A DIY Box Fan Air Filter for the Room Your Animals Are In: The Build, the Measured Numbers, and Where to Put It

The smoke came in three days ago, the air purifier you meant to buy is out of stock everywhere within driving distance, and the animals are shut in the back bedroom because that is the room with the fewest windows. There is a box fan in the garage. There is a hardware store still open. Somewhere on the internet somebody has told you that a furnace filter taped to that fan will clean the room.

That is broadly true, it is published by a federal agency, and the numbers behind it are better than most people expect. It is also the option EPA ranks last of three, for reasons it states plainly, and almost everything published about it was written about a room with people in it rather than a room with a dog in it.

This page is about the build and the operation. What to buy, how to assemble it, what each configuration actually delivered in a laboratory, how to size it against the room, how often the filter has to change, how loud it is, how much power it draws, where to stand it when an animal shares the floor with it, and the specific things it does not do. Where a source stops and this site starts reasoning, the page says so in those words.

What This Page Owns, and What Lives Elsewhere

Three neighboring pages already carry pieces of this, and repeating them here would only give you two versions of the same thing.

The trigger belongs to the threshold page. When to stop walking the dog, what a veterinary college published band by band, and why no source that page could locate publishes a number of minutes are all on what AQI is safe to walk my dog. That page is also where EPA’s box fan condition already appears on this site, quoted in full, attached to a veterinary college’s mention of a do-it-yourself setup. This page repeats the quotation deliberately, because a build page that omitted the safety condition attached to the device would be worse than a duplicate, and it flags below where the two pages are saying the same thing.

Buying a commercial unit belongs to the wildfire page. Wildfire smoke and pets carries the portable air cleaner comparison and the clean air delivery rate selector for commercial units sized to room square footage. If the answer to your problem is a purchase, that is the page, and EPA agrees it should be, for reasons in the next section.

The heat question belongs to the no-air-conditioning page. A heat emergency with no air conditioning works through the published fan guidance, including the fact that agencies disagree about the temperature above which a fan stops helping a person, and that a fan is aimed at a cooling route a dog barely uses. The fan on this page is a filtration device. It is not a cooling device for an animal, and nothing here should be read as one.

What is left, and what this page owns, is the object itself.

Before You Build: EPA’s Own Order of Preference

This is the part most build guides skip, and skipping it changes what the rest of the page means.

EPA’s Wildfires and Indoor Air Quality page, last updated August 6, 2026, states the ranking directly: “EPA recommends using DIY air cleaners only when products of known performance (such as commercially available portable air cleaners) are not available or affordable. DIY air cleaners are not recommended as a permanent alternative to these products.”

The reason follows in the same paragraph, and it is worth reading whole rather than in the encouraging half:

“Several studies indicate that well-built DIY air cleaners can perform similarly to commercial portable air cleaners in reducing airborne particles such as those in wildfire smoke. However, their performance depends on their design and how well they are put together. Each DIY air cleaner may perform differently, even after reassembling the same unit when changing filters. On the other hand, commercial air cleaners are more standardized from unit to unit, so performance testing of these products is more reliable.”

Note what that says about you specifically. Not only can two builds differ, but the same build can differ from itself after you change a filter. There is no serial number on the thing you are about to make and no test report for it.

EPA’s Indoor Air Filtration fact sheet, EPA-452/F-26-006, encodes the same ranking as a table. Its row asking when each of three options is a good choice gives, for the do-it-yourself column, one sentence: “Portable air cleaners are not available or affordable.” EPA’s cleaner air room fact sheet, EPA-452/F-26-005, leads its own do-it-yourself bullets with the same instruction: “Get a portable air cleaner if you can. DIY air cleaner performance can be more variable depending on how they are built and maintained, and they are often louder than commercial air cleaners.”

And here is the finding a pet-owning reader should have in front of them before building anything. EPA publishes a fact sheet specifically about animals, Protect Your Pets From Wildfire Smoke, EPA-452/F-26-007. Read whole on August 15, 2026, that document tells owners to “Plan to keep your home cool and filter the indoor air. Buy a portable air cleaner and high-efficiency HVAC filters if they are compatible with your central air system”, and to “Think about creating a cleaner air room in your house with a portable air cleaner.” The words box fan and do-it-yourself do not appear anywhere in it.

So nothing on this page should be read as EPA recommending this build for animals. EPA recommends buying a portable air cleaner for animals. What EPA publishes about the do-it-yourself build is general household guidance, written for people, that a household with animals in it can use when the recommended option is not on the shelf. That distinction survives every section below.

The counterweight, in EPA’s own framing, is that the build is genuinely worth having when the alternative is nothing. EPA’s cleaner air room fact sheet describes it as “an effective, less expensive option to clean the air when portable air cleaners are not available or affordable.” That is the situation this page is written for.

The Build, From EPA’s Own Two-Page Infographic

EPA publishes build instructions as a two-page infographic titled DIY Air Cleaner to Reduce Wildfire Smoke Indoors. Everything in this section is from that document, from the fact sheets, or from the California Air Resources Board’s own build steps, quoted or described as such. Nothing here is a design this site invented.

The basic design

EPA’s Basic Design page lists three materials: a 20 by 20 inch box fan, a 20 by 20 inch air filter that is either 1 or 4 inches deep with a suggested rating of MERV 13, and one of three fasteners, listed as clamps, duct tape or bungee cords. Under the fan graphic it carries the condition that travels with this whole build: “Only use certified fans with UL or ETL marking (2012 model or newer)”.

Its assembly steps are three lines long. Step one: “Attach the air filter to the back of the box fan using either clamps, duct tape or bungee cords.” Step two: “Check the filter for the direction of the air flow (marked on the side of the filter).” Step three: “Replace filters when dirty.”

The fact sheets add the parts of the technique that decide whether it works. EPA’s cleaner air room fact sheet states: “When assembling a DIY air cleaner, use a high-efficiency filter(s)—preferably rated MERV 13 or higher—for better filtration. Align the arrows on the filter(s) with the direction of the air flow through the fan. Try to get a good seal between the fan and the filter. Change the filter when it appears dirty or starts to smell.”

The California Air Resources Board publishes the same assembly with the arrow direction spelled out from the other side, and adds an instruction about the room that the EPA build pages leave to the cleaner air room sheet: “Attach a high-efficiency air filter (MERV 13 rating or higher) to the back of the fan using duct tape or a bungee cord, with the arrow printed on the filter pointing toward the fan (in the same direction as the airflow). Close all windows and doors when the box fan filter is being used. Change the air filter when it gets dirty.”

Two agencies, one arrow rule stated two ways that agree: the arrow points the way the air is going, which on a filter taped to the back of a fan means the arrow points at the fan.

On the filter rating, EPA’s research page explains why MERV 13 is the number everyone repeats: “The Minimum Efficiency Reporting Value (MERV) rating of the filter strongly impacts the Clean Air Delivery Rate (CADR). It is recommended to use a MERV 13 filter(s) to remove the very small particles in wildfire smoke.”

The three tiers EPA publishes

EPA’s second infographic page is headed DIY Air Cleaner Designs: Beyond the Basic, and it lays out three columns labeled Good, Better and Best.

Good is the basic supplies list: the 20 by 20 inch box fan, one 20 by 20 inch MERV 13 filter at either 1 or 4 inches deep, a 20 by 20 inch cardboard shroud described as a “(cutout the size of the fan blades)”, and clamps, duct tape or bungee cords.

Better adds, in EPA’s words, “Two - MERV 13 air filters” and a “Triangle cardboard cutout for base on top”.

