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.

Question about AFE4490

Other Parts Discussed in Thread: AFE4490SPO2EVM, AFE4490, AFE4400

Dear Sir,

    I bought an AFE4490SPO2EVM from TI. I had some questions about the chip setting. Please help me to know more on it. I just list below:

(1) What applications do you suggest to use in such high pulse repetition frequency (5KHz) and low pulse repetition frequency (61Hz)? 

(2) What kind of Photo-detector do you suggest?? Can you comment on the photo diode and photo transistor??

(3) What is the real ADC clock in AFE4490??  (4MHz or 20KHz??)

(4) Why did you need High resolution timer module in AFE4490 (16Bits counter)? 

(5) What wavelength (LED) is the best choice for heart rate monitor?? Why??

Thanks for your time and help.

  • JingLin,

    (1) What applications do you suggest to use in such high pulse repetition frequency (5KHz) and low pulse repetition frequency (61Hz)?

    LP:   Applications looking at heart rate variability and those looking to do PPG waveform analysis need high data rate.  For heart and simple SPO2 analysis slower rate is fine.  

    (2) What kind of Photo-detector do you suggest?? Can you comment on the photo diode and photo transistor??

    LP:  AFE4400 was designed to work best with photo-diodes.  Photo-diodes tend to be bigger compared to photo transistors, but require special biasing and potentially burn more power.   I suggest searching the e2e forums on using photo-transistor and photodiode with AFE4400.  Much has been written already on the subject.

    (3) What is the real ADC clock in AFE4490??  (4MHz or 20KHz??)

    LP:  4MHz.  ADC is a oversampling delta sigma type, as explained in the datasheet.

    (5) What wavelength (LED) is the best choice for heart rate monitor?? Why??

    LP:  Heart rate can found with many wavelengths.  You'll find papers and articles on the internet that say RED/IR/GREEN are good choices. 

    Good luck.

  • Thanks a lot for your reply. Three more questions.

    (6) Can I turn on 2*LED simultaneously?? I mean Two LED "ON" at the same time.

    (7) I found 3 different PRF number. Which one is correct??

         (a)GUI = 2KSPS   

         (b)Website = 1.3 KSPS

         (c)Datasheet = 5 KSPS
    (8) Why did you need high resolution (16 bits timer) for timing adjustment?? Any purpose or reason??

    Thank you.

  • Hello JingLin,

    (6)  Yes, it is possible.  The current limit is set by user registers.  So the two LEDs will be half as bright as having one on.  We encourage you to use the EVM to test out these things.

    (7)  The datasheet is always best source for information.  We will update the websites soon.  Sorry for the confusion.

    (8)  16-bits is to give user more flexibility and control over the timing.

    Good Luck.

  • Thank you. Sorry to bother you again. 
    (9) As mention in document : AFE4490

    page 5, Output current settling time need 7 us 

    page 31: Time instant t1 = t3 = 6000 ,
    t1 = Start of sample LED2 pulse
    t3 = Start of LED2 pulse

    => My question is:
    Does LED2 need some settling time to settle down and than sampled it?? ex: t3= 6000, t1 = 6028
    Thanks.

  • JingLin,

    We are looking into this and want to verify with our design team. We will get back to you as soon as possible.

  •  This is the time for the current coming out of hte TXP/TXM pin to get to full scale( 0 to 150mA) or down.Some of this will depend on the capacitance on the TXP/TXM pins as well. This is not a transient spec, rather the time it takes for the current reference to switch modes.

    The Page 31 specs on the other hand are related to transients.

    Though in theory you could start the sampling window at the same time as the LED pulse, you might get lower noise if you were to slightly offset (delay) the sampling window with respect to the LED pulse start. The reason is that the LED signals rising transients do not get sampled by the receiver.  

    So, the sampling window is typically adjusted so that we sample it after the optical waveform from the LED is stable. The device gives all the flexibility to adjust the time based on the conditions and system parameters.

    Typically the closer you get to full scale (0mA to 150mA) the more delay you will need.