Box Fan Filter Adapter aka the Hamster Thunderdome

Before continuing there are a couple of questions you need to answer yes to and be honest it's just between you and me.

Do you have a room with some dirty air that needs to be filtered?

Are you a 3D printing YouTube celebrity with a wall of unused free filament or scored a deal on some filament that required a pallet to be delivered? 

Have 100+ hours to kill on a highspeed printer for such stockpile of filament?

A spouse/significant other who won't roll their eyes at thinking you are crazy but just thinks yep that's on-brand for you?

A 20" box fan collecting dust (likely from the dirty air in said room above)?

Have a handful of M5x16 BHCS and a sprinkling of M5x35 BHCS laying around?

Have an extra roll of foam tape and a roll of metal foil ducting tape?

Some scrap wood and screws if you plan on cleaning said room from up on the wall?

Have the ability to ignore the voice in the back your head saying I…probably…shouldn't…print…thissss…?

If you answered yes to all those and want to embark on a journey of printing what my wife calls the Thunderdome for Hamsters then by all means lets begin.

This all started 2 months ago when my new years resolution was to give my lungs some lovin and keep my garage workshop a little cleaner. You see I tinker with a lot of crap in my garage from sophomoric attempts at woodworking, worse attempts at metal fabrication, auto repairs and a steadfast determination on learning to make tig welds that I don't want to immediately hide. My first step was a dust collector with a canister filter. Then I needed some air filtration in my garage for everything else and don't have the space for a Corsi-Rosenthal setup on the floor. I knew I also wanted more than a 20x20x1" air filter so I wasn't choking the airflow of my newly ordered Air King fan. I also didn't want to replace filters every time I walked into the garage. My next progression was well what about 20x20x4" filters they would allow for better airflow. Well turns out they exist but are not common around here for me. They seemed more expensive than the 20x25x4" ones I saw. Well as it turns out those 20x25x4 are a pretty standard/common size and on a whim I grabbed one while at the local home store. So this journey began with a fan, a filter ,mediocre planning and some boxes off brand filament sitting in a closet that I scored a good Amazon price on before the holidays.

My initial plan was to design this in Fusion360 and make the parts from wood actually but remember what my comment said in the previous paragraph. Seriously though my first wood prototype wasn't horrible but what I didn't like was the weight of it as I got close to finishing it. It was turning into heavier than I wanted dangling off the wall. So I modified my plans in Fusion to accommodate a 3D printed design. Export the first corner and put the on my slicer buildplate to gauge how much time and filament would be needed. This was my first calculation mistake. The slicer came back with like 170m of filament and being late at night I went yeah OK there's a 1000m of filament on a normal spool so I'll only need 2 spools give or take, awesome. Yeah for that 1 reader making it this far, it's 1000 grams not meters on a spool and depending on who you read there is only about 300-330m on a normal spool. I loaded a brand new spool and started. When my print finished and I was queuing up the next one I glanced and my recently new spool to see it was quite low, like no chance it would do another one low. I busted out my abacus and slid around some heaven and earth beads to figure out my folly. Insert forehead smack emoji. Being persistent (nice way to say stubborn?) I sliced up a smaller piece and soldiered on. Well 6 spools and about 150 print hours later at 200mm/s on my Voron 2.4 I got my working model. I had some extra printing in there because of a couple failed prints that took some diagnosing. Turns out nearly 100 continuous hours printing at 200mm/s stresses things and I broke a silicone stepper wire somewhere inside one of the wires of the cable chain.

The printed parts were designed with M5 threaded holes to assemble and use M5x16 BHCS. Aside from foil tape and mounting plate that's all mine is using and it feels pretty solid. After it's assembled I used some leftover thick adhesive foam tape on the small side of the Thunderdome errrr adapter. Now by the time I got to mounting it up I was exhausted with this project. I removed the M5 screws on one side and took a scrap piece of plywood, clamping it in place and from the inside carefully used the M5 holes and a template into the wood. I then ditched 9 of the M5x16 BHCS and replaced them with M5x35 BHCS this time through the wood into the molded threads. Wrapped the seams with foil tape to ensure minimal leaking then for the first time in my life made a French cleat about 20 inches long. I attached one some to the plywood on the fan and yep you guessed it the other on the wall. So far so good, it moves a good amount of air and is quieter than I thought it would be. I'm happy with how it turned out but in retrospect I should have listened to that voice in the back of my head and built a Corsi-Rosenthal variant that could mount to the wall with 20" filters. Would have been a lot faster, cheaper and easier. What can I say I'm “persistent”.

I printed at .30 layer height, 3 perimeters, 3 top/bottom solid layers, 10% rectilinear infill, no supports or brim all in PLA @ 200mm/s. With those settings the slicer estimated a little over 9 hours on the corners and reality was about 8-½. I attached all the STL parts for simplicity but some are duplicates like corner 1 & 4 are the same etc. I've also included the STEP file for those who want to make tweaks.