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.

MSPM0G1506: About operating in high-humidity environments.

Part Number: MSPM0G1506
Other Parts Discussed in Thread: SYSCONFIG

Hello,
My customer is asking about operating in high-humidity environments.
They are using MSPM0G1506 in their application and have observed a phenomenon where ADC conversion values fluctuate in high-humidity environments.
(Question)
1) Is a guaranteed operating humidity range specified for the MSPM0G150x series?
We couldn't find it in the datasheet but would the following links be helpful?

Qualification summary
www.ti.com/.../home

JESD22-A102/JESD22-A118
JESD22-A101/JESD22-A110

2)About impact on internal OPA, ADC, and Vref in high-humidity environments(over 80%).
We have observed a drop in the output voltage of the I-V conversion circuit using an internal OP amp under conditions where humidity exceeds 80%.
The output voltage is being converted by an internal ADC, and since fluctuations are observed  in the AD counts, they guess that the internal reference voltage (Vref) is changing.
Is there a possibility that high-humidity environments could affect the internal OPA, ADC or Vref in the MSPM0G150x series?

This issue occurs with a frequency of approximately one in ten units.
Would you give comments and advice about the operating under high-humidity environments?

Regards,
Tao_2199

  • Hello Tao,

    In theory, the high humidity won't affect MCU work. Could you please help to confirm more details with customer. 

    • How customer control the high humidity? Through control temperature? Could you please share more details of customer test environment?
    • You mean ADC conversion result will have fluctuate. Could you please help to quantify the error?
    • Is it possible that the high humidity affect other device on customer board, such as the voltage source? 

    BR,

    Janz Bai

  • Helllo Jnaz,

    I have received a reply from them.

    >How customer control the high humidity? Through control temperature? Could you please share more details of customer test environment?

    ->The testing environment is a chamber with controllable temperature and humidity. The generated conditions were 25–25.3°C and 75–77% humidity.

    >You mean ADC conversion result will have fluctuate. Could you please help to quantify the error?

    ->The fluctuation in the count value is 33 counts (compared to an environment of 22.8°C and 57.6% humidity; standard laboratory conditions).

    >Is it possible that the high humidity affect other device on customer board, such as the voltage source? 

    ->I would like to contact you via email because they wish to share the circuit diagrams, which are considered confidential information.

    Regards,
    Tao_2199

  • Hello Tao,

    Thanks for your quick response. About the ADC conversion fluctuation, I found you mean is 33 count. Could you please help to give more explanation about the 33counts, does it mean "if the ADC is 12bit, the max count is 4096 (2^12), the ADC conversion fluctuation is just 3, the error is just Vref/4096 * 3?"

    And it seems that the ADC conversion result is compared under "22.8C and 57.6%" and "25-25.3C and 75-77%", could customer try to compare the ADC conversion result under "22.8C and 80%" and "22.8C and 57.6%" or compare the result under "25-25.3C and 75-77%" and "25-25.3C and > 80%". In other words, keep the temperature same.

    And about the system schematic, if customer share with us, we can just review the MCU module. About whether the device on the board will be affected by humidity, it needs customer to check accroding to the other device spec in their datasheet. 

    BR,

    Janz Bai

  • Helllo Jnaz,

    Thank for your quick response.

    >. Could you please help to give more explanation about the 33counts, 

    ->The "33 counts" means the fluctuation (diffrence) in the conversion results between normal humidity (57.6%) and high humidity (75–77%).
    33 counts = AD count @ 57.6% − AD count @ 75–77%

    ADC is 12-bit, and Vref is 1.4 V.
    The maximum count is 4096 (2^12), meaning the fluctuation value corresponds to Vref/4096 × 33.

    >could customer try to compare the ADC conversion result under "22.8C and 80%" and "22.8C and 57.6%" or compare the result under "25-25.3C and 75-77%" and "25-25.3C and > 80%". 

    ->They will verify whether it occurs under the same conditions.

    I have requested friendship to you for private message

    I would like to share their circuit diagrams with you, so please let me know how I should go about doing so.

    Regards,
    Tao_2199

  • Helllo Jnaz,

    Would you let me know current status about high-humidity environments and thire circuit diagrams?

    Regards,
    Tao_2199

  • Helllo Jnaz,

    They checked the ADC count values ​​and current values ​​under conditions of 25°C/50% RH and 25°C/75% RH.

    25.2℃/51.7%RH  -> ADC count/Current = 3038/96157 (nA)

    25.1℃/75.1%RH  -> ADC count/Current = 3167/82358 (nA)

    The current value(nA) is calculated from the ADC count value using software.

    The ADC count difference fluctuated by approximately 129 counts.

