• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 30 subscribers
  • Views 1660 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.

AFE4404: Can the AFE44xx series ICs be used for NIRS applications?

Robert Marosi
Prodigy 50 points
Part Number: AFE4404

Hi all,

I'm currently designing a device which will be used for functional near-infrared spectroscopy (fNIRS) and I have a couple of questions to find out if the AFE44xx series ICs is suitable for what I'm trying to do. The concept of operation is very similar to pulse-oximetry, but the spacing between the LEDs and photodiode is larger (> 2.5cm) and I'm going to be monitoring changes/oscillations in oxy- and deoxy-hemoglobin concentrations, rather than absolute concentrations.

  • I will require approximately 8+ individual sensors. Is it possible to multiplex the SPI interface of several AFE44xx chips? What about the chips with an I2C interface? Could I possibly use an additional I2C multiplexer to control and sample each of the AFE44xx chips sequentially? I would be polling the averaged red/IR ADC values periodically.
  • Is it necessary to use the CLK output of these chips to clock my microcontroller or processor? If I'm to use more than one of these ICs, I probably can't use a common clock for all of them due to how far they are spaced apart. Will there be significant downsides to operating without a central clock? The fundamental frequency of what I'm trying to measure is quite low (< 10Hz).
  • Can the AFE44xx ICs also drive laser diodes? Since the spacing between the LEDs and photodiode would be relatively large, the attenuation of light will be higher. I might need the increased light intensity.

Thanks,

Robert

  • 0
    TI__Genius 12575 points
    Hi Robert,
    Here are the answers.
    1) Do you need multiple LED s and multiple photodiodes (PD) or multiple LED and single PD? In either case you can use several AFEs . But there are some points to be considered such as clock to all AFEs should be same and timing engine ( of each AFE) should be enabled at the same time. You can control AFEs using external switches in both the interfaces (SPI or I2C).
    2) You can run your controller on its own clock. For deterministic operation it is advised to synchronise all the AFEs with a common clock. That being said this also depends on the application, such as if separation between sampling instances in time domain is used in post processing of the signal. One way to clock all the AFEs is to generate a single clock from MCU and feed it to all the AFEs. Multiple GPIOs of MCU can be used to generate a same clock and assign each GPIO to a single AFE.
    3) Laser diodes can be used as long as AFE's requirements are fulfilled. Please refer to following FAQ page (#20) for such requirements/considerations.

    e2e.ti.com/.../369445


    Regards,
    Prabin
  • 0 in reply to Prabin Yadav
    Prodigy 50 points

    Hi Prabin,

    Thank you for the information.
    1) Each sensor module will consist of 2+ LEDs and a common photodiode, like in the usual pulse-oximetry application. The LEDs and photodiodes will not be shared between different sensors. In this case, the sensors can operate independently from each other, as long as light from the sensors are not interfering with neighboring sensors. I will likely use an external I2C multiplexer to control and sample from each sensor individually.
    2) I'll consider clocking the sensor modules from the MCU. If GPIOs are limited, could a fanout clock buffer also work for clock distribution? In either case, the clock lines will all need to be length matched and may require some series terminations depending on frequency and trace length, correct?
    3) Thank you. I'll narrow down my selection of emitters based on those requirements.

    Thanks,
    Robert