TPMS Light: Steady and Flashing Mean Different Things

The tyre pressure monitoring light is the most misread instrument on a modern dashboard, and the reason is structural rather than anything to do with the driver. On most cars it is a single amber lamp shaped like a horseshoe with an exclamation mark in it, and that single lamp has been handed two jobs that have almost nothing in common. One job is to tell you a tyre is low. The other is to tell you the system that tells you a tyre is low has stopped working. The car separates the two signals clearly and deliberately, and hardly any driver has been told that it does.

What follows is built on the federal rule that puts the lamp there, because the rule is unusually explicit about this and because reading it settles arguments that the internet has been having for years. TPMS is mandated in the United States by Federal Motor Vehicle Safety Standard No. 138, and the text quoted throughout this page comes from the current version of that standard as published in the Code of Federal Regulations.

What the standard actually requires

Start with what the system is defined to be, because the definition is narrower than the name suggests. The standard defines a tyre pressure monitoring system as one that detects when one or more of a vehicle’s tyres is significantly under-inflated and illuminates a low tire pressure warning telltale. Note the word significantly, and note that the whole purpose clause is about warning drivers of significant under-inflation and the safety problems that follow from it. This was never designed as a pressure gauge. It was designed as an alarm, and an alarm has a threshold.

The threshold is written down. The detection requirements oblige the system to illuminate the low tire pressure warning telltale not more than 20 minutes after the pressure in one or more tyres, up to a total of four, falls to or below whichever is higher of two figures: 25 percent below the vehicle manufacturer’s recommended cold inflation pressure, or a fixed floor set out in Table 1 of the standard for the corresponding type of tyre. For a standard-load P-metric tyre — the ordinary passenger-car case — that table floor is twenty psi.

Two things follow immediately, and they matter more than anything else on this page for day-to-day driving.

An unlit lamp certifies nothing about your pressures

The first is that a dark dashboard is not evidence of correct pressures. A tyre sitting some way below the placard figure but not yet down by a quarter of it will not light the lamp, because it has not met the condition the regulation sets. The standard is candid about this in a passage most owners never read: the required owner’s manual wording tells drivers in so many words that the TPMS is not a substitute for proper tyre maintenance, and that it remains the driver’s responsibility to maintain correct pressure even when under-inflation has not reached the level that triggers the telltale. The regulator wrote that sentence because it anticipated exactly the misunderstanding that has since taken hold.

The second is that when the lamp does light, the tyre is already well down. This is not a nag. By the time the condition in the standard has been met, the tyre is deforming more than it should on every revolution, running hotter than it should, and steering and stopping the car worse than it did.

On applicability, the standard reaches passenger cars, multipurpose passenger vehicles, trucks and buses with a gross vehicle weight rating of 10,000 pounds or less, excluding vehicles with dual wheels on an axle. The phase-in ran from October 5, 2005, at a fifth of production, rose to seventy percent for the year beginning September 1, 2006, and became universal for vehicles manufactured on or after September 1, 2007. Vehicles built in two or more stages, and vehicles altered after certification, had until September 1, 2008.

There is a wrinkle in that timetable which is genuinely useful when you are looking at older stock, and it is easy to miss. During both phase-in years, manufacturers were allowed to count a vehicle as complying even though it did not meet the malfunction provisions, unless they chose to certify to those as well. In other words, a car built in that window can lawfully have a working low-pressure warning and no malfunction indicator at all. On such a car, a dead sensor may produce no distinctive signal whatsoever — which is a very different proposition from the same fault on a car built after the requirement became universal.

Steady or flashing: one lamp, two messages

Here is the part that makes this page worth reading, and it turns on a paragraph of the regulation that is not widely quoted.

The standard requires a second warning entirely, separate from the low-pressure one. It says the vehicle must be equipped with a system that includes a telltale providing a warning to the driver not more than 20 minutes after the occurrence of a malfunction that affects the generation or transmission of control or response signals in the vehicle’s tyre pressure monitoring system. That is a warning about the plumbing of the system itself: a sensor that is no longer transmitting, a signal that is no longer arriving, a module that is no longer receiving.

Manufacturers are then given a choice of two ways to deliver that warning, and which one your car uses decides what you see.