    Regards,
    Tao_2199

  • Hello Tao,

    Sorry for the late response. In theory, the humidity won’t affect MCU work, but for locating the root cause, we still need to further confirm some points with customer and below are my comments:

    • You can share the MCU module schematic, better do some notes, especially what is the OPA input, etc. And please also share sysconfig file here, I want to do a simply check for the OPA and ADC configuration. You can give these to our field team and let them email to me.
    • According to the last time test under same temperature and different humidity, the ADC conversion result has about 129 counts error, nearly about 3%. We need to locate which part in the system is the root cause. So please try to do below tests or check below points:
      • Just run a OPA demo, and input a very stable signal into OPA, output the OPA, check whether the OPA will be affected by the humidity
      • Just run a ADC demo, and input a very stable signal into ADC, read the ADC conversion results, check whether the ADC will be affected by the humidity
      • Don’t use VREF as reference of ADC, use VDDA as reference of ADC, and confirm the VDDA is stable and won’t be affected by humidity, then check the ADC conversion results, to check whether the VREF will be affected by humidity.
      • Please keep temperature same, just change the humidity when doing above tests.
      • Whether other thing will be affected by the humidity change, for example, the OPA input signal, the noise level of the OPA input signal, the voltage source, etc.
      • If all the external thing won’t be affected by humidity, we can check MCU continuously. Could you try below tests
        • Just run a OPA demo, and input a very stable signal into OPA, output the OPA, check whether the OPA will be affected by the humidity
        • Just run a ADC demo, and input a very stable signal into ADC, read the ADC conversion results, check whether the ADC will be affected by the humidity
        • Don’t use VREF as reference of ADC, use VDDA as reference of ADC, and confirm the VDDA is stable and won’t be affected by humidity, then check the ADC conversion results, to check whether the VREF will be affected by humidity.
        • Please keep temperature same, just change the humidity when doing above tests.

    BR,

    Janz Bai

  • Hello Jnaz,

    Thank for advice and comments.
    I will announce to my customer them.
    As there wasn't explanation regarding how to share the schematic, I sent it to  TIJ FAE(Takigasaki-san) by email.
    (Question)
    > Just run a OPA demo,
    > Just run a ADC demo,
    For example, which demo should they use in SDK?
    Would you tell us specific details?

    Regards,
    Tao_2199

  • Hello Tao, 

    Yes, you can share the schematic with our FAE, and let him send to me by email.

    About running OPA demo, customer can use the same OPA configuration as they used in their application, input a very stable signal into OPA, and output the OPA result through OPA_out pin, and check it.

    About running ADC demo, customer can use the same ADC configuration as they used in their application, but don't use OPA as ADC input, but input a very stable signal through ADC input pin, and then check the ADC conversion results. Or customer can also use the simple ADC demo in SDK to do tests, such as the "adc12_single_conversion"

    BR,

    Janz Bai

  • Hello Jnaz,

    Now mu customer are verifying your comments and advice.

    I sent the circuit diagrams the other day; would you provide any comments or advice regarding them?

    Regards,
    Tao_2199

  • Hello Tao,

    I have checked this part circuit. To be honest, I am still a little confused about this circuit. For example:

    • Are HCT_HC_DAC and DAC_OUT signal are same, both generated by DAC12?
    • HCT_HC_DAC is connected to inverting input of the internal OPA, which OPA, OPA1?
    • Customer said, OPA1 generates 200mV at HCT-HC, so HCT_HC is the output of OPA1? 
    • If HCT_HC is the output of internal OPA, but from the calculation formula in the figure, HCT_HC and HCT_SIG_OUT is calculated by HCT_HC_DAC and the external R9. R10, R11. So I want to understand the logic

    I think we need more detailed explanation about this circuit, or at least complete signal connection figure with MCU module.

    BR,

    Janz Bai

  • Hello Jnaz,

    Thank for your quick response.

    I have announced your comments to them.

    I will let you know as soon as I receive a response from them.

    Regards,
    Tao_2199

  • Hello Jnaz,

    Sorry for the late response. The results of their verification are reported below.
    An investigation into the effects of humidity suggested that the cause lay elsewhere than the MSPM0G150x itself.
    They confirmed that cleaning resistors R05 and R06—which generate the reference voltage (Ref voltage) for HCT_HC—with anhydrous ethanol resolved the issue. This leads us to speculate that water-soluble flux residue remaining on the resistor surfaces was the contributing factor.
    They believe that flux residue caused leakage current on the resistor surfaces, altering the effective resistance values; this changed the voltage division ratio established by R05 and R06, thereby affecting the reference voltage. In particular, R05 has a very high resistance value (in the MΩ range), making it highly susceptible to the effects of even minute leakage currents.

    As a countermeasure, they reduced the resistance values while maintaining the same voltage division ratio for R05 and R06, thereby reduction the impact of leakage currents caused by flux residue. Additionally, because thick-film chip resistors are prone to significant resistance fluctuations due to moisture absorption, we switched to metal-film resistors.

    Regards,
    Tao_2199

  • Hello Jnaz,

    That summarizes the situation; since the customer has resolved the issue, I will now close this thread.

    Thank you for all your support .

    Regards,
    Tao_2199