The filter in your hot-air central heating system \u2013 known as a \u2018furnace\u2019 in American terminology \u2013 is the cheapest form of protection in the home, and most homeowners forget about it until something stops working. The usual advice is to change it every 90 days, but this figure doesn\u2019t apply to a large number of households. A single person living in a small flat with no pets can keep the same filter for six months. A family of five, with two dogs and a cat, living in a dusty neighbourhood, may need a new one every three weeks. The difference comes down to four factors, and you can work out the right frequency for your own home in less than ten minutes, simply by checking it against the light.<\/p>\n
The filter traps dust, pet hair and dander, pollen and the rest of the dirt floating around the house, before the air is drawn into the central heating system and blown back out as warm air. As long as it\u2019s doing its job, the indoor air stays cleaner and the system runs smoothly. When it becomes clogged, everything downstream suffers.<\/p>\n
A dirty filter does two bad things at once. Firstly, it restricts the air flow. The boiler needs a certain volume of air to pass over the heat exchanger in order to operate safely and efficiently. When the flow is restricted, the fan is put under greater strain, the heat exchanger reaches higher temperatures than designed, and comfort levels drop whilst energy consumption tends to rise. Furthermore, the dirt itself becomes a problem: some of the trapped particles re-enter the airstream, whilst those that slip past the filter settle on internal components and, over time, build up on motors and bearings.<\/p>\n
The US Department of Energy estimates that replacing a dirty filter with a clean one can reduce the air conditioner\u2019s energy consumption by 5 to 15 per cent. This figure applies to air conditioning, and during the cold season the same mechanism \u2013 restricted airflow \u2013 affects the central heating system. The estimate comes from the department\u2019s official guidelines on best practices for furnace maintenance<\/a>, published in the Energy Saver programme materials. A clean filter is one of the few household expenses that pays for itself straight away.<\/p>\n The three-month interval is a good starting point for an average home in a moderate climate, with two occupants and no pets. Most households do not fit this description. The actual interval depends on four factors, each of which requires adjustment.<\/p>\n Pets are the biggest variable. A golden retriever sheds enough hair and dander to clog a filter in about half the usual time, so if you have a pet that sheds, halve the interval at the very least. The number of people matters almost as much, as every extra person brings in dust from outside, sheds skin cells and stirs up what was already on the floor. As a rule of thumb, a household of five people uses up filters about twice as quickly as a couple.<\/p>\n The type of filter also affects the calculation. A simple fibreglass filter only traps large particles and remains permeable for longer. A high-efficiency pleated filter, with a MERV rating of 11 or higher, traps far more fine particles, so it becomes clogged more quickly and needs replacing sooner. Then there\u2019s the environment around the house: building works, the wildfire smoke season, a dirt track or a neighbourhood with heavy traffic all send extra particles into the system.<\/p>\n The 90-day rule has a flaw that has nothing to do with dust. It requires a person to remember, at a specific moment, a task for which there are no external cues. Psychologists call this ability prospective memory \u2013 that is, the memory of deferred intentions \u2013 and its experimental study began with a paper by Gilles Einstein and Mark McDaniel. In 1990, the two published an experiment in the *Journal of Experimental Psychology: Learning, Memory, and Cognition* in which participants carried out a routine task whilst simultaneously having to press a key whenever a specific word appeared on the screen.<\/p>\n One distinction from that paper is still used today. An event-bound intention is triggered when a cue appears: I see the letterbox and remember the letter. A time-bound intention requires the person to check their watch or calendar themselves, without any external cue, and is therefore more easily lost.<\/p>\n Changing the filter every three months is a time-based intention in its most fragile form. The interval is long, the filter sits in a basement or a cupboard, and the boiler runs just as well on day 89 as it does on day 91. The practical advice given by technicians all serves the same purpose: to transform a time-based intention into one linked to an event. The date written on the cardboard frame, the monthly check against the light, and the pack of spares kept next to the boiler are cues placed in one\u2019s path, so that the intention no longer depends on memory.