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.

Why the ambient signals of AFE4400 EV kits always drift?

Other Parts Discussed in Thread: AFE4400, AFE4490

Hi,

I am using AFE4400 EV kits to do a university project. I am writing to ask why the ambient signal given by EV kits
always drift over time. I did the experiments in the following situations.

Case 1: The finger clip, coming with the purchased EV kits, is connected to the EV kits. The
ambient light is drifting over time. The noise floor is about ADC count of 150.

Case 2: Disconnect the finger clip. Neither LED or PD is connected to EV kits. But a resistor of
500Kohm is connected between INN and INP of the AFE4400 chip. The ambient signal (ADC count of
-3000) with noise floor of 2000 count, always drift.

Case 3, 4, 5 and 6: The setup is the same as case 2 except that the 500Kohm resistor is replaced
by 100K, 20K, 10K and 5K respectively. In those cases, the average DC count of ambient signal
are 2.4e4, 7.1e4, 14.6e4 and 28.2e4. The noise floors are respectively ADC count of 600, 1000,
2000 and 4000.
I have the following questions.

1. Why does ambient signal (ADC count) change from positive to negative when comparing case 2
to case 3~6? Is it because that 500Kohm resistor makes the fully differential transimpedance
amplifier unstable?

2. By comparing case 4~6, we found the ambient signal level is inversely linearly proportional
to the resistance of the used resistor. It makes sense to me by considering the effect of the
offset voltage of the transimpedance amplifier. But by comparing case 3 and 4, the ambient
signal level is not inversely linearly proportional to the resistance of the used resistor. Do
you know why?

3. Why the ambient signal always drift no matter how long (10 minutes, 20 minutes or 1 hour) I
keep the EV kits on? Besides thermal effect, I think there should be another effect making the
ambient signal always drift. Could you explain what the signal might be? Could you also tell me
the typical offset voltage of the transimpedance amplifier used in AFE4400? And what's the drift
value per 1 degree Celsius?

The drift has been confused to me for a quite bit long time. It will be very helpful if you can
let me know the reasons. Thank you very much for your help in advance,

Bin

  • Hi Bin,

    We have not seen this ambient drift. Have you made any modifications to the EVM or firmware? I will replicate your experiments to see if I get similar results.

    Can you explain what you are trying to determine by adding various resistors to the input of the TIA? By connecting a resistor across the positive and negative inputs of the amplifier, any voltage across it due to thermal noise or bias currents will see open loop gain. This is not a good way to model the behavior of a photodiode since a photodiode can roughly be modeled as a current source with a parallel capacitor.

    Can you tell me what rate the signal is drifting at? Is this constant or does it change? Is it different for the different test cases?

    Regards,

    Gui

  • Hello Gui,

            I've seen the same ambient drift on AFE4490 EVM. From the AFE4490 Datasheet, the diagram in fig. 74 shows how LED bias current passes 50uA whenever LED_ON occurs.  In the ambient sampled state, there are still 1uA current pass through the LED.
           My question is:
                     What is the reason that you want to put 1uA current pass through the LED when LED_OFF?? Any concern??
                     I think this is the reason why that Bin can see the ambient drift.
    Thanks.
    JingLin

  • JingLin,

    The 1uA is the leakage current when LED_ON = 0. We have not seen the ambient drift on our EVM's despite this leakage using the finger clip sensor that come with the kit; are you using this finger clip sensor?

    Regards,

    Gui