Best adds “Four or five - MERV 13 air filters” and one instruction that is easy to miss and matters structurally: “If using five filter design, use leg supports (e.g., blocks) to allow airflow through bottom”.

The same page carries a Ways to Improve Effectiveness list that is really the summary of EPA’s whole research finding: add a cardboard shroud, described by EPA as a “(no-cost improvement)”; use thicker filters, 4 inch rather than 1 inch; and use multiple filters, in what EPA writes as “(2-5 filter designs)”. Its Key Reminders list repeats the fan condition, adds “Keep extra filters on hand”, and repeats “Replace filters when dirty”.

EPA’s Wildfires and Indoor Air Quality page describes the same family of designs in prose, and its sentence is the clearest short statement of the shape of the thing:

“There are many ways to put together a DIY air cleaner. Some common designs are to place one filter flat against the fan, two filters taped with cardboard to form a triangle against the fan, or four or five filters taped against the fan to form a cube. In general, DIY air cleaner designs that use more filters are more effective. Using thicker filters (2-4” rather than 1“ deep) or covering the outside corners of the front of the box fan (sometimes called a shroud) can also improve performance.“

The four or five filter cube is the design known as the Corsi-Rosenthal box. Its own originators describe how it came about, in the peer-reviewed paper that characterized it: “The Corsi-Rosenthal Box was originally proposed by Richard Corsi on Twitter and with Jim Rosenthal making the first prototype”.

The shroud, which is the free part

If you build only one improvement into the basic design, EPA’s own research points at this one. Its research page states: “The addition of a cardboard shroud increases the CADR by 40% without any change in the cost or physical footprint.”

A shroud is a piece of cardboard covering the corners of the fan’s outlet, leaving a round opening roughly the size of the swept area of the blades. EPA’s infographic describes cutting the opening “the size of the fan blades”. EPA’s Science Matters newsletter item states what that accomplishes: “DIY air cleaner performance could be improved by the addition of a no-cost cardboard shroud (see Figure 2) to block recirculating air flow past the fan blade tips.” The peer-reviewed paper describes its own version as a roughly circular shroud that “covers the box fan corners and prevents backflow of unfiltered air into the fan”, and notes the idea was proposed for square box fans by David Elfstrom.

That paper also records a detail about airflow direction that is worth understanding before you tape anything: on the cube design, “The box fan is oriented such that the fan blows out of the constructed filter box. This creates a slight negative pressure that may help to seal the box and limit leaks, although any persistent leaks from e.g., holes in the filters or the tape would be independent of the flow direction.”

What Each Design Actually Delivers

This is the table nobody hands you at the hardware store.

EPA measured the designs in a laboratory. Its research page explains the method in its own words: “The research was conducted in laboratories in Research Triangle Park, NC. Simulated wildfire smoke made from smoldering pine needles was used in a room-sized chamber to determine the Clean Air Delivery Rate (CADR) of the air cleaners.” It also defines what the metric is: “The CADR is a standard measure of how well a commercial air cleaner can remove particles of a specific size from a standardized room.”

Here is EPA’s published table. Both columns reproduce EPA’s own wording, including its inconsistent spacing of MERV13 and MERV 13.

Design	Wildfire smoke CADR
Box fan with one 1” MERV13 filter	111 ± 1
Box fan with one 1” MERV13 filter and cardboard shroud	156 ± 4
Box fan with one 4” MERV13 filter and cardboard shroud	248 ± 15
Box fan with two 1” MERV13 filters and cardboard shroud	263 ± 22
Corsi-Rosenthal Box with four 1” MERV 13 filters and cardboard shroud	401 ± 31

Source: US EPA Office of Research and Development, Research on DIY Air Cleaners to Reduce Wildfire Smoke Indoors, page states last updated on July 6, 2026, read August 15, 2026. Row labels, figures and plus-or-minus values are EPA’s, reproduced as written. The underlying study is Holder, Halliday and Virtaranta, Indoor Air, volume 32, issue 11, November 2022, which is how EPA cites it.

Read down that column and the practical lesson is stark. Going from the bare filter to the same filter with a piece of cardboard on it gains more than a third. Going from one 1 inch filter to a four-filter cube with a shroud is a difference of roughly three and a half times.

EPA also publishes the trade-off, because more filters is not free in every sense: “Increasing the number of filters used in the design increases the cost of the raw materials to construct the air cleaner, the complexity of the assembly, and the space the air cleaner takes up. However, designs with more filters are much more effective at removing particles and able to clean larger sized rooms more typical of the living spaces in many homes.”

And for the household that does not have floor space for a cube, EPA names the compromise explicitly: “While the most cost-effective designs are those with multiple filters, use of a single 4” MERV 13 filter is also highly effective (increased CADR by 123%) and may be more suitable for smaller areas with minimal floor space that cannot accommodate the multi-filter designs. Using multiple single-filter units in the same room is also worth considering, when balancing performance, costs, space, and ease of assembly for your specific needs.”

For a household where a cube on the floor is a hazard because of what lives on that floor, that sentence is the useful one. A slimmer single-filter unit with a shroud is a published design with a published number, not a downgrade you invented.

The second set of numbers, and why they must not be merged with the first

There is a second body of measurement, and it produced larger figures for the cube. It must be read separately, because it measured a different thing.

Dal Porto, Kunz, Pistochini, Corsi and Cappa published a characterization of a five-panel Corsi-Rosenthal air cleaner in Aerosol Science and Technology, volume 56, issue 6, pages 564 to 572. Their abstract reports that the size-integrated effective clean air delivery rate rises with fan speed, from about 600 to about 850 cubic feet per minute as determined with an aerodynamic particle sizer.

Four differences separate that from EPA’s 401 in the table above.

Different aerosol. EPA used simulated wildfire smoke from smoldering pine needles. The UC Davis team used salt particles generated by a nebulizer, and reports its results for particles larger than 0.5 microns with a number-weighted median particle diameter of 1.2 microns.

Different build. EPA’s row is a four-filter cube with 1 inch filters. The paper’s unit is a five-panel build using 2 inch filters.

Different setting. EPA used a room-sized chamber. The paper measured in two real rooms, which it describes as “a furnished but not occupied 5926 ft3 (168 m3) classroom and a 1277 ft3 (36.2 m3) furnished but not occupied home office”.

Different method ceiling. The paper states that its method “allows for determination of CADR values above the 450 ft3 min-1 (765 m3 h-1) upper-limit of the standard method”, which is a statement about why the standard test cannot report numbers this high at all.

Those are two honest measurements of related objects, not two estimates of one quantity. Presenting them as a single range, or writing that studies find somewhere between 400 and 850, would be manufacturing a figure neither group published. This page does not do that, and you should be skeptical of any page that does.

ASHRAE’s current Position Document on Filtration and Air Cleaning, approved by its board on June 26, 2024 and marked as expiring June 26, 2027, states the general form of the caution both sets of numbers deserve:

“In-situ effectiveness and safety of filtration and air-cleaning technologies are a strong function of air distribution, maintenance, aging, degradation impacts, component face air velocities, system airflow rates, installation, space volume, and other application and contextual factors. Accordingly, filtration and air-cleaning effectiveness and safety in a specific application should not be assumed equivalent to performance determined using controlled laboratory tests under specific conditions.”

One scope note on that document, because absence claims are easy to overstate. This specific position document, read whole on August 15, 2026, contains no do-it-yourself, box fan or Corsi-Rosenthal guidance. That is a statement about this document, not about everything ASHRAE has ever published. The nearest thing to an ASHRAE view on the build that surfaced in this research is a remark attributed by the volunteer site Clean Air Crew to Bill Bahnfleth, then chair of ASHRAE’s Epidemic Task Force, on December 10, 2021: “The CR box is not specifically referenced in ASHRAE guidance, but they are completely consistent with guidance about upgrading filters and using portable air cleaners to improve clean air delivery rate where needed. You can certainly quote me on that.” That is one named individual’s quoted remark relayed by an aggregator, not an ASHRAE position, and it is included here because it happens to match what reading the position document shows.