<\/p>\n This distinction has its limits, which you should be aware of before relying on it. Laboratory experiments last minutes, whereas the filter requires a memory span of months. Furthermore, subsequent research has shown that the boundary between the two types of intention is less clear-cut than it seemed. A phone alarm instantly transforms a time-based task into an event-based one, and the results obtained with tasks of the latter type vary from one study to another. The model remains a useful framework for deciding where to set the cue, though it cannot predict who will forget and who will not.<\/p>\n The calendar isn\u2019t set in stone. A visual check takes fifteen seconds and tells you more than any schedule. Once a month, in the middle of the heating season, take the filter out of its housing and hold it up to a window or a light source.<\/p>\n If the light passes clearly through the material and the surface looks mostly white or light grey, the filter still has some life left in it. If it\u2019s uniformly grey or brown, replace it. If you can see clumps of dust, hair or debris on the surface, you\u2019ve waited too long, and the filter has already been costing you for several weeks.<\/p>\n A habit worth adopting from HVAC technicians: write the date on the cardboard frame with a marker the moment you fit the new filter. This removes any guesswork about how long the current filter has been in use. Check it monthly and, after two or three changes, you\u2019ll know your home\u2019s actual usage pattern.<\/p>\n Nobody sets out to forget about the filter. It slips your mind because it\u2019s tucked away in a cellar or a cupboard, out of sight and out of your daily routine. Then, one freezing January morning, the temperature in the house is 58 degrees Fahrenheit \u2013 that\u2019s about 14 degrees Celsius \u2013 and the boiler is running non-stop without catching up, and that\u2019s when the price of putting off maintenance comes home to bite.<\/p>\n A boiler that runs for a whole season with a blocked filter can overheat its heat exchanger to the point of cracking. It sounds dramatic, but a blocked filter is a well-known cause of emergency call-outs in the depths of winter. A cracked heat exchanger also poses a risk of carbon monoxide, turning a comfort issue into a safety hazard. The EPA lists worn-out, poorly adjusted or poorly maintained combustion appliances \u2013 including central heating systems and boilers \u2013 amongst the sources of carbon monoxide in homes, according to its own indoor air quality guidance<\/a> on combustion pollutants.<\/p>\n It also shows up on your bill, although less so than is usually claimed. Restricted airflow can prolong the operating cycles until the set temperature is reached, and longer cycles mean higher gas or electricity consumption for the same result. Exactly how much is lost depends on the system, and the greatest cost of a $15 filter left in place for too long usually lies in the wear and tear on the equipment.<\/p>\n The figures circulating about the energy waste caused by a dirty filter \u2013 such as \u201810 to 20 per cent of the heating budget\u2019 \u2013 have no traceable source, and published measurements indicate smaller effects. In 2010, Brent Stephens, Jeffrey Siegel and Atila Novoselac, from the University of Texas at Austin, monitored 17 forced-air systems in homes and small commercial buildings whilst they were in cooling mode. They compared low-efficiency filters with MERV 11 and 12 pleated filters, which offer greater resistance to airflow, and the energy consumption parameters changed only slightly.<\/p>\n The explanation lies in the way small systems are constructed. When resistance increases, a standard fan moves less air rather than running faster, so its energy consumption does not rise as it does in large buildings, where automation maintains the airflow rate. Variable-speed motors in newer systems compensate for the resistance with slightly higher electricity consumption. A review carried out for ASHRAE by Alavy and Siegel reached the same conclusion for domestic properties: more efficient filters have only a minor effect on the system\u2019s energy consumption.<\/p>\n There is a limitation to these results that must be clearly stated: clean filters of different grades were measured during the cooling season, whereas the issue here concerns filters that have been neglected for an entire winter. The findings suggest that the damage caused by a clogged filter is less evident on the energy bill and more so in the heat exchanger, which receives insufficient air. An overheated heat exchanger causes the temperature limiter to shut off the burner, and the boiler enters short start-stop cycles, which wear out the parts. The compelling argument for timely replacement is, therefore, the protection of the equipment, whilst savings on the energy bill come second.