---

---

---

Drive to Build a Cheaper Compact Exhaust Fan

Issues With Current Fans

I have made quite a few electronics now and did nearly all of it either in a shed back in undergrad or my apartment office in graduate school. At no point during this time did I have an exhaust fan but I wanted one since the solder fumes are not exactly good for you. Also I could not exactly solder on my small 5x5 balcony as I lacked both a plug to do so and the space. Commercial options such as the Hakko Fa400 at nearly $80 are northing more than a PC fan and a charcoal filter which has been shown to not do a whole lot. The more expensive fa-430 that includes a HEPA filter is nearly $800. So for the time being I bought a large amount of charcoal filters and threw them in front of a PC fan and mounted that on a machining coolant hose goose neck with copper wire core for rigidity based off of Adam Savages light build and you can see it below.

It was built using a Chinese desk clamp and Plastic Coolant Line and then a standard 12v DC jack with a wire running up the center of the coolant tube to the fan. As nice as this was I wanted something a bit more with PWM control of fan and if its going to be over my work surface might as well have some LEDs as well. So I designed a rather large board with way too many LEDs and originally I relied on the voltage drop of the LEDs to regulate current along with a PWM transistor. It contained a atmega328p to control the PWM by reading the analog value of a potentiometer. However I decided to use chinese LEDs and the "heat pad" in the datasheet turned out to be tied to the positive side of the LED meaning the first board was shorted and the arudino didn't have all its pins connected properly which was my oversight.

Version two fixed the LED issue and I switch to a AP3019A for current regulation and I fixed the atmega pins however I used too small a inductor for the AP3019a not looking at the datasheet and had LED lights that didn't current regulate (they always stayed on the lower side as the inductor didn't have enough power)

Enter V3, it currently is still using the atmega but I will likely switch to a 555 timer soon as it means no programming is required

I may have over done the lights a bit as combined this outputs around 3000 lumens of light at full power but I want my work surface illuminated. This board will slot into the front of a 3d printer enclosure I designed that will connect to the same gooseneck, it will house charcoal filters behind the fan and then a hepa behind that.