Sizing It to the Room the Animals Are In

Three bodies publish a rule connecting clean air delivery rate to room size, and they are compatible once you notice that they are answering slightly different questions.

Source (read 2026-08-15)	What it publishes about sizing	What it is about
AHAM Verifide, AHAM’s Air Filtration Standards	“As a rule of thumb, the CADR of your air cleaner should be equal to at least two-thirds of the room’s area.” And: “For wild fire smoke, AHAM recommends a Smoke CADR equal to the size of the room in square feet.” Also: “If your ceilings are higher than 8 feet, an air cleaner rated for a larger room will be necessary.”	Air cleaners generally, with a separate stricter rule for wildfire smoke. AHAM’s own page writes wild fire as two words
US EPA, cleaner air room fact sheet EPA-452/F-26-005	“Choose a device with a clean air delivery rate (CADR) for tobacco smoke that is at least two-thirds of the room area in square feet.”	Written in the fact sheet’s portable air cleaner option, not in its DIY option
US EPA, Indoor Air Filtration fact sheet EPA-452/F-26-006	“Choose a PAC with a clean air delivery rate (CADR) for smoke that is at least two-thirds of the room area in square feet.”	Written in the portable air cleaner column of the three-option table. The DIY column carries no room-size rule at all
US EPA, DIY air cleaner research page	“The Association for Home Appliance Manufacturers suggest using an air cleaner with a CADR equal to the square footage of the room to be cleaned.” (EPA’s own wording, reproduced as written)	Written in the answer to whether DIY air cleaners are effective, immediately before EPA’s CADR table

Notice which cell is empty. Neither EPA fact sheet attaches a room-size rule to the do-it-yourself option. The only place EPA puts a sizing rule next to the DIY numbers is its research page, and there it is attributing the rule to AHAM rather than issuing it.

Here is the step this page takes, and it is the site’s own reasoning rather than a published instruction. If you want to know roughly how large a room a given design suits, the only arithmetic available is to take EPA’s chamber figure for that design and hold it against AHAM’s wildfire smoke rule, which asks for a smoke clean air delivery rate equal to the room’s floor area in square feet. On that basis EPA’s four-filter cube figure of 401 lines up with a room in the low hundreds of square feet, and the bare single filter figure of 111 lines up with something much smaller. We are not going to publish a per-design square-footage table off the back of that, because three separate things would have to be assumed to make one: that a chamber figure transfers to your room, that AHAM’s rule for tested commercial units transfers to a home build, and that your build performs like EPA’s. EPA states that each build can differ, ASHRAE states that laboratory performance should not be assumed equivalent in a specific application, and AHAM’s rule assumes an eight-foot ceiling by its own note. Three assumptions stacked produces a number that looks like a measurement and is not one.

What survives the caution, and is worth acting on:

• More filters and a shroud beat fewer filters and no shroud, by a wide margin, in the one controlled comparison a federal agency has published. That ranking is robust even if the absolute figures do not transfer.
• The room-size rule that does exist is stricter during smoke than at other times. AHAM’s own page gives two-thirds as the everyday rule of thumb and equal-to-square-footage for wildfire smoke. Use the stricter one when smoke is why you are building this.
• A tall room needs more than its floor area suggests. AHAM’s own note about ceilings above eight feet is the reason.
• Two units in one room is a published option, not a fudge. EPA’s research page states: “Using multiple single-filter units in the same room is also worth considering, when balancing performance, costs, space, and ease of assembly for your specific needs.” It also states, in a passage quoted whole in the next section: “You may benefit from having more than one air cleaner if smoke is very thick or your home is not well sealed from the outdoor environment. However, if you can designate one room in your home as a cleaner air room that people spend time in when it is smoky out, then one air cleaner may be sufficient.”

If you are choosing a commercial unit instead, the clean air delivery rate selector on our wildfire smoke and pets page does that job with manufacturer-published figures, and this page does not duplicate it.

The room EPA had in mind

One more sentence from EPA’s research page is worth reading slowly, because of who is in the room:

“You may benefit from having more than one air cleaner if smoke is very thick or your home is not well sealed from the outdoor environment. However, if you can designate one room in your home as a cleaner air room that people spend time in when it is smoky out, then one air cleaner may be sufficient. Some people have found it to be effective to move one air cleaner from room to room.”

A room that people spend time in. EPA’s cleaner air room fact sheet defines the room the same way in its first step: “Pick a room big enough for everyone in your household to be comfortable. A bedroom with an attached bathroom is a good choice.”

For a lot of pet-owning households those are not the same room. The dog is crated in the utility room, the cat has claimed the spare bedroom, the parrot’s cage is in the living room because that is where the light is, and the humans are in a fourth room. EPA’s guidance does not address that split, and this page will not pretend it does. What follows from the sources without invention is that one unit cleans one room, that EPA names moving a single unit between rooms as something people have found effective, and that a household with animals in more than one room is choosing which room gets the filtration rather than covering the house. Making that choice for your own household is not something a page can do for you, and it is worth making it deliberately before the smoke arrives rather than at midnight during it.

EPA’s pet fact sheet does give one relocation instruction that helps here: “Bring outdoor pets into a less smoky, cooler environment, like a utility room, garage, or bathroom.” Note that a garage carries a separate carbon monoxide hazard that a bedroom does not, which our generators, carbon monoxide and pets page covers in full.

The 2012 Fan Condition, Quoted From Both Agencies

This is the single most-repeated claim about this build and the one most often stated wrongly.

What is actually published

EPA, in the cleaner air room fact sheet EPA-452/F-26-005, reads in full:

“Use a newer box fan (made since 2012) with a UL or ETL logo. These fans have thermal safety fuses that reduce the risk of fires or burns from overheating the fan’s motor and outer surfaces. Look for fans that have a listing for electrical safety and have a mark from a certifying organization, such as UL or Intertek (the label will show the UL logo or Intertek’s ETL logo, for example). If you must use an older fan, never leave the fan running unattended while you are away or sleeping.”

That last sentence is the one that disappears when this rule gets summarized, and it is the sentence that tells you EPA is issuing a recommendation with a fallback rather than a prohibition. Our own AQI threshold page publishes that same passage, attached to a veterinary college’s mention of a do-it-yourself setup, and the two pages are quoting the same fact sheet on purpose.

EPA, in the Indoor Air Filtration fact sheet EPA-452/F-26-006, compresses it into one line in the do-it-yourself column of its comparison table: “Use a box fan made since 2012 with a UL or ETL logo to reduce the risk of overheating.”

EPA, on its research page, words the older-fan half more strongly than the fact sheets do: “If you build a DIY air cleaner, use a newer model box fan (2012 or later). The newer models have added safety features. Fans built prior to 2012 were not tested and pose known fire risks.” It adds a verification instruction the fact sheets do not: “Use fans that have been verified by an accredited third party to meet the UL 507 safety standard for electric fans or equivalent. To find a verified fan, look for one with a UL or ETL safety marking.” And it states EPA’s position on older fans as a recommendation with the same fallback: “EPA does not recommend using DIY air cleaners built with older model box fans (before 2012). If they are used, they should not be used unattended or while sleeping.”

