

    For the weigh scale signal chain, are you using load cells or potentiometer on a wheatstone bridge?
  • 0 in reply to Praveen Aroul
    Intellectual 625 points

    Hello Praveen,

    Find below the pics of VLDO/VCC abd VLDO/CLK captured on my scope. I see that the VLDO is 1.7V/ Vcc is 3.3V and clock is close to 1MHz with high being at 3.3V.

    For Weight scale am using a 1Mohm and 22k pot in series and am measuring voltage across 22k pot.

    Regards,

    Shankar

  • 0 in reply to SHANKARNARAYAN PS
    TI__Genius 12575 points

    Hi Shankar,

    Are you using circuit as shown below to generate input for weight scale?

    Regards,

    Prabin

  • 0 in reply to Prabin Yadav
    Intellectual 625 points

    Hello Prabin,

    Yes this is the circuit am using for weight scale but please note that on the 22k POT, I use one end and the center tap and not the other end (Resistance between two ends will always be 22k) . I also set the pot to such a position that the voltage drop is about 20mV (measured it with multimeter) and then conencted the signals to INP1 and INM1

    Regards,

    Shankar

  • 0 in reply to SHANKARNARAYAN PS
    TI__Genius 12575 points

    Hi Shankar,

    Thank you for confirming, but I am afraid you can't use this circuit to generate the input.
    The problem with this setup is that it doesn't meet the common mode requirement (i.e. 0.9V) of input. You differential signal should have a common mode voltage of 0.9V.
    You can use two such circuit to generate the required differential input.
    For e.g you need differential of 20mV, then
    set one resistor divider to : 900 + (20/2) = 910mV and
    set other divider to : 900-(20/2) = 890mV
    So in this case (910mV and 890mV) you have differential input of 20mV with common mode 0.9V

    Regards,
    Prabin

  • 0 in reply to Prabin Yadav
    Intellectual 625 points
    Hello Prabin,

    Thanks for the answer, I completely missed that. Ok I will try putting a differential signal the way you have suggested.

    Is there a similar requirements for BMI part? I am running with another issue with BMI

    1) I set the mode to BMI, disabled IQ mode and set DAC frequency to 64KHz
    2) I measured the impedance on calibration terminals (100 ohm and 1K) and calculated the slope.
    3) I connected 560 ohm, 680ohm and 1kohm for three terminals (I shorted IOUTx and VOUTx since I was passing the current measuring voltage across same points)
    4) Then I changed the ISW_MATRIX modes to 0x0408, 0x1020 and 0x4080 respectively to measure the resistances. I got 566ohm, 686ohm and 1022ohm fair enough! I was really happy at this stage.

    However, when I remove the 1Kohm resistance and try to put my right hand and right limb in the same place through wires, I always get a saturated 2Kohm + values.

    I rechecked with lot of standard resistances and I get agreeable values but nothing turns up with my hands or legs. Am I missing something? Please help.

    regards,
    Shankar
  • 0 in reply to SHANKARNARAYAN PS
    TI__Genius 12575 points
    Hi Shankar,

    In BCM signal chain, the common mode is set by the device.
    To avoid the saturation you can reduce the excitation current by placing an external resistor between DAC_OUT and DAC_FILT_IN.
    Since the excitation current is changed, you have to calibration once again.

    Regards,
    Prabin
  • 0 in reply to Prabin Yadav
    Intellectual 625 points

    Hello Prabin,

    Ok but in that case what are the typical values of Body impedance that I can expect? Do you have any documents that can support me in that? That will be really helpful. The issue is that I do not know what typical value of impedance I can expect what I put my body in the circuit. AllI know is some standard resistances that are being correctly measured by the device. This is what I have currently

    Calibration done using 100 ohm and 1kohm on the board
    Measured across Measured value Expected 
    100 ohm 108 ohm Yes
    560 ohm 562 ohm Yes
    680 ohm 686 ohm  Yes
    Person 1 (hand to Hand) 200 ohm ??
    Person 2 (Hand to Hand) 74ohm - 100ohm (in different trials) ??



    Please do help me with supporting documents as to what are typical body impedance values that I can expect normally.

    Regards,
    Shankar

  • 0 in reply to Prabin Yadav
    Intellectual 625 points

    Hello,

    Well am in a bit of trouble here. According to the documents the range of the impedance measured from hand to hand is to be in range of 1k ohm. Unfortunately when I pass current from right arm to left arm and measure the voltage across same points, I measure 100 ohm 200 ohm and 500 ohm for three different people. But however, the standard resistances measure agreeably.

    I checked several times all the signals and clocking. Everyting seem to be fine. Am unsure why I have values that are too less. Can you please provide me any leads that can help? Where can I look in to?

    Regards,

    Shankar