3d printer filament dryer

in summer 2024, i built a 3D printer filament dryer. filament which has absorbed moisture produces poor quality prints and results in failures. at the time, many drying boxes on the market were expensive and the more economical ones performed poorly due to a lack of exhaust, so i wanted to build my own dryer to solve these issues.

my goal was to create a low-cost dryer which can maintain a temperature of 60°C, sufficient to dry some basic filaments including petg and tpu. to provide the thermal energy, i considered resistive heating elements, infrared heaters, and peltier (thermoelectric) modules. i landed on a peltier with the idea of using it as a heat pump, with the hot side facing inward to leverage the colder surroundings.

i used waterjet polycarbonate for the walls and abs 3d printed components for the peltier support, rollers and brackets. the primary criterion in material selection was to ensure that the heat deflection temperature of all the components was greater than 60°C. off-the-shelf components included 608 bearings due to their availability, since this is a low-load application. additionally, a computer fan for humid air evacuation, and an aluminum heatsink which i modified to integrate with the system. I chose aluminum over copper since it provided 2/3 of the thermal conductivity for 1/4 of the price.

for the electronics, i used an arduino uno, relays, a dht22 temperature/humidity sensor, the peltier, an external 12v power supply, and an lcd display for the current and setpoint temperatures.

i implemented bang-bang control with a +/- 2 degree threshold from the setpoint, since i wasn't expecting rapid temperature fluctuations. i implemented the setpoint target with two buttons and added a display timeout to show the real-time temperature and humidity.

unfortunately, the project was a flop. it was a case of "you don't know what you don't know". there were a few issues stemming from both a lack of physical knowledge and intuition, and not knowing "how things are done". issues included not understanding the thermals and the power required to heat up such a volume of air given the poor insulation, not understanding the peltier, its inefficiency and the requirement of cooling the cold side (i cooked the module i bought in this way), not understanding the insulation requirements for containing the dry air, and not understanding how electronics would be much cleaner on a simple custom board.

there's not much motivation to revisit the project since cheap drying boxes have gotten better in the past couple of years, but if i were to revisit, i would more clearly define my strategy and requirements. for example, is this an active high-temperature drying box, or a low-humidity long term storage box? i'd appropriately size thermals and not cook my components, and shrink the messy electronics into a custom pcb.

it's true that you don't know what you don't know, but you won't learn without doing, which is precisely what i gained from this project. that the low efficiency of the peltier would have me better off sticking the whole module inside and using as a normal resistive heater. that a bare arduino and jumpers are great for prototyping, but terrible for productionizing. that insulation and boundary conditions play a huge part in maintaining temperature.

despite the fact this project was a failure on paper, the learning and growth were far more valuable than any physical outcome.