The two ways a manufacturer may deliver the malfunction warning

A dedicated telltale. The vehicle can fit a second, separate lamp for malfunctions, mounted in clear view of the driver and identified by the word TPMS, as described in Standard No. 101. It must stay lit for as long as the malfunction exists whenever the ignition is in the run position, and it gets its own bulb check at start-up. On a car built this way there is no ambiguity at all: one lamp means low, a second lamp spelling out TPMS means broken.

A combination telltale. The far more common arrangement uses a single lamp for both jobs, and the standard tells it exactly how to disambiguate. In the words of the regulation, the combined telltale “Flashes for a period of at least 60 seconds but no longer than 90 seconds upon detection of any condition specified in [the malfunction paragraph] after the ignition locking system is activated to the ‘On’ (‘Run’) position. After each period of prescribed flashing, the telltale must remain continuously illuminated as long as a malfunction exists and the ignition locking system is in the ‘On’ (‘Run’) position.” The same paragraph adds that the sequence must be repeated on every subsequent start until the situation causing the malfunction has been corrected.

Read that back slowly, because it is the whole answer to a question thousands of drivers ask every day. A flash-then-steady lamp is not the car being indecisive, or a bulb on its way out, or a low tyre that the system is still thinking about. It is a defined, regulated, deliberate signal, chosen precisely so that a single lamp could carry two meanings without either being lost. The flashing is the qualifier. The steady illumination that follows it is the same lamp continuing to say something, and what it is saying is not the same thing it says when it comes on steady from the start.

This is why the classic frustrating scenario resolves so neatly. Somebody checks all four tyres with a decent gauge, finds every one of them spot on, drives off, and the lamp comes back. They inflate a couple of psi over for good measure. It comes back. They begin to suspect the gauge, the garage, the tyres, and eventually the car. Nothing is wrong with any of them. The tyres are fine and always were. The system is broken, it has been telling them so in the first minute after every single start, and nobody was watching the lamp during the only part of the cycle when it was saying anything different.

The state where nothing is being monitored

It is worth dwelling on the malfunction state a little longer, because there is a category error buried in how most people react to it.

Warning lamps are normally read as reports about the car. Oil pressure low, coolant hot, brake fluid down — each names a physical condition, and when the condition goes away the lamp goes out. The TPMS malfunction indication belongs to a smaller and stranger family: lamps that report on the state of an instrument rather than the state of the machine. It is the dashboard equivalent of a smoke alarm chirping about its own battery. The chirp is not a fire, and treating it as a nuisance rather than as information gets the risk backwards, because the alarm is now decorative.

The practical consequence is that the malfunction state is more dangerous than it feels, in exactly the way that the low-pressure state is more dangerous than it feels. A steady lamp is a problem you can see: pressures are down, go and put air in. A flash-then-steady lamp is the absence of a problem you can see, and it stays that way right up until the moment you have a tyre going down on a motorway and no warning arrives, because there is nothing left in the system to send one.

It also means that the standard advice — go and check your pressures — is the wrong first move in this state. It is not harmful, and checking pressures monthly is a good habit regardless, but it will not resolve anything and it burns weeks of driving while the owner assumes the problem is elusive rather than diagnosed. The right first move on a flash-then-steady lamp is to have the system read, which takes a TPMS tool a couple of minutes and identifies which sensor has stopped answering.

Direct and indirect systems, and why one of them must be told

There are two fundamentally different ways to build a TPMS, the standard mandates neither, and the difference decides most of what you will experience as an owner.

A direct system measures pressure. A small sensor sits inside each wheel, usually built into the valve stem or strapped to the drop centre of the rim, and it contains a pressure transducer, a temperature sensor, a radio transmitter and a battery. It sends a reading and a unique identifier to a receiver in the car, and the module compares the reading against a threshold derived from the placard pressure. Because it measures the actual thing, it can tell you a number, it can tell you which corner, and it works when the car is standing still.

An indirect system measures nothing of the kind. It infers pressure from wheel speed, using the sensors that are already on the car for the anti-lock brakes. An under-inflated tyre has a slightly smaller rolling radius, so it turns marginally faster than its neighbours to cover the same ground, and the module watches for that discrepancy. More sophisticated versions also examine the vibration signature of each wheel, because a softer tyre resonates differently. There is no sensor in the wheel and no battery to die.