<\/p>\n The risk of carbon monoxide requires the same level of attention. The annual report of the US Consumer Product Safety Commission (CPSC), published in February 2025, estimates that in 2021 there were 204 accidental deaths from carbon monoxide poisoning, unrelated to fires, associated with consumer products. Heating systems were linked to 63 of these deaths \u2013 31 per cent \u2013 and rank second, after power tools, a category dominated by portable generators. Between 2019 and 2021, more than half of the deaths occurred between November and February.<\/p>\n The same report contains a detail that directly concerns those who change their own filters. Among the conditions found at the heating systems involved in incidents, the CPSC mentions, alongside compromised heat exchangers, filter compartment doors or covers that have been removed or are not properly sealed. After each change, the cover must be replaced and closed securely. The Commission also recommends one carbon monoxide detector on each floor of the house, near sleeping areas.<\/p>\n There is no one-size-fits-all answer. The schedule is tailored to the household\u2019s profile and must remain realistic. As a guide, depending on what the filter is exposed to:<\/p>\n Combined with a monthly visual check, this schedule catches problems before they become costly. Buy filters in bulk, so you don\u2019t end up at the DIY store on a Sunday at nine in the evening \u2013 which is exactly when, for some reason, you always seem to remember you need them. Keep your spares near the boiler, not in the garage, hidden behind the festive decorations.<\/p>\n These guidelines are in line with the recommendations of the ENERGY STAR programme, run by the EPA: check the filter every month, especially in winter and summer; replace it if it looks dirty; and don\u2019t keep it for longer than three months.<\/p>\n Changing the filter solves most of the boiler\u2019s day-to-day problems, but not all of them. Strange noises, frequent start-ups and shut-downs, a burner that won\u2019t ignite, or rooms that are heated unevenly indicate faults that a new filter won\u2019t fix. That\u2019s when you need someone to open up the boiler, test the components and find the actual fault.<\/p>\n A qualified technician can spot things you can\u2019t see for yourself. They check the fan motor, look for cracks in the heat exchanger, measure the gas pressure and clean the burners. An annual service catches minor faults whilst they\u2019re still cheap to fix. Without it, the failure of a minor part can lead to a decision to replace the whole unit, and that\u2019s how people end up spending thousands of dollars when a few hundred would have sufficed.<\/p>\n When the problem goes beyond a dirty filter, act quickly. A boiler that continues to run with obvious issues \u2013 such as rumbling, creaking or blowing out cold air \u2013 needs a professional as soon as possible. The longer it operates in a degraded state, the more secondary damage accumulates. In this scenario, a timely repair<\/a> saves you from having to replace the unit prematurely, as a part left to fail for weeks on end tends to drag the rest of the system down with it.<\/p>\n The advice to act quickly assumes something that maintenance guides usually leave unsaid, namely that you already know who to call. Finding a technician is in itself a task that gets put off, just like changing the filter. The thought \u2018I should find a reliable engineer before winter\u2019 has no set date or trigger, and as long as the boiler is working, it costs nothing to put it off. The intention is eventually carried out on a January morning when the temperature in the house is 14 degrees, when the first phone number that answers is dialled.<\/p>\n That moment is the worst time to make a choice, as the market works against the hurried buyer. The Bureau of Labour Statistics estimates around 40,000 vacancies for HVAC mechanics and fitters each year throughout the decade, many of which are to replace those leaving the trade. The ENERGY STAR guide to heating and cooling also notes that firms are busiest in summer and winter and recommends having your boiler serviced in the autumn. Anyone who calls at the peak of the season has to take whatever\u2019s available and doesn\u2019t have time to check anything.<\/p>\n The solution takes the same form as the date written on the frame. The search ties in with an event that happens anyway: the first filter change in the autumn. That same evening, draw up a shortlist of two or three firms; check each one carefully, and write the name of the chosen firm \u2013 along with their telephone number \u2013 on the boiler casing, next to the filter date. The list could start with a neighbour\u2019s recommendation or a category of HVAC contractors from a human-edited directory<\/a>, where firms are already grouped by trade.<\/p>\n The quality of a repair job cannot be assessed from the pavement. However, there are a few things about a company that can be checked via official sources, the first being its licence. The United States does not have a national HVAC licence, so you should start by checking the state or local register. In Utah, for example, contractors are licensed by the Division of Professional Licensing, whose public register allows you to search by name or licence number and shows whether the licence is active.