For tests 1-3 (22°C, 45-50%RH), multiple data points were analyzed, including relative humidity, absolute humidity, dew point, RH slew rate, and AH slew rate, across varied time windows of 1, 5, and 10 seconds. Absolute humidity and dew point were calculated using the following equations, where AH is the absolute humidity (in g/m³), T_d is the dew point (in °C), RH is the HDC3020 relative humidity measurement (in %RH), and T is the HDC3020 temperature measurement (in °C). The equations are derived from the designed for gas equation and Magnus-Tetens formula, using ⎦values approximated by Alduchov and Eskridge (1996).

Data analysis showed that merely observing the RH, AH, or T_d does not offer as clear an indicator of a water ingress event as the RH slew rate does. Moreover, placing fixed thresholds on RH or AH for a semi-open system can cause false alarms, since ambient humidity fluctuations can also increase RH over time. Figure 5-1 and Figure 5-2 show the RH, AH, and T_d over time for test 2 (22°C, 45%RH, 0.07mL water).

Figure 5-1 Test 2 (22°C, 45%RH, 0.07mL water): Relative Humidity and Absolute Humidity vs. Time

Figure 5-2 Test 2 (22°C, 45%RH, 0.07mL Water): Dew Point vs. Time

The best indicator for the ingress event, especially for smaller volumes of water, was the RH 10-second slew rate. This is because the RH slew rate quickly and dramatically increases upon the introduction of water to the system, as shown in Figure 5-3. The AH slew rate data is too noisy to effectively and reliably indicate an ingress event, as shown in Figure 5-4. In addition, this is better to use a 5 or 10 second window. Slew rate calculations across a smaller time window are dominated by noise, making this difficult to distinguish a true water ingress event, as illustrated in Figure 5-3 and Figure 5-4. Using a 10-second window allows to eliminate noise effects and distinguish the water ingress event from normal humidity fluctuations.

Figure 5-3 Test 2 (22°C, 45%RH, 0.07mL water): RH Slew Rate Calculated Across Different Time Windows

Figure 5-4 Test 2 (22°C, 45%RH, 0.07mL water): AH Slew Rate Calculated Across Different Time Windows

Furthermore, the test results confirmed that a greater volume of water induces a larger change in RH, resulting in a larger peak RH slew rate. Figure 5-5 and Figure 5-6 compare the response for all 3 water volume tests at indoor ambient conditions, where the ingress events have been normalized to occur at time = 0. The full flow (100mL) and 3 drop (0.07mL) events are easily distinguishable using the RH 10-second slew rate. The one drop (0.023mL) event also had a noticeable increase in RH slew rate, however the data was noisier. Thus, this is difficult to reliably distinguish a single-drop ingress event from normal ambient humidity fluctuations using an RH slew rate threshold, as shown in Figure 5-6. This shows that the slew rate threshold can be used to detect small volumes of water ingress ≤0.07mL, however there can be limitations in detecting even smaller water volumes.

Figure 5-5 Test 3 (22°C, 50%RH), Full Flow (100mL) Test

Figure 5-6 Tests 1-2 (22°C, 45%RH), 1 Drop (0.023mL) and 3 Drop (0.07mL) Tests

The full flow ingress event was also repeated at indoor ambient temperatures and elevated humidity levels of 70%RH. Figure 5-7 shows the response for the full flow ingress event at 70%RH, where the ingress event have been normalized to occur at time = 0. At higher RH, the change in RH slew rate due to the ingress event is less pronounced but still comfortably exceeded the 10m%RH/s threshold, and is thus distinguishable from typical ambient fluctuations.

Figure 5-7 Test 4 (22°C, 70%RH), Full Flow (100mL) Test