The row that catches people out is the reset. On an indirect system, the module has no absolute reference — it only knows that one wheel is turning differently from the others. When you inflate the tyres correctly and then drive off without resetting, the system carries on comparing against whatever it learned last, which may well have been the pattern produced by the soft tyre you have just fixed. Worse, if you reset it while a tyre is low, you have just taught it that low is normal, and it will not complain until that tyre goes lower still. The regulation itself acknowledges the step: its own compliance procedure has the tester set or reset the system in accordance with the owner’s manual before the calibration and learning phase begins, and it allows the manual to explain how to use a reset button where one is provided.

The last row matters at the tyre shop rather than in the driveway, and it is covered in its own section further down.

If you want to know which type a specific car uses before you go and look at it, the federal VIN decoder returns it as a field, which our guide to what a VIN can and cannot tell you about tyres and wheels walks through. It is one of the more useful fields in the decode, and it changes what questions are worth asking.

Why it appears on cold mornings and is gone by the afternoon

This is the single most common way people meet the TPMS light, and it is also the one where the explanation is almost boringly physical.

The air in a tyre is a fixed quantity of gas in a container of nearly fixed volume. Cool that gas and its pressure falls; warm it and the pressure rises again. Nothing has escaped, no seal has failed, and nothing needs repairing. The first genuinely cold night of autumn drops every tyre on every car in the street by the same proportion, and the ones that were already sitting a little low are the ones that cross the threshold the standard sets.

Then the sun comes up. The car warms, the tyres warm further as you drive on them, the pressure climbs back above the point where the system objects, and the lamp goes out on its own. By lunchtime the whole episode looks like a glitch. It was not a glitch. It was an accurate report of a real condition that has since reversed itself, and it carried a piece of information most people throw away: those tyres were close enough to the threshold that a cold night was enough to push them over it.

The regulation is quietly aware of all this. Its test conditions specify an ambient temperature band running from 32 to 104 degrees Fahrenheit, and they go so far as to require that the vehicle’s tyres be shaded from direct sun while it is parked. A rule-writer only bothers to control for something that moves the result.

Two failure modes hide behind the seasonal one. A slow leak looks identical on any given morning — the distinguishing feature is that it does not recover, and that the same corner keeps needing air. And a tyre that has been repeatedly topped up in cold weather can end up substantially over-inflated when summer arrives, which wears the centre of the tread and rides badly. Set to the placard, not to whatever makes the light go out.

Whether the spare is watched, and why nobody can tell you in general

A recurring question, and the honest answer is that it depends on the car and that the regulation deliberately leaves it open.

The compliance test is run on the tyres fitted to the vehicle at the point of first sale, and the standard explicitly excludes the spare from that set — while adding that the spare may be used for malfunction testing purposes. Separately, the standard lists what a manufacturer may optionally add to the owner’s manual, and one of the permitted additions is whether the tyre pressure monitoring system functions with the vehicle’s spare tyre, where one is provided. Something described as an optional disclosure is, by construction, not a requirement.

So there are three real-world cases and you cannot infer which one you have from the outside. Some vehicles carry a fifth sensor in a full-size spare and monitor it like any other wheel. Some carry a sensor in the spare that is not monitored while stowed but is picked up once it is fitted and the system relearns. Many carry nothing at all in the spare, which is universally true of the temporary space-saver type and of the tyre-sealant-and-compressor kits that have replaced spares on a lot of modern cars.

The consequences are practical. Fit an unsensored spare and, on a direct system, you have created exactly the condition the standard treats as a malfunction: a wheel position from which no signal arrives. Expect the flash-then-steady sequence within a short drive, expect it to persist until the proper wheel goes back on, and do not spend that week wondering what else you have broken. Conversely, a spare that is never monitored can sit flat in its well for years while every dashboard indication says the car is fine, which is the least helpful moment to discover the fact.

The sensor battery: a real cost that never reaches the service history

Now the part that concerns the purchase rather than the repair.

Each direct-system sensor is a sealed unit. The battery inside it is not a serviceable item, is not a coin cell behind a cover, and cannot be replaced by anybody without the sensor being cut open, which destroys it. When the battery is exhausted the sensor stops transmitting, the module stops hearing from that wheel position, and the car does precisely what the regulation requires it to do: it flags a malfunction. The remedy is a new sensor.

Two features of that make it an unusually awkward cost on a used car.