EPA, on its Wildfires and Indoor Air Quality page, hedges the risk differently from its own research page: “Use a newer model box fan (2012 or later) and look for one with a UL (Underwriters Laboratory) or ETL (Intertek) safety marking. These newer models have added safety features. Fans built before 2012 may pose fire risks. If you must use a fan built before 2012, do not leave it unattended or use it while sleeping.”

Those two EPA phrasings are not identical. The research page says pre-2012 fans “pose known fire risks.” The indoor air quality page says they “may pose fire risks.” Both are EPA. This page quotes each to the page it came from and does not blend them, because blending them would produce a sentence EPA has not written.

The California Air Resources Board publishes the date independently, in a fact sheet page that carries the date August 20, 2021: “Only box fans manufactured in or after 2012 should be used to construct a DIY air cleaner – these fans will have a fused plug, which will prevent electrical fires if the device is knocked over.” It carries the same fallback: “If an older fan is used it should never be left unattended or operated while the occupant is sleeping.”

EPA’s Science Matters newsletter item, published September 6, 2023, names both safety features in one sentence: the team “identified several best practices to safely construct and operate a DIY air cleaner, including using a newer box fan (2012 model or newer) equipped with safety features like fused plugs and thermal cutoffs.”

What is not published, as far as this page could reach

Two agencies publish the 2012 date. Neither publishes it as a legal requirement, and this page located no source establishing one.

Specifically: we found no regulation, no dated revision of a safety standard, and no product recall establishing a mandate that took effect in 2012. Searching the full text of the UL Research Institutes report that EPA relies on for the safety findings returns no occurrence of 2012 at all. UL 507 itself, the electric fan safety standard both EPA and the report name, is sold rather than published openly, so its edition and revision history could not be read from here at all. That last point is a limit on our retrieval, not a finding about the standard. What can be said is what the agencies say: this is a recommendation, published twice, with a stated fallback for an older fan, and the word requirement should not be attached to it.

One document that EPA names does speak to the date, and it is worth having. EPA’s research page sends readers who ask how to build one to the Missoula City-County Public Health Department, a partner in its Wildfire ASPIRE Study, and that link resolves to the department’s Montana Wildfire Smoke site. Its do-it-yourself fan filter page, read August 15, 2026, carries a warning of its own and then the only sentence located here that connects the date to what is actually on the shelves: “Use a newer fan equipped with a fused plug. These are not 3 prong, just slightly larger. All fans sold in the US after 2012 will have this.” That is a county and city health department’s statement about what is on the shelves, not a citation of a standard. It names no regulation, no standard and no revision date, so it does not change the paragraph above. It does tell you what to look at when you pick up a fan, and the same page adds that “Using older box fans is not recommended as they are more likely to catch fire.”

Two other documents that might have spoken to the date or to the build could not be opened from here on August 15, 2026, and their contents are therefore unknown to this page rather than absent from the world. A CDC wildfire air filtration PDF returned an HTTP 403. The National Collaborating Centre for Environmental Health review of do-it-yourself air cleaners, which EPA lists in its own references, returned an HTTP 403 as well. Separately, the peer-reviewed study behind EPA’s CADR table, Holder, Halliday and Virtaranta 2022, was verified only by its bibliographic record; its own sentences were not read, so this page cites those figures the way EPA’s page presents them and does not quote the paper.

Why the mechanism matters more in a house with animals in it

Three of those sources name a physical mechanism: EPA’s Science Matters item names fused plugs and thermal cutoffs, the Missoula health department names the fused plug, and the California Air Resources Board names the fused plug too. Only one of the three also names the scenario the mechanism is for, and that one is the most useful sentence on this whole subject for a pet-owning household. The California Air Resources Board’s sentence gives the mechanism as a fused plug and the scenario as the device being knocked over.

That is the sentence to hold onto. Not because a dog knocking over a fan is going to burn the house down, which is a claim nothing here supports, but because the reason the date exists at all is a scenario in which the device ends up on the floor on its side. In a household with a large dog, a cat that jumps onto things, or a bird that is out of its cage in the evenings, that scenario is not exotic. It is the ordinary Tuesday version of the room.

Placement, With an Animal in the Room

Here is where the published record thins out, and it is worth being exact about what each source does and does not establish.

What the agencies publish about placement

EPA’s Indoor Air Filtration fact sheet gives three instructions for using and maintaining a portable or do-it-yourself air cleaner: “Use a PAC or DIY air cleaner where you spend the most time.” “Run it continuously on the highest fan setting.” And the one this section turns on: “Make sure air can flow freely around the air cleaner.”

EPA’s research page states the same idea as a conclusion from the safety work: DIY air cleaners made with newer model box fans “are unlikely to pose a fire or burn risk but should be kept clear of obstructions and operated with common sense precautions.”

EPA’s Wildfires and Indoor Air Quality page adds the box fan manufacturer’s own rules, which EPA relays: “Follow the box fan manufacturer’s instructions, such as: do not leave children unattended when the fan is in use; do not use an extension cord; do not use a damaged or malfunctioning fan; and ensure that there are working smoke detectors throughout the home.”

Two of those four transfer to an animal with no reasoning required. Do not use a damaged fan. Have working smoke detectors. The no-extension-cord instruction is worth reading twice in a pet household, because an extension cord is exactly what people reach for when they want the unit somewhere other than next to the outlet, and a cord across a floor is a thing animals chew and trip over. EPA’s instruction is not to use one.

What UL measured when the fan was blocked and tipped

Chemical Insights Research Institute of UL Research Institutes published a final report in July 2021, An Evaluation of DIY Air Filtration, conducted with support of the Office of Research and Development at US EPA. It tested five box fan models with a single filter taped to the back, which it chose deliberately: “this single filter construction was selected to put the largest stress on the box fan since a single filter results in a larger pressure drop than multi-filter configurations.”

It also ran an obstruction test that reads, in its description, almost exactly like a pet accident: “Two-sided obstruction (i.e., complete obstruction on the back side and the fan tipped onto its front face, resulting in blockage on both sides).” That configuration was run for at least seven hours.

Its findings, quoted rather than characterized:

• “Fire ignition was not achieved with any of the filter/fan test scenarios.”
• “Exterior surfaces that can come into direct contact with people remained below 36 °C, which is below the first-degree burn potential of 47.5 °C.”
• “Temperatures at the guard, switch, and output air were near or lower than ambient room air temperature for all test scenarios except for the two-sided obstruction tests where temperatures increased at all three locations. Even with the temperature increase, the parts remained below temperature thresholds that may pose a burn hazard.”
• “The highest temperature was observed at/near the winding on fans with a dust-laden filter, where the peak increase in temperature (relative to ambient) measured 57 °C.”
• “Air flow reductions were observed when filters were placed on the fans with a maximum of 29% flow reduction for the addition of clean filters, 31% reduction for smoke/soot-laden filters, and 70% reduction for dust-laden filters.”

EPA’s own summary of that work adds one number from the outside surfaces that matters for an animal that likes warm objects: “Temperatures on the external fan components remained near room air temperature when filters were attached.” And for the blocked case: “Temperatures on the external fan components were elevated above ambient temperature, but below hazard thresholds for the extreme scenario with both fan sides blocked.”

So the honest summary is this. A pet lying against the unit, sitting on it, or knocking it over is a real problem, and the problem is that the room stops being cleaned rather than that the house catches fire.