<\/p>\n At federal level, there is only one clear requirement. Section 608 of the Clean Air Act requires EPA certification for any technician who opens a refrigerant circuit, which applies more to air conditioning and heat pumps than to gas burners. NATE certification is voluntary and attests to the technician\u2019s knowledge, without saying anything about the company. Proof of liability insurance must be provided in writing, and the quote must specify the part, labour and timeframe.<\/p>\n Then there is the company\u2019s public profile, i.e. how it appears in listings: the same name, the same address and the same telephone number everywhere, opening hours that match reality, and signs that someone is looking after the site. An article on the Jasmine Directory blog describes in detail what makes a company listing trustworthy<\/a>, starting with a search carried out at three in the morning, with water running under the sink. The scene is similar to that of the central heating system breaking down in January, only now it\u2019s in another corner of the house.<\/p>\n For the plumbing firm, the lesson is once again about the filter. Out-of-date opening hours or an old telephone number quietly accumulate in the public listing, and the customer in a hurry moves on to the next result without explaining why.<\/p>\n Business directories<\/a> are not all the same, and this difference is precisely what matters when it comes to verification. An aggregator automatically gathers its entries from other databases, and no one checks the company\u2019s website before publication. In an edited directory, a person reads every submission. At Jasmine Directory, an editor visits the website before the listing appears; descriptions written in an advertising style are shortened or sent back for rewriting, and regular link checks remove from the directory any companies whose websites are no longer responsive. Anyone wanting more details can find a comparison between aggregated directories and those with editorial selection<\/a> on the directory\u2019s blog.<\/p>\n The limitations must be stated just as clearly. A listing in an edited directory confirms that the company exists, that it operates in the category in which it appears, and that it can be found again in a month\u2019s time. The quality of the workmanship is not covered, nor is the fair price. The listing does not constitute a recommendation and does not replace the register of the licensing authority, where verification is carried out directly against the company\u2019s name.<\/p>\n There is also a practical limitation. On the day of a breakdown, nobody browses through categories; they ring the first result that comes up on the phone, and speed is what counts there. A verified list is only helpful if it has been compiled in advance, which brings us back to the signal from last autumn: the new filter is fitted, and the technician\u2019s name is written on the boiler casing.<\/p>\n A single winter of consistent monitoring tells you more than any general recommendation. Make a note of the date you fit each filter, check it monthly by the light, and note down when each one needed replacing. By spring, you\u2019ll have established your own schedule, tailored to your home, your pets and the local air quality. This habit is worth more than any rough rule of thumb, as it\u2019s based on your own home. Set your first reminder in your calendar today and check the filter right now, whilst it\u2019s still fresh in your mind.<\/p>\n","protected":false},"excerpt":{"rendered":" The filter in your hot-air central heating system \u2013 known as a \u2018furnace\u2019 in American terminology \u2013 is the cheapest form of protection in the home, and most homeowners forget about it until something stops working. The usual advice is to change it every 90 days, but this figure doesn\u2019t apply to a large number […]<\/p>\n","protected":false},"author":1,"featured_media":30760,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[58],"tags":[],"class_list":["post-30759","post","type-post","status-publish","format-standard","has-post-thumbnail","category-editors-corner"],"yoast_head":"\nWhy the 90-day rule doesn\u2019t hold true<\/h2>\n
Why we forget the filter: time- and event-related factors<\/h2>\n
How to check the filter without guessing<\/h2>\n
What happens when you keep putting it off<\/h2>\n
What the measurements reveal about bills, wear and tear, and carbon monoxide<\/h2>\n
A replacement programme that works<\/h2>\n
\n
Why professional maintenance still matters<\/h2>\n
It\u2019s best to call a technician in the autumn, when you change the first filter<\/h2>\n
What can be checked from the outside<\/h2>\n
What a listing confirms and what still needs to be checked<\/h2>\n
Keep track of your own data<\/h2>\n