Why mileage is no defence, and why they go together

It is driven by time and transmission cycles rather than by mileage. The sensor is broadcasting whenever the wheel is turning, and drawing a trickle even when it is not. A garage queen ages its sensors much the same way a commuter does. This is one of very few wear items on a car where a low odometer reading offers almost no protection, which is precisely the sort of item a used-car buyer should care about and almost never asks about. Any specific lifespan you see quoted is a manufacturer’s estimate or a trade rule of thumb, not a regulated figure, and it is not in the standard — so treat a confident number in years with the same scepticism you would apply to any other unsourced claim about a car.

They fail as a cohort. Four sensors fitted on the same day at the same factory, on the same duty cycle for the same number of years, arrive at the end of their lives at roughly the same time. The common pattern is one failure, a single sensor replaced, and then the remaining three following it over the next year or two — three more visits, three more tyre-machine charges, three more relearns. Which is why, when the first one goes on a car of any age, the question worth asking the workshop is whether replacing the set makes more sense than replacing the one.

The bill itself is not enormous per wheel, but it is a bill with a shape: a sensor, a service kit, a tyre broken down off the rim and refitted, rebalancing, and a relearn procedure. Multiply by four and it stops being trivial. It is also, unhelpfully, a cost that clusters with the other tyre-related decisions on an older car, because the sensible moment to change sensors is when the tyres are coming off anyway — and if the tyres are old enough to be near replacement themselves, the two jobs arrive together. Tyre ageing is a separate clock that runs whether the car moves or not, and our guide to how long tyres last, and how to read the date moulded into the sidewall covers how to establish where a set actually is in its life.

What happens to sensors when tyres are changed

Every tyre change on a direct system is a procedure involving the sensor, whether or not anyone tells you so.

The tyre has to be broken away from the rim and levered off, and the sensor is sitting right there in the path of the machine. A careless bead breaker snaps the stem or cracks the housing. Then the sensor’s seals, valve core and retaining nut — collectively the service kit, and a genuinely cheap item — are meant to be renewed at every tyre change, because a perished grommet leaks slowly through the very component fitted to detect leaks. Then the wheel is rebalanced, since the sensor is a mass on the rim. Then, on many vehicles, the system needs a relearn so the module knows which identifier is now at which corner, which matters as soon as the car tries to tell you which tyre is low.

Rotating tyres between axles raises the same relearn question for the same reason. On some vehicles the system works it out on its own after a period of driving. On others it needs a tool, a specific sequence with the ignition, or a controlled deflation at each corner in turn. There is no universal procedure, which is exactly why the standard leaves the reset method to the owner’s manual rather than prescribing one.

Two things follow for a buyer. First, a car whose tyres were changed cheaply is a car where the sensors may have been treated cheaply too — and a snapped sensor stem is often quietly replaced with a plain rubber valve, which leaves that wheel permanently silent and the system permanently in malfunction. Second, the malfunction lamp is the honest record of that decision, and it will keep telling you at every start-up because the standard requires it to. The regulation itself lists installing a tyre or wheel that is incompatible with the TPMS alongside disconnecting a component’s power supply as equivalent ways of deliberately inducing a malfunction for test purposes. The rule-writers regarded the wrong wheel and a cut wire as the same category of event, which tells you how the car will regard it too.

Aftermarket wheels, and the car with no sensors at all

Of all the ways a TPMS ends up switched off, this one carries the largest bill and is the hardest to see in an advert.

Wheels are one of the most common modifications on a used car, and a set of aftermarket wheels does not come with sensors in it. Someone fitting them either transfers the originals across, buys a new set of sensors and has them programmed to the car, or fits nothing and accepts the light. The third option costs nothing on the day and is therefore extremely common, and the car that results has a permanently flashing-then-steady lamp, no tyre pressure monitoring of any kind, and an owner who has stopped seeing the dashboard.

The required owner’s manual wording anticipates exactly this. It tells drivers that TPMS malfunctions may occur for a variety of reasons, including the installation of replacement or alternate tyres or wheels that prevent the system from functioning properly, and it instructs them to check the malfunction telltale after replacing one or more tyres or wheels for that reason. That instruction is in the regulation, which means it is in the manual of every compliant car, which means it has been sitting in millions of gloveboxes being ignored.

When you are buying, the sequence to keep straight is this. Aftermarket wheels are not a defect. Aftermarket wheels with no sensors are a known, quantifiable cost — four sensors, programming, four tyres off and on, four rebalances — and it belongs in the negotiation rather than in the surprise pile. Where a seller says the original wheels are included, ask to see them and ask whether the sensors are still in them, because a set of originals stripped of their sensors is the same bill in a different box. And where the modification story is larger than a set of wheels, it is worth checking what the car’s recorded history says before you drive out to see it; you can look the VIN up against title and history records in the time it takes to read the advert. A proper pre-purchase inspection treats wheels and tyres as one of the areas where a photograph is worth almost nothing.