Be exact about which number measures what, because the two sets above are not measurements of the same thing. The 29, 31 and 70 percent figures are what attaching a clean, a smoke-laden or a dust-laden filter costs in airflow compared with a bare fan; the report’s method is to assume “the air flow measured without filters to be 100%” and read the reduction off that, and it notes that those “Air flow measurements showed a large standard deviation (of up to a half of the values)”. They are not a measurement of a blocked fan. For the blocked and tipped scenarios UL published temperatures rather than an airflow figure, and what it published is that the fans ran for at least seven hours without ignition and below burn thresholds. The reason to keep an animal off the unit is not in those numbers at all. It is EPA’s instruction that air must be able to “flow freely around the air cleaner”, next to EPA’s separate finding that “DIY air cleaners were almost completely ineffective with dirty filters”. An obstructed or dirty unit leaves the animal in the room breathing air nobody is cleaning. That is the harm. Writing that a cat on the fan will start a fire would be inventing a hazard that the controlled obstruction test did not find, and the one place that testing stops is named next.

The one gap UL names, and why the safety marking is the answer to it

The report closes its safety discussion with a passage a pet-owning reader should have whole:

“This study assessed DIY air cleaners for use in indoor environments at room air temperatures near 20 °C. Further analysis may be required for situations where ambient room air may be at higher temperatures (e.g., 40 °C). Should the blade turning be impeded in some way, it is possible that further heating could be created resulting in an additional hazard. However, fans that have been shown to meet the requirements of UL 507 are also required for a test that eliminates this impedance hazard.”

Read the second half carefully. UL is naming a scenario its testing did not cover, in which the blade itself is stopped or slowed, and pointing at third-party certification as the thing that covers it. A tail, a paw, a nose or a mouthful of fur reaching the blade is blade impedance. So is a toy pushed through the guard.

That is the strongest published reason a household with animals should treat the UL or ETL marking as non-negotiable rather than as a nice-to-have. The certification is not decoration on the box. It is the mechanism both EPA and the California Air Resources Board point at, and it is the one UL names for exactly the failure mode a curious animal creates.

One further limit belongs with those quotations. The report states of its own results: “Research results do not imply that the fans or fans with filters used in this study and with the described filter conditions meet UL 507; that can only be confirmed with compliant safety testing requirements and third-party verifications.” Testing five fans is not certifying a category, and EPA says the same of the whole body of work: “EPA cannot assure these results are representative of all scenarios. In addition, there are inherent risks associated with operating any electrical device.”

The inverted build, and why the people who measured it advise against it

There is one published design change aimed directly at the falling-object problem, and the group that tested it recommends against it.

The UC Davis team built a variant with the fan pointing at the floor rather than upward. Their stated reason is the one you are probably already thinking about: “An inverted CR Box would potentially be more robust against potential foreign objects being dropped into the fan.”

Then they measured it. “The CADR values for the inverted CR Box were all suppressed relative to the standard CR Box orientation (with the fan pointed upwards; Table S1).” In the supplementary table the inverted unit on low reads 481 against roughly 600 for the standard orientation. They offer two candidate explanations, short-circuiting of the exhausted clean air back into the filters, or increased shear forces resuspending particles that had settled on the floor, and they close with a recommendation:

“Therefore, we suggest that orienting a CR Box (or likely any air filter) such that the fan exhaust is towards the floor be avoided.”

That is a genuinely useful negative result for this page. The obvious modification for a household worried about something going into the fan is the one modification the measurement argues against, and part of the stated reason, blowing settled floor dust back into the air, is a reason that gets stronger rather than weaker in a room with a shedding animal in it. The remedy is the physical arrangement of the room instead, which is where the next part comes in.

What this site concludes, marked as this site’s own

No source located in this research addresses an animal sharing a room with a do-it-yourself air cleaner. Not EPA, not the California Air Resources Board, not ASHRAE’s current filtration position document, not the UL report, not the peer-reviewed characterization of the Corsi-Rosenthal box, and not EPA’s own pet fact sheet, which recommends buying a portable air cleaner and never mentions this build.

What the sources establish, separately, is that the fused plug in a post-2012 fan exists because of the device being knocked over, that a unit needs air to flow freely around it and to be kept clear of obstructions, that a blocked unit’s measured penalty is a large loss of airflow rather than a fire, that blade impedance is a hazard the safety marking is meant to cover and the testing did not reach, and that pointing the exhaust at the floor makes it work worse.

Putting those against a dog that leans on things, a cat that seeks out warm surfaces, and a bird or a small mammal that could be out of its enclosure is this page’s own reasoning, not an agency’s instruction. With that label attached, here is what follows from it:

• Stand it where the animal cannot lean on it, sleep against it, or topple it, and where the exhaust is not aimed at a bed, a crate or a cage. EPA’s own instruction to keep air flowing freely is about the machine. A dog that has chosen to sleep against the warm object is the most likely way that instruction gets broken in a pet household, and the cost is that the room stops being cleaned.
• Route the cord where nothing can chew or catch it, and do not solve a placement problem with an extension cord, because EPA relays the manufacturer instruction not to use one.
• Keep the filter faces clear. On a cube build with the fan on top, the filter faces are vertical walls at animal height, and a bed, a blanket or a body against one of them is an obstruction in exactly the sense EPA means.
• If you cannot place it out of reach in the room, consider a barrier around it rather than a modification to it, because the modification aimed at this problem is the inverted build that its own testers advise against.
• Check it when you check the animal. A unit that has been nudged, tipped or leaned on all afternoon is not cleaning the air, and EPA’s daily filter check is already an errand you are running in that room.

None of that is a published protocol. It is what the published sentences imply once an animal is in the room, and it is offered in this site’s voice rather than borrowed from a federal one.

Filters: the Part That Decides Whether Any of This Works

If you take one operating rule from this page, take this one. The build is not the hard part. Keeping the filter clean is.

What EPA publishes

EPA’s research page states the finding: “DIY air cleaners were almost completely ineffective with dirty filters, highlighting the need for frequent filter replacement during smoke events; the duration of filter lifespan will vary with use and conditions.” Its Science Matters item states it again in plainer words: “DIY air cleaners with dirty filters loaded with smoke or dust were almost completely ineffective, so be sure to replace dirty filters.”

EPA publishes the check and the trigger: “With all DIY air cleaner designs, filters must be changed frequently during smoke events as they may quickly load up with particulate matter. Check the filter daily during a wildfire smoke event. When the filter is dark brown or grey or smells like smoke, it is time to change it for a new filter.”

EPA’s Wildfires and Indoor Air Quality page gives a planning horizon: “Have extra filters on hand and change the filter when it appears dirty or starts to release smoke odors. During smoke events, filters may need to be changed every few weeks or days.”

The cleaner air room fact sheet gives the shortest version of the trigger, “Change the filter when it appears dirty or starts to smell.” Its Key Reminders equivalent on the infographic is “Keep extra filters on hand”. EPA’s Guide to Air Cleaners in the Home, 2nd edition, August 2018, generalizes it to every filter in the house: “All filters need regular replacement. If a filter is dirty and overloaded, it won’t work well.”

Buy the spares before smoke season. EPA’s own stated reason the do-it-yourself build exists is that filtration becomes hard to get exactly when everyone wants it, and that logic applies to the replacement filters as much as to the fan.

The pet-hair question, and what we could not source

There is a claim that travels with this build in pet-owning circles: that animal hair loads a MERV 13 filter far faster than dust does, so a pet household should expect to replace filters much sooner. It sounds obviously right, which is exactly why it needs checking rather than repeating.

We could not source it, and we are not going to publish it. On August 15, 2026 we read EPA’s DIY research page, EPA’s Wildfires and Indoor Air Quality page, EPA’s Science Matters item, EPA’s three wildfire guide fact sheets, EPA’s Guide to Air Cleaners in the Home, the California Air Resources Board’s wildfire air cleaner and odors pages, ASHRAE’s current filtration position document, the UL Research Institutes report and the peer-reviewed characterization of the Corsi-Rosenthal box, and none of them addresses animal hair as a filter-loading agent or publishes any figure for the effect. That is a statement about those documents, read in full, rather than a claim about everything that exists anywhere.

