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.

AFE4490: About the clock oscillator circuit of AFE4490

Part Number: AFE4490
Other Parts Discussed in Thread: AFE4403

I am checking the hardware by newly manufacturing a circuit using AFE4490. The signal line of AFE4490 connected to CPU such as MSP430 is not connected.
SPIMOSI, SPIMISO, SCK , LED-ALM and so on.
Even if an 8MHz crystal oscillator and a capacitor of 18pF are connected to the AFE4490, the oscillation waveform cannot be observed at the Xout terminal.
Using 4.7pF, 10pF, 22pF, I was able to oscillate this 8MHz crystal oscillator with 74HCU04.
Since it was possible to oscillate with all capacitors, the crystal oscillator is normal and the load capacitance of 18pF is also appropriate.
For checking the inverter and 1/2 frequency divider circuit inside the AFE4490,
I connected the 8MHz clock oscillated by 74HCU04 to the Xin terminal.
When I observed the 4MHz clock terminal, there are times when the 4MHz clock can be observed and times when it cannot be observed.
I cannot see the 4MHz waveform constantly. Please refer to attached waveform.
The possible causes are as follows.
1. The drive capacity of the inverter inside the AFE4490 is weak and cannot oscillate.
2.74HCU04 and AFE4490 have different inverters, so it cannot oscillate with the same capacitor.
3. The inverter or 1/2 frequency divider circuit is out of order.
If you have any other possible causes, please let me know.

8MHz_4MHz_waveforms.pdf

  • Hello Matsumura-san, 

    Do you have a schematic of your implementation of the AFE4490? There may be an incorrect configuration. 

    Kind regards,
    Nick Z

  • Thank you immediately answer.
    The design circuit of AFE4490 is the same as the circuit of AFE44x0SPO2EVM.
    There may be a wiring mistake in the board production, so I will check it.
    8MHz is the same, but the crystal oscillator itself is different.
    Please let me know if there is a detailed specification of the equivalent constant in the crystal oscillator. See attached figure.

    Clock oscillation occurs when the crystal oscillator and capacitor are connected, so
    Using another AFE4490, I will try to experiment with the oscillator circuit only.
    Apply 3V to the following terminals, connect 1 to 2.2uF to 17pin, 18pin, and 0.1uF to others.
    RX_ANA_SUP (33, 39pin), RX_DIG_SUP (31pin), TX_CTRL_SUP (11pin),
    LED_DRV_SUP (17, 18pin),
    GND (3, 8, 12, 13, 16, 19, 41, 32, 36, 40)
    Does it oscillate with nothing connected to terminals other than the above?
    Does it oscillate without applying 3V to RX_ANA_SUP, TX_CTRL_SUP, LED_DRV_SUP?
    I would like to experiment with the smallest circuit. Please tell me.
    Crystal_Osc_Sch.pdf

  • Hello Matsumura-san, 

    Working on this, please allow 24 hours for a response. 

    Kind regards,
    Nick Z

  • Hello Matsumura-san,

    I think ESR might be the problem. Since the clock characteristics of the AFE4490 is the same as the AFE4403, in the AFE4403 datasheet (https://www.ti.com/lit/ds/symlink/afe4403.pdf), section 8.3.2, on page 32, we include a little extra information on circuitry and how to find the characteristics for a crystal to reduce the ESR and make sure that the crystal oscillator is not being undesirably altered.

    While the oscillator itself should function fine with only RX_DIG_SUP and RX_ANA_SUP, correct device functionality will require TX_CTRL_SUP, LED_DRV_SUP be connected and supplied. 

    Kind regards,
    Nick Z

  • Thank you for your support.
    I consider ESR.
    Apply 3V to RX_DIG_SUP and RX_ANA_SUP and
    If 0.1 μF is connected to each, the clock will oscillate.
    I understood that.
    Pin.20,21,22,23,24,25,26,27,28,29,30 not connected anything.
    Some ICs go to sleep when there is no external signal.
    Does AFE4490 have such a function?
    Clock oscillation stops in sleep mode.

  • Hello Matsumura-san,

    The AFE4490 does not have a sleep mode, so with pins 20-30 disconnected the device clock should function fine. 
    However, I would recommend you connect pins 24-27 to your microcontroller so you can read output and see if the device is functioning properly. Pins 20 and 29 should be kept high instead of left floating. 

    Kind regards,
    Nick Z

  • Dear Nick-sun,Ti_report20220526.pdf

    I confirmed the clock oscillation.

    Please refer to the attached document.

    Thank you for your support.