Resetting the system, in general terms

There is no single reset procedure, and any page that gives you one as though there were is describing one manufacturer’s method and hoping. What can be said in general is what the reset is for, and that is the part people get wrong.

On a direct system there is usually nothing to reset after ordinary inflation. Put the pressures right, drive the car, and the sensors report a value above the threshold; the module sees it and puts the lamp out by itself. A reset or relearn on a direct system is for a different event — a sensor replaced, wheels rotated, a winter set swapped on — where the module needs to be told which identifier now sits at which position. Some cars do this automatically after a period of driving, some need a scan tool or a dedicated TPMS tool, and some use a magnet or a controlled deflation at each wheel in a set order.

On an indirect system the reset is not optional housekeeping. It is the whole calibration. You inflate all four tyres to the placard figure, then tell the system that this is now correct, and from that point it watches for deviation from the state you have just declared. Skip it and the baseline is wrong. Perform it with a tyre low and the baseline is wrong in the more dangerous direction. The button or menu item is generally in the vehicle settings, the glovebox, or below the steering column, and the owner’s manual is the only reliable authority on which.

Why you cannot simply drive on with it lit

The temptation is understandable. The car drives normally, the lamp is amber rather than red, and nothing about the experience feels like an emergency. Here is what you are accepting in each of the two states.

With a steady lamp, at least one tyre is significantly under-inflated by the standard’s own definition, and the consequences are the ones the regulation was written to prevent. The required owner’s manual text sets them out directly: driving on a significantly under-inflated tyre causes the tyre to overheat and can lead to tyre failure, and under-inflation also reduces fuel efficiency and tread life and may affect the vehicle’s handling and stopping ability. A soft tyre flexes more with every revolution, and that flexing is what generates the heat that destroys the structure from the inside. The failure, when it comes, is not a slow deflation. It is a tread separation or a sidewall letting go, and it picks its own moment to do it.

With a flash-then-steady lamp, you have no tyre pressure monitoring. Every argument above still applies to your car; you have simply removed the thing that would have told you about it. And you have removed it permanently rather than for a journey, because a malfunction does not fix itself.

Two further consequences are worth stating because they are the ones people discover late. Several states include the tyre pressure monitoring system in the annual safety inspection, so a car in either state can fail a test it would otherwise have passed. And a lamp that has been lit for a long time is a lamp the owner has stopped seeing, which means the next warning — a genuine one, about a genuinely deflating tyre — arrives on a dashboard where it is indistinguishable from the wallpaper.

Using this at a viewing

Everything above collapses into a short routine that costs nothing and takes about a minute, and it has to happen at a specific moment that most viewings skip past.

The reason to run this at the start of the viewing rather than the end is that it is one of the very few checks that gives you a hard answer for free, and the answer changes what the rest of the visit is for.

Putting it in the price, and one thing it is not

A TPMS fault is a good example of a defect that ought to be easy to negotiate and usually is not, because neither side knows what it costs.

Price it as what it is: a parts-and-labour job with a known shape, repeated once per affected wheel, plus a relearn. Get a figure from a tyre shop before the conversation rather than during it, because a seller who is asked to knock something off for a light is being asked for a favour, while a seller who is shown a written quote is being shown an arithmetic problem. If the sensors are simply old rather than dead, the honest framing is that they are near the end of a life you will be paying for shortly, in the same category as a cambelt with a year left on it.

Be clear-eyed about who owns the problem after the money moves. Most private sales, and a great many dealer sales, are made on an as-is basis under the federal Used Car Rule’s window form, and a fault visible on the dashboard at the point of sale is about as squarely the buyer’s problem as a fault can be. There is nothing improper about that. It is simply the reason to establish the cost before signing rather than afterwards.

Two things worth checking while you have the VIN in your hand, both free. Open safety recalls are searched by VIN rather than by symptom, and the remedy is free to whoever holds the car at the time. And the recorded history of the car is worth reading for what it says about how it has been treated generally — you can pull a history report from the VIN in a couple of minutes, and it costs nothing to know before you commit.

