Based on the results from my last test, I decided to adjust the sensor layout inside the dry box to better support automated heating control. This time, I moved the SONOFF TH Elite sensor to the area near the PTC outlet, so it can directly monitor the outlet temperature and serve as the trigger for heating cycles.
In addition, I placed two SNZB-02LD sensors inside the box: one with its metal probe inserted into the center area of the filament spool, and the other mounted near the top of the enclosure. This arrangement gives me a clearer picture of how heat spreads from the PTC outlet to the rest of the sealed box over time.
To avoid having to constantly watch the readings manually, and to support a longer-term temperature control experiment, I connected both the TH Elite and SNZB-02LD sensors to my local iHost platform. I then created a dedicated CAST monitoring page, which displays live temperature graphs from each sensor for easy observation.
The results from the chart are quite clear—and very promising. I initially set the TH Elite to turn off the PTC at 55°C, and turn it back on when the temperature drops to 50°C. The system started running around 9 AM and continued until about 3–4 PM, a total of roughly six hours of continuous operation.
The temperature graph from the TH Elite shows that the device consistently maintained the outlet temperature between 50°C and 55°C, as configured.
At the same time, both SNZB-02LD sensors—one at the filament center and one near the top – reported temperatures holding steady around 50°C. This confirms that the current control logic effectively maintains a stable internal environment around the target temperature for drying.
To further verify performance at higher temperatures, I updated the TH Elite’s automation logic to switch off the PTC at 65°C and turn it back on when the temperature dropped below 60°C. I then let the system continue running under this new setting for over five more hours.
From the new set of temperature graphs, it’s clear that the two SNZB-02LD sensors consistently reported readings right around 60°C during the entire period.
This confirms that even in this higher range, the enclosure maintained stable and reliable thermal control. The overall performance looks excellent.
Just when I was feeling satisfied with the results, I ran into a frustrating realization. The TH Elite sensor, as currently positioned, is mounted above the turbine fan, which means it’s not directly in the airflow path of the PTC heater.
As a result, its readings are lower than the actual outlet temperature, and the system has been underestimating how hot the air from the PTC really is.
Although the temperature charts show the TH Elite holding within the desired range, the actual PTC output was likely far higher. This became clear when I noticed that the PETG-printed base structure had deformed – a sign that the airflow temperature had gone well beyond safe levels for that material.
Even though the overall temperature control inside the sealed box worked beautifully, the fact that the 50W PTC heater still caused deformation in the bottom structure means that the current dry box design is still not practically viable. If the structural components can’t withstand the localized heat,
then the whole system will eventually fail.
Now I see two main paths forward: either I use a lower-wattage PTC module
and rely on longer, staged heating to gradually raise the temperature, or I must completely redesign the PTC installation so that its high-temperature airflow never touches any 3D-printed parts or filament.
Leave a Reply