HDC3022: How to determine if condensation water is generated when taking out equipment from the cold storage

Hello,

I agree that using the dew point temperature is not appropriate in this case. What is the HDC3022 reading when the condensation occurs? Are you seeing RH increase or decrease as it is removed from storage? When you get a water droplet onto our HDC3020 (no filter), the RH will briefly drop to 0% before recovering to ~40%RH, you may see that signature. On HDC3022, you may see that RH drops slowly to a very low RH value when water gets on the filter instead of quickly dropping to 0%. 

Regards

-Alex Thompson

4300.log.txt

HI Alex  

The attachment is the log of my device entering and removing from the refrigerator. It can be seen that after the device is taken out for a period of time, there are small white water droplets on the PCB board, which then evaporate.
I did observe that at the moment when HDC3020 was blocked by water droplets, the relative humidity was 0%. But only for a brief moment. I'm worried that my device won't be able to capture this moment to make a judgment. Is it reliable to use this as a basis for judgment?
HDC3022 has a waterproof film, why is there still condensation blocking the hole? Why is the phenomenon different from HDC3020? I have tried soaking HC3022 directly in water for 24 hours, and the relative humidity reading remained at 99-100%.

Hello,

Alex is on paternity leave so I will try to help with this issue. 

On the HDC3020, we have consistently observed that: 

• in the presence of ~100%RH moisture before condensation, the sensor will read close to 100%RH and as saturation occurs there will be a sudden drop in the RH reading due to water drop formation on the polymer surface. 
• if water drops are on the sensor itself, there will be a sudden drop in RH reading to 0%RH, and then a sudden increase in RH reading when the water regresses from the polymer surface. 

We haven't done condensation testing with the HDC3022 yet, so we would need to check if it behaves in the same way or not. The HDC3022 has a IP67 filter for dust and condensation protection, so that doesn't mean condensation won't happen but the filter should protect the polymer in this case. 

If condensation is detected, you can run the heater to get rid of the condensation. The duration of heating will be dependent on your system parameters (the board thickness and layout, temperature, etc) so you may need to do some experimentation to determine the optimal amount of time. I would recommend running the heater at max power. 

Best regards,

Sakeenah Khan

So what should I do in the current situation (taking it out of the cold storage)?
When should I turn on the heater? Add a threshold for relative humidity? What is the threshold for this?

Does this heater have a thermal shutdown function? If the MCU controlling HDC3022 crashes during the heating process, no one can turn off the heater.

Hello,

Regarding the data log you shared previously, it looks like the device was removed from cold storage and the RH measurement suddenly jumped up close to 100%RH, then leveled off to ambient humidity. Did condensation occur in this case as well? Did any water drops occur on the sensor? Also, have you consistently observed this trend, or do you sometimes see the RH measurement drop to 0%RH? 

Assuming the HDC3022 data you shared holds true for all cases, you could use a rate-of-change threshold to detect if the RH measurement is rapidly increasing and condensation is likely to occur. Then the heater can be turned on accordingly. When your device was taken out of cold storage, the RH measurement suddenly jumped over the course of 38 sec, with a slope of +0.89%RH/sec. You could use a rate-of-change threshold around 0.7-0.8%RH/sec. You might need to do a few experiments/tests though to make sure this is an optimal threshold. Keep in mind the RH response differ depending on the ambient temperature/humidity conditions (if the ambient conditions are not controlled/indoor). 

Another option: if condensation consistently occurs whenever you remove the device from cold storage, then you can opt to turn the heater on when the temperature increases. Monitor the HDC3022 temperature reading for a sudden increase in temperature and then enable the heater. You can use a rate-of-change threshold again in this case. Looking at your data again, you could use a threshold of 0.4-0.5C/sec, but again it's better to test that out and the response could differ depending on ambient temperature. 

The integrated heater can be turned on/off using the enable/disable i2c command sequence, detailed in datasheet section 7.5.7.7.

Best regards,

Sakeenah Khan

HI 

1. Condensation occurred, but the amount of water was relatively small, and there was a layer of small white water droplets on the PCB board

2. There are also small water droplets on the sensor

3.During the experiment, no humidity reaching 0% was found. I once immersed HDC3022 directly in water for 23 hours, and during this process, I did not observe the humidity becoming 0%.

But if HDC3020 is soaked in water, the relative humidity will briefly drop to 0%. What is the reason for the relative humidity becoming 0%? What principle is this?

4. Is the method of using a change rate threshold to determine whether condensation has occurred reliable? Is there such a precedent? Why can't we directly use the threshold of relative humidity to determine the generation of condensation water? For example, if the relative humidity exceeds 99%, it is judged that condensation has occurred.

Best regards,

Vincent Tang

HI Sakeenah

Thank you for your response

My goal is that HDC3022 can work normally all the time, even if condensation affects its function, it can be eliminated by self heating, and then continue to work normally.

The picture you sent is crashed，like this --》。Could you post this picture again？

Best regards,

Vincent Tang

Hi Victor,

I tried to attach the .png image again, I hope you can see it. 

Best regards,

Sakeenah Khan