Two further points keep the record straight.

The one controlled comparison that does exist runs the other way on the comparison the claim assumes. The UL report loaded filters three ways: clean, smoke-laden, and dust-laden, the last of which it describes as prepared with a standardized dust composition “to represent a worst-case scenario loading from typical blockage generated by extended household use.” Dust produced the largest airflow penalty of the three, at up to 70 percent, and the highest measured winding temperature. That is a finding about dust, not about hair, and it is included here because it is the closest measurement to the question and it does not support the assumption.

Nothing about the operating rule depends on settling it. EPA’s instruction is to look at the filter every day during a smoke event and to change it when it is dark brown or grey or smells like smoke. That instruction works identically whether the loading came from smoke, from household dust, from a shedding retriever, or from all three. You do not need a loading rate. You need to look.

The screen prefilter, and exactly what it rests on

There is a modification that circulates specifically as a pet-household answer, and it deserves to be described precisely rather than either repeated as fact or silently omitted.

Clean Air Crew, a volunteer site that aggregates do-it-yourself air cleaner material and is not an authority, publishes a section headed with the words screen as prefilter. Its own summary sentence reads: “This modification, adding fiberglass screen material, can keep pet hair out of filters and prolong their useful life, with barely noticeable flow reduction!”

Underneath it, the page reproduces a social media post by a named individual, David Elfstrom, dated November 14, 2021, which reads: “Barely noticable flow reduction, just 1.27% loss!!” and “Mesh is 18×16 per inch, diameter 0.011, made by Adfors.” The misspelling of noticeable is the source’s own and is reproduced here rather than corrected.

Here is the honest accounting of that.

The 1.27 percent figure is one named individual’s own bench measurement of airflow loss on his own build, relayed by an aggregator quoting a social media post. It is not an agency figure, not a peer-reviewed figure, and not something anyone appears to have reproduced. That does not make it wrong. It makes it a hobbyist measurement, and it should be described as one wherever it is repeated.

The second half of the aggregator’s summary sentence, that the screen keeps pet hair out of filters and prolongs their useful life, carries no measurement at all in that source. It is an assertion, and it is the same assertion the previous section could not source anywhere. Adding a screen may well do what it says. Nothing located here measures whether it does, by how much, or at what cost in airflow on a build other than the one measured.

So if you add a screen, add it knowing that the airflow figure is one person’s measurement and the filter-life benefit is unmeasured, and keep doing EPA’s daily filter check regardless, since the screen changes nothing about EPA’s published trigger.

The same aggregator’s page is also the only place the fan-shroud diameter figures for specific fan brands surfaced in this research, attributed to the same individual. Those are hobbyist figures too. EPA’s own published instruction, cutting the shroud opening the size of the fan blades, is the federal version of the same idea and is the one this page recommends.

Noise and Power, Measured

Two specifications matter more than usual when the machine is going to run continuously in a closed room with an animal in it, and only one source located here measures either.

Dal Porto, Kunz, Pistochini, Corsi and Cappa measured both on the five-panel build, using a decibel monitor placed 5 feet from the center of the unit and perpendicular to the air exhaust, and a power logger.

Noise. They report: “The CR Box loudness varied from 58 dB (low speed) to 67 dB (high speed)”, and offer their own comparison points: “For reference, a modern refrigerator has a noise rating of about 50 dB and a LEED certified vacuum must be <70 dB.” Set against the two commercial machines they tested, one at a time, the build is not the quiet option: “The low speed loudness is similar to that measured for HEPA #1 (59 dB) but higher than that for HEPA #2 (54 dB).” What changes the comparison is holding the clean air delivery rate equal, and they work that out too: “To attain a CADR equivalent to the CR Box on low speed would require about two HEPA #1 units and 4 HEPA #2 units, which would yield 62 dB and 60 dB, respectively.” Both of those figures are above the 58 dB they measured for the CR Box on low.

Power. They report: “The power draw for the CR Box varied from 67 W (low speed) to 98 W (high speed)”, against 89 watts for one of the commercial HEPA units they tested and 43 watts for the other.

EPA names both as known drawbacks of the do-it-yourself approach. Its cleaner air room fact sheet states that these units “are often louder than commercial air cleaners.” Its research page lists “increased noise and heat generation from the fan motor, and limited data on how well DIY air cleaners filter smoke particles” as the drawbacks compared with a commercial air cleaner, and, in its list of what makes one design more suitable than another, includes “potential noise disruption” alongside the space and activity considerations. Its Science Matters newsletter item states the trade in one sentence: “Although the DIY air cleaners were just as effective as commercial air cleaners, they were louder and required more power to operate.”

Now the part this page will not do. No source located here publishes a sound level at which a dog, a cat, a bird or a small mammal is harmed or distressed, and there is no animal threshold on this page to compare 58 to 67 decibels against. Our quiet generator versus battery power station comparison carries a sourced figure for how common noise aversion is in dogs and for what it looks like, and that is the page to read if noise is your animal’s issue. Our honest review of dog smoke masks carries manufacturer and independently measured decibel figures for a commercial purifier, if you want a comparison point from the category this build substitutes for.

The behavioral readout is the one you actually have. An animal that will not enter the room, or that hides, paces or will not settle once the unit is running, is telling you something no specification does. A fearful or noise-averse animal is a question for your veterinarian, and it is worth asking before smoke season rather than during it.

On the power figure, one useful connection and one refusal. The connection: 67 to 98 watts puts this build in the same range as the ordinary box fan load our portable power stations for pets page already does runtime arithmetic around, which means a do-it-yourself air cleaner is a load you can plan for rather than a mystery. The refusal: we are not publishing a runtime for it, because the honest version of that arithmetic starts at your own fan’s data plate and your own power station’s rated capacity, and our watt-hour budget page walks through where to find the first and what the division leaves out. Note also that EPA writes a power outage into its reasons to leave the cleaner air room rather than into a reason to run it on batteries.

What This Build Does Not Do

Three limits, each with a source, and one of them is this site’s own final step rather than a published finding.

Gases, odors and ozone

A pleated particle filter traps particles. That is the whole of its job.

The California Air Resources Board states the limit for the category: “Many commonly used residential air cleaners will not effectively remove odors or gaseous pollutants. For example, HEPA filters are very useful for removing particles but cannot remove gases, which pass through the filter material.”

EPA’s Guide to Air Cleaners in the Home, 2nd edition, August 2018, states the same split and adds what is needed instead: “No air cleaner or filter will eliminate all of the air pollutants in your home. Note that most filters are designed to filter either particles or gases.” And: “To filter gases, choose a portable air cleaner with an activated carbon filter or other filter designed to remove gases. Note that there are no widely used performance rating systems for portable air cleaners or filters designed to remove gases. The CADR rating system is for particles only.”

That last clause is worth carrying away from the CADR table earlier in this page. Every number in that table is a particle number. None of them says anything about a gas.

Neither of those sources names ozone in the same sentence as a MERV 13 filter. So the final step, that ozone is a gas and this build therefore does not remove it, is this site’s own reasoning rather than a published finding, and it is labeled as such here.