Finally, the boundary of what this lamp can tell you. TPMS reports on inflation and on its own health. It says nothing about tread depth, nothing about the age of the rubber, nothing about whether the four tyres match, and nothing about the alignment or suspension faults that quietly destroy tyres from one edge inwards. A car can pass every test on this page with four tyres that need replacing. Those are separate checks, they are worth more money than this one, and none of them is answered by the lamp.

One closing comparison, because the contrast is instructive. The other amber lamp people worry about can be switched off in under a minute by anyone with a code reader, which is why that page is largely about detecting a light that has been erased. This one behaves in the opposite way: the standard requires it to come back at every start until the underlying situation is corrected, so it cannot be quietly deleted — but it can be misdescribed, and “it just needs air” is the misdescription to watch for. The other half of that dashboard argument — how to tell whether a check engine light was cleared before you arrived — is set out separately.

Common questions

What does the TPMS light mean?

It means one of two different things depending on how it behaves. Illuminated steadily from start-up, it means at least one tyre has reached the under-inflation threshold set by FMVSS No. 138 — a quarter below the pressure on the vehicle placard, or the floor in the standard’s own table, whichever is higher. Flashing for roughly the first minute after the ignition goes to run and then staying on continuously, it means the system has detected a malfunction in itself, and in that state it may not be able to detect or signal low pressure at all.

Why is my TPMS light flashing?

Because the system has a fault rather than because a tyre is low. The standard requires a vehicle that uses one combined lamp for both warnings to flash it for at least 60 seconds but no longer than 90 seconds at start-up when a malfunction is present, then hold it steady, and to repeat that sequence every time the vehicle is started until the cause is corrected. The usual causes are a sensor whose sealed battery has died, a sensor damaged or omitted during a tyre change, a wheel fitted with no sensor in it, or a sensor that has never been paired to the car.

Can I drive with the TPMS light on?

You can, and you should not do it indefinitely in either state. With a steady lamp a tyre is significantly under-inflated, which the standard’s own required owner’s manual wording links to overheating and tyre failure, worse fuel consumption, shorter tread life and degraded handling and stopping. With a flashing-then-steady lamp the car is not monitoring tyre pressure at all, so the next slow puncture arrives with no warning. Check the pressures cold and set them to the placard figure; if the lamp flashes, get the system read rather than continuing to check pressures that are probably fine.

Why does the light come on when it is cold and go off later?

Because tyre pressure falls as the air inside cools and rises again as it warms. A cold night can take a tyre that was already slightly low across the threshold, and by the afternoon the sun and the driving have put the pressure back above it. Nothing is broken, but you have been told something useful: those tyres were close enough to the limit that a temperature swing was enough. Set them cold to the placard figure. If it happens again after that, one tyre is genuinely losing air.

Does the TPMS monitor the spare tyre?

Sometimes, and there is no general answer. The standard runs its compliance test on the tyres fitted at first sale and specifically excludes the spare, and it treats a statement about whether the system works with the spare as an optional addition to the owner’s manual rather than a requirement. Some cars monitor a full-size spare, some pick it up only once it is fitted, and many carry nothing in the spare at all — which is always the case with space-savers and with the sealant-and-compressor kits that have replaced spares on many modern cars. Fitting an unsensored spare to a direct system will usually trigger the malfunction warning within a short drive.

Do TPMS sensors need replacing on a used car?

Eventually, on any car with a direct system. Each sensor contains a sealed battery that is not replaceable, so when it is exhausted the sensor is replaced as a complete unit. Because they were all fitted on the same day and have run the same duty cycle, they tend to reach the end together, and the first failure is often followed by the other three. It is a genuine deferred cost that will not appear in a service history, because no service schedule lists it — nothing maintains a sensor, it simply stops. The same bill arrives in one lump on a car wearing aftermarket wheels with no sensors transferred into them, which the regulation itself names as a cause of malfunction, so treat sensors, programming and four tyre fittings as a line item on any modified car.

Can a seller just switch the light off before I arrive?

Not in the way the check engine light can be erased. The standard requires the malfunction sequence to restart at every ignition cycle until the situation causing it has been corrected, so it will be back on the next start. What a seller can do is inflate the tyres on the morning of the viewing, which legitimately clears a steady low-pressure lamp and tells you nothing about whether a tyre has a slow leak, and describe a flashing lamp as needing air. Start the car yourself from cold, watch the first minute, and you have the answer without needing anyone’s account of it.