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.

PPG signal cut-off with AFE4400

Other Parts Discussed in Thread: AFE4400, AFE4490

Hi everyone, I do experiment with AFE4400 board made by myself with DCM03 LED/Photodiode sensor.

Below is my register setting: 

Register : 0x21       Value: 0x000005   //Ambient DAC 0uA, stage 2 amplify disable, stage 2 gain 0 dB, Cf= 5 pF, Rf=  10k

Register : 0x22       Value: 0x01FFFF   //Current for LED = 50mA

Register : 0x23       Value: 0x020100   //H-bridge

It is strange that my collected data seems to be cut-off even though they are still in range of signed 22 bit. You can see below photo to understand more.

Sorry for bad picture. All value of data are negative (<0)

Could you guys help me to solve this problem.

Thank you very much

  • Linh,

    We have received your post and hope to have a response back to you soon.
  • Hi, Amy

    Please help me.

    Thank you,
  • Linh,

    Is this data at rest or with motion?
  • Thanks for reply, Amy

    These data was measured at rest. I tried many times with some different configurations but outputs seem to be the same.

    Data only continues in range of ~32000 (2^15), if data is below or above this band, they  will immediately  be raised up or cut down by ~66000

    You can look at below attached image

    Value in the vertical axis is read by AFE4400's ADC.

  • Linh,

    I assume this is using a custom board and not our EVM?

    Have you looked at our sample code that we provide to make sure you are interpreting the data correctly?

    www.ti.com/.../toolssoftware
  • Hi Amy,

    Yes, I am using custom board  for experiment. 

    I wrote my AFE4400  controlling functions based on TI's sample code.

    I also checked the code by setting and reading register values of AFE chip. All the values are written and read correctly so I don't think there is problem with it.

    It seems like the ADC of AFE working wrong.

  • Hi Amy,

    Can you help me please.

  • I see in your first post you are in the range of signed 22-bit data. What do you mean by signed? The reason I ask is because this is a 22bit 2's compliment ADC output code.

    To interpret the ADC codes, please follow the FAQ section (question # 18) in the link below to get an understanding on how to get HR data from the ADC output.

    e2e.ti.com/.../369445
  • Two's complement data IS signed.  Bit 21 is the sign bit in 22 bit data.  Why is this thing giving negative numbers for light levels?  Negative light is not possible.

    SPI capture of config registers:

    Low power setting:

    Reg 22 power level = 0C0C
    2A = FFFC50 = 3FFC50 (22 bits) = bit 21 is 1 = negative number
    2B = FFFC2F = 3FFC2F (22 bits) = bit 21 is 1 = negative number
    2C = FFFC47 = 3FFC47 (22 bits) = bit 21 is 1 = negative number

    High power setting:

    Reg 22 power level = F0F0
    2A = FFFC41 = 3FFC41 (22 bits) = bit 21 is 1 = negative number
    2B = FFFC55 = 3FFC55 (22 bits) = bit 21 is 1 = negative number
    2C = FFFC53 = 3FFC53 (22 bits) = bit 21 is 1 = negative number

  • Scott,

    Section 8.4.1 (page 38) of the datasheet (assuming you are working with the AFE4400) describes the ADC output code for different input voltages.

    The output format of the ADC is 22-bit two's complement and the 2 MSB bits can be ignored.

    You may notice very small negative voltages during the ambient phases due to the channel offset of the AFE4400/AFE4490.

    The ADC full scale voltage can go from -1V to +1V. Since the photodiode (PD) current is one sided current, only one side (0 to 1V) of the ADC range is used. Now with no PD current during the ambient phases, we could be getting 0V ideally but with the channel offset this will be slightly higher or lower. This should not be an issue and is as expected.

    The channel offset will be cancelled out since the ADC value during the ambient phase is subtracted out from the ADC value during the LED phase (RED – RED Ambient or IR – IR Ambient).