One thing that runs the other way and should not be confused with the above: this build does not make ozone either. EPA’s Indoor Air Filtration fact sheet states that “Mechanical air cleaners and filters pull air through a filter that traps particles. For example, HEPA filters and other pleated filters are mechanical filters. Mechanical air cleaners and filters, especially higher-efficiency filters, effectively trap particles from smoke and do not produce ozone.” It is the electronic category EPA warns about: “Electronic air cleaners are not recommended for smoke removal because it is not clear that they can effectively remove smoke particles from the air and they may produce ozone.” Other technologies get the same treatment, EPA stating that photocatalytic oxidation, hydroxyl generators and devices with ultraviolet light components “are not recommended for smoke removal because they are not designed to remove particles from the air and may produce ozone or other harmful air pollutants.”

If you were considering an ionizer instead of building this, that is the paragraph to read twice. The California Air Resources Board certifies air cleaners sold in California for ozone emissions, stating that “Ozone emissions must stay within a concentration limit of 0.050 parts per million”, and it also states the limit of what that certification means: “Note: CARB certification does not reflect how effective an air cleaner is. CARB does not evaluate the effectiveness of air cleaners at removing pollutants.” A home-built assembly is not a certified device under any of those schemes, which is a fact about the paperwork rather than about the ozone, since a plain fan and a pleated filter is the mechanical category EPA says does not produce it.

ASHRAE’s position document makes the labeling point in its own general form, stating that “All electrically powered filtration and air-cleaning devices should only be used when tested and labeled for ozone emission in accordance with Underwriters Laboratory (UL) 2998 or an equivalent international standardized test to avoid hazards associated with ozone.” A device you assembled on the kitchen floor cannot satisfy a labeling condition, and that is worth knowing about the whole DIY category rather than treated as an alarm about this particular build.

It does not replace ventilation

EPA is explicit that filtration is one part of a plan rather than the plan: “Portable air cleaners, including DIY air cleaners, are only part of a comprehensive indoor air quality strategy. They do not replace the need for ventilation and should be used in conjunction with other appropriate health measures.”

EPA’s own cleaner air room routine builds that in. Its step six states: “Spend as much time as you can in the cleaner air room to get the most benefit from it. When the air quality improves, even temporarily, air out the cleaner air room to freshen the air.”

The other half of clean indoor air is not adding to it. EPA’s cleaner air room step five states: “Avoid activities that create smoke or other air pollutants indoors, such as smoking or vaping, burning candles, using aerosols, and vacuuming, unless you use a vacuum with a high-efficiency particulate air (HEPA) filter.” EPA’s pet fact sheet writes the same instruction for a household with animals: “Keep indoor air clean. Avoid cooking, especially frying or broiling, and do not vacuum, burn candles, use a fireplace or woodstove, or smoke or vape. These activities add particles to your home.” For a pet household the vacuuming line is the one that bites, because a shut-in animal and a shedding coat is exactly when you want to vacuum, and unless the machine has a HEPA filter EPA’s instruction is not to.

It does not cool the room

This is a filtration device with a fan in it, not an air conditioner, and treating it as cooling is the mistake that turns a good idea into a hazard.

EPA’s own research page lists “heat generation from the fan motor” among the drawbacks of a DIY unit compared with a commercial one. The unit adds heat to the room rather than removing it.

That distinction resolves something a reader moving between our pages will otherwise trip on. Our heat emergency with no air conditioning page works through what published fan guidance says about cooling a body, including the fact that four bodies publish four different temperature cutoffs for fan use and that none of them was written about an animal, and one of the published rules it carries is about running an electric fan in a closed room for cooling. The fan on this page is not doing that job. It is pulling room air through a filter, and EPA writes the closed room into the filtration instruction with its own condition attached, quoted in the next section. If your problem is the temperature in that room rather than the particles in it, the heat page is the one to read, and this one has nothing to add to it.

Heat Is the Limit That Outranks the Filter

Smoke season and heat season are the same season, and this build asks you to close a room.

EPA conditions the closed room in its own words. The Indoor Air Filtration fact sheet states: “To maximize effectiveness, use the air cleaner in a room with doors and windows closed, when it is safe to do so.” The final clause is EPA’s, and it is the clause that gets dropped when this instruction gets summarized.

The cleaner air room fact sheet keeps the exit open in step two, “Prevent smoke from entering the room. Close windows and doors, but do not do anything that makes it hard to leave the room in an emergency.” Its step three assumes you have a way to keep the room cool: “Stay cool. Run fans, window air conditioners, or central air conditioning.” Read the first item there against our heat emergency with no air conditioning page rather than as permission, because that page carries four different published temperatures above which agencies say a fan stops helping a person, none of them written about an animal. And the unit on this page is not one of the three things in that sentence. It is a filter, and EPA lists “heat generation from the fan motor” among its drawbacks, so it adds heat to the room rather than removing it.

Then it publishes the abandon condition, in a box on the page, quoted whole:

“In Some Cases, It May Be Best to Seek Shelter Elsewhere”

“If you can’t stay cool at home, the power goes out, or too much smoke is still getting in your home after you have followed the steps above, staying in a cleaner air room may not be the best option for you. Consider staying with friends or family, going to a public cleaner air shelter, or seeking relief from the smoke in another location with air conditioning and good air filtration.”

EPA’s research page carries the same instinct in its closing answer about using a DIY cleaner: “It is also important to take other steps to reduce your exposure to particles during wildfires, and to keep an eye on temperature and make sure you have a way to stay cool, even if that means going somewhere else.”

And there is a scope limit on the safety testing that belongs in this section rather than buried in the placement one. The UL report states: “This study assessed DIY air cleaners for use in indoor environments at room air temperatures near 20 °C. Further analysis may be required for situations where ambient room air may be at higher temperatures (e.g., 40 °C).” Room air near 20 degrees Celsius is about 68 degrees Fahrenheit. The reassuring temperature findings in that report were produced in a room at about that temperature, and the report itself says a much hotter room may need further analysis.

None of the sentences above is written about an animal. EPA does not leave the animal case entirely to inference, though. Its pet fact sheet states it in its opening paragraph, and the two sentences EPA prints immediately after are the ones that name which animals it bites hardest for and what an owner is told to do about it: “Evidence in humans indicates that the effects of exposure to smoke and heat together are much worse than to smoke alone. This is also likely true for your pets. Animals that are older, younger or pregnant or have heart or lung disease are especially at risk from smoke or heat and should be closely watched during all periods of poor air quality. Pet owners should consult with their veterinarian for advice to manage their pets during a smoke event, or a combined smoke and heat event, and to make an evacuation plan.” That is EPA, on the one document it wrote about animals, saying the combination is worse than the half this build addresses, naming the animals it is worst for, and sending owners to a veterinarian for both the management question and the evacuation plan. If your household holds an animal that is older, younger, pregnant, or living with heart or lung disease, that veterinary conversation is a thing to have before smoke season rather than during it.

The extension to your own house takes one step this page has to add in its own voice. The animals may not be in the room the household is judging. This page’s own split-room section says so. So put a thermometer in the room the animal is in, check it when you check the filter, and treat EPA’s reason for leaving a cleaner air room as applying to that room and not only to yours. An animal behind a closed door cannot leave a room that has become too hot, and it is the one occupant whose discomfort nobody in the house will feel. Our indoor temperature decay and species thresholds page carries the two-reading method for turning a thermometer into a rate for your own building, and is honest about why that rate will not finish the arithmetic into an hour. Putting the thermometer in the animal’s room is this page’s own reasoning rather than an agency’s instruction.

If leaving becomes the answer, the thing to have settled beforehand is whether the place you would go admits animals, because that is set locally and is never safe to assume. Our pets at cooling and warming centers page covers how to find that out by phone before you need it, and our AQI threshold page carries the human heat limit in the context of a multi-day smoke stretch.

