Some notes on general filter 'sack' design, if you can imagine the drawing as a sheet of filter media #IAQ

@WilliamRamsey The fan mount plate could be extended enough to where two rubber straps form high areas and a U shape in the metal allows an external rubber strap to clamp it securely in place. Another external rubber strap could be used to keep it air tight.

The two internal rubber straps would be before and after the U in the metal so the external rubber strap can go into the U section. The internal rubber straps raise the filter medium so the external rubber strap can pull a better seal without putting excessive stress on the medium. It's also going to be reusable when fitting another bag. A similar approach is used to form water-resistant seals in electronics.

I would also recommend an internal wireframe to support the bag so it performs better in either push or pull configuration. The fan shroud could also use rubber straps to allow for a quick change bag system. A plug and play style that integrates the support wireframe with the metal connector that mates to the fan shroud. A swap would just need a trash bag to go over the bag frame, kill the fans, remove the strap, pull it into the trash bag, push the new filter on, reattach bans and turn on the fans.

The metal frame can be taken outside, removed from the old filter medium, thoroughly cleaned, dried and new filter fitted. It's like a K&N air filter but without the long cleaning and oiling procedure or having to replace or clean your MAF sensor.

I saw your cylindrical filter post. I believe that a conical filter would provide better results. To put it in perspective, it would resemble your bag filter where more uniform negative or positive pressures are allowed. The majority of the air will come from the rear and middle sections with higher pressures allowing the area closer to the front to pass air.

Yes, it's interesting but I don't know if those are made for anything but vehicles. It seems obvious that a large cylinder would restrict airflow more due to higher internal volume and uniform shape despite internal pressure differences. Conical filters make more horsepower and tomatoes if you catch my drift. It just seems like lazy engineering. The only cylindrical filter used in motorized equipment I know of is on the Caterpillars used in quarries. The reason why is because they need to breathe even after the more favored sections are partially clogged. Those are also very large filters that need to be replaced in 200hr of operation or more in dusty environments.

@AmpBenzScientist Thanks for the replies, I appreciate the feedback! Here are a couple examples of framing - a mini plant cage, and a hanger. I think If designers could have some general confidence in fan performance as a function of surface area, and access to factory rolls of filter media, they could offer more integrative solutions. To that, some valve engineer put together a spreadsheet to estimate performance. I don't have link to his wiki but here's a copy:
https://drive.google.com/file/d/10876l7wQnEMgXvk6rNdkifDWkJxTmRru/view?usp=drivesdk

@WilliamRamsey I saw the plate and immediately thought it was for a server rack cabinet (and there's no reason why that would be a bad choice). I saw the plant portion later and realized that yes, it is a good option for pest prevention. A server rack cabinet could be a mostly closed system that doesn't need to use pesticides.

For the power supply, a rack mounted server will have dual PSUs with absurd efficiency ratings. The 12v could easily be tapped from these especially if it is a used rack server. They could be separated so humidity and exhaust air exit each chamber. It's an option if one is willing to take it. A good condition rack server is rather inexpensive for the performance and it could justify the cabinet itself. Solid metal racks can hold a lot of weight and containment chambers would be rather easy to construct. As long as humidity is monitored, it would work out well for the server too.