The Rest of the Household

Birds. EPA’s pet fact sheet singles them out and routes them to the room this page is about: “Pet birds are particularly susceptible to smoke. Keep them inside, in a cleaner air room, if possible, when smoke is present.” What EPA does not do is say anything about what device belongs in that room for a bird specifically, and no source located here addresses a do-it-yourself air cleaner and a bird together. A bird’s respiratory system is not a mammal’s, which is a reason to take the question to an avian veterinarian rather than to reason across from a dog. Our bird and parrot emergency preparedness guide covers the enclosure and transport side of a smoke event.

Cats and dogs. EPA’s instructions for a smoke event are written for pets as a group: keep them indoors as much as you can with doors and windows closed if you can stay cool, keep outdoor bathroom breaks short, make water available at all times, and move potentially dangerous products such as pesticides out of their reach. That last one earns a mention on a build page, because building this involves duct tape, a cutting blade or scissors, cardboard and a trip to the hardware store, and the offcuts and packaging end up on the floor of the room the animal is shut in.

Short-nosed breeds. The same EPA fact sheet singles them out in the bullet list it writes for when smoke and heat are both present: “Short-nosed dogs, such as pugs and bulldogs, can have trouble breathing if there is smoke or heat. Use extra caution for these breeds.” That is EPA naming a group for whom the two hazards this section is about are harder, and it sits alongside EPA’s opening sentence about animals that are older, younger or pregnant or have heart or lung disease. Neither sentence says anything about a filter. Both are reasons to have the veterinary conversation EPA asks for before a smoke event rather than during one.

Reptiles, fish and small mammals. No source located in this research addresses an enclosure and a do-it-yourself air cleaner together. What is worth saying without inventing anything is that the exhaust of this unit is a directed airstream, that an enclosure relies on its own temperature and humidity being stable, and that aiming a continuous airstream at a terrarium or a tank changes the conditions inside it. Do not aim it at an enclosure. That is this page’s own caution rather than a published rule, and the temperature and humidity question for a specific species belongs with the exotics practice that already sees your animal.

Where more than one room has animals in it. EPA names moving a single unit between rooms as something people have found effective, and names building or buying more than one where a home is not well sealed or the smoke is thick. Neither is written about a household split across four rooms by species. Deciding which room gets the filtration is a household decision, and our one room with incompatible animals page covers the partitioning problem when animals that cannot share a space end up sharing one.

The Build and Operate Checklist

• Try to buy a portable air cleaner first. EPA recommends the DIY build only when products of known performance are not available or affordable, and states it is not recommended as a permanent alternative.
• Use a box fan made since 2012 carrying a UL or ETL mark. EPA and the California Air Resources Board both publish that condition, and EPA tells you to look for the certifying organization’s mark on the label.
• If the only fan you have is older, follow the fallback both agencies publish rather than improvising one: never leave it running unattended while you are away or sleeping.
• Buy MERV 13 filters, and buy spares at the same time. EPA’s own reason this build exists is that filtration is hard to get during a smoke event.
• Align the arrows on the filter with the direction of airflow through the fan, and get a good seal between the fan and the filter.
• Add the cardboard shroud, cut the size of the fan blades. EPA measured it at a 40 percent increase in clean air delivery rate at no cost and no extra footprint.
• Add filters if you have the floor space, or use a single thicker filter if you do not. Both are EPA-published designs with EPA-published numbers.
• If you build the five-filter design, use leg supports so air can flow through the bottom, as EPA’s infographic instructs.
• Do not orient it with the fan exhaust pointing at the floor. The team that measured that configuration found lower clean air delivery rates and advises against it.
• Stand it where the animal cannot lean on it, sleep against it or topple it, with the exhaust not aimed at a bed, a crate, a cage, a tank or a terrarium. Keeping it clear of obstructions is EPA’s instruction; applying it to an animal is this page’s reasoning.
• Do not run it on an extension cord. EPA relays that as one of the box fan manufacturer’s own instructions.
• Run it continuously on the highest fan setting, with the room’s doors and windows closed when it is safe to do so, which is EPA’s own conditional.
• Look at the filter every day during a smoke event. EPA’s change trigger is when it is dark brown or grey or smells like smoke.
• Do not add particles to the room you just cleaned: no candles, no aerosols, no frying or broiling, no woodstove, and no vacuuming unless the vacuum has a HEPA filter.
• Check working smoke detectors throughout the home, which EPA relays as part of the same manufacturer instruction set.
• Watch the temperature, not just the air quality. EPA’s own trigger to leave a cleaner air room includes not being able to stay cool and the power going out.
• Put a thermometer in the room the animal is in, not only the room you are in, and read it whenever you check the filter. The animal behind the closed door is the occupant whose discomfort nobody in the house will feel.
• Confirm by phone, in advance, whether any cooling center or cleaner air shelter you might use admits animals. It is set locally and it is never safe to assume.
• Treat the signs EPA lists as a reason to call your veterinarian rather than to adjust the fan: excessive coughing or gagging, trouble breathing, red or watery eyes, nasal discharge, excessive panting, fatigue or weakness, and reduced appetite or thirst. For cats EPA adds a second set: “In addition, cats may exhibit restless behavior, search for a cool spot, pant, drool, and have sweaty paws.”
• If any animal in the house is older, younger, pregnant, or has heart or lung disease, EPA names it as especially at risk from smoke or heat and tells owners to consult their veterinarian about managing that animal and about making an evacuation plan.

The Version You Do on a Quiet Afternoon

Three things done before a smoke event turn all of the above from a scramble into an errand.

Buy the fan now and check the label while you are in the shop. The condition is a mark from a certifying organization and a manufacture date, and both are on the fan itself rather than on a listing. EPA’s instruction is to look for a listing for electrical safety and a mark from a certifying organization such as UL or Intertek. Checking that in a lit aisle in October is a different experience from checking it by phone light during an evacuation warning.

Buy filters, then buy more filters. EPA says filters may need changing every few weeks or days during a smoke event and that a dirty filter makes the unit almost completely ineffective. The spare filters are not an accessory to this build. They are the build.

Decide the room now, with the animals in mind. EPA’s guidance picks the room by where people spend time. Your household may need a different answer, or may need to accept that one room gets the unit and another does not. Work out where the crate goes, where the cord runs, where the unit stands so nothing can lean on it, and whether the door stays shut with an animal behind it. Then run the thing for an hour on a clear day, with the animals in the room, and find out what they do about the noise while it still costs you nothing to learn.

Where to Go Next

This page is the build-and-operate spoke of our pet emergency playbooks hub, and it covers one object.

If you have not decided whether to keep the dog in yet. What AQI is safe to walk my dog has the veterinary band table and the number nobody publishes, and it is the page that decides whether you need this one.

If buying is still an option. Wildfire smoke and pets has the commercial purifier comparison and the room-size selector, which is what EPA would rather you did.

If the dog has energy and nowhere to put it. Indoor enrichment for smoke days covers the gear that tires a dog without raising its breathing rate, which is the other half of a shut-in day.

If the hazard is dust rather than smoke. Dust storms, haboobs and pets covers the differences, including why manufacturers publish a separate dust rating.

If the power is out or the heat is the real problem. Pets and power outages covers the wider failure, the watt-hour budget page covers sizing any load from its own data plate, and a heat emergency with no air conditioning covers the limit that outranks every filtration decision on this page.

Then do the one thing that costs nothing today. Go and look at the box fan you already own and find out whether it carries a UL or ETL mark and when it was made. If it does not, you have learned that on a clear afternoon with time to fix it, rather than at the point where the answer decides whether you run it overnight in a room with your animals in it.