A tire loses air whether or not anyone drives it. Rubber is slightly porous. Cold weather shrinks the air inside, and a valve seals well but not perfectly. So the pressure set at the last service is not the pressure in the tire today. The tread keeps a record of the difference.
What is proper tire inflation?
Proper tire inflation is the cold pressure the vehicle manufacturer specifies for that vehicle. It is printed on a placard inside the driver’s door. The number molded into the tire sidewall is a different figure, and it is the tire’s maximum. NHTSA states the distinction plainly: the correct pressure is what the vehicle manufacturer has listed, and not what is listed on the tire itself.
That placard figure is specific in a way a general rule cannot be. It is chosen for one car’s weight, its balance front to rear, and the handling the maker tuned for. Front and rear pressures can differ, and on some vehicles they do. None of that survives being swapped for a number somebody remembers.
How often should you check tire pressure?
Check tire pressure monthly, and check it cold. Cold means parked for at least three hours. The US Tire Manufacturers Association sets both halves of that rule, and says a tire pressure monitoring system does not replace a monthly gauge check.
Two things pull pressure down between checks. Air permeates out through the rubber at 1 to 2 psi a month. Temperature takes more, moving pressure one to two pounds for every 10 degrees of change. A 30-degree drop between September and January costs 3 to 6 psi on its own. That is why warning lights arrive in clusters on the first cold morning of the year.
What the warning light does and does not tell you
The warning light is a floor, and the floor is lower than most drivers assume. FMVSS 138 requires the telltale no later than 25 percent below the placard pressure, or a fixed minimum for that tire type, whichever is higher. Nothing requires a system to speak before that.
On a car placarded at 35 psi, 25 percent below is 26.25 psi. A tire can sit almost 9 psi light, month after month, and stay legal, silent, and normal to the eye. The light was set to catch a safety problem. It was never set to catch a bill.
How inflation pressure shapes tread wear
Inflation pressure decides which part of the tread carries the vehicle, and whichever part carries it wears first. The pressurized casing supports the tread band from below. Change the pressure, and you change the shape of the contact patch.
| Inflation state | Where the tread wears | What is happening |
|---|---|---|
| Under-inflated | Both outer shoulders, center stays deeper | The casing stops supporting the crown, so the two shoulders take the load |
| Over-inflated | The center ribs; both shoulders stay deeper | The crown rounds outward and a narrow center strip carries the car |
| At placard pressure | Evenly across the full width | Load spreads across the whole contact patch |
Both shoulders wearing together points at pressure. One shoulder wearing on its own points at alignment, which is a different repair on a different bill. Cupping, feathering, and the rest of the tire wear patterns each name a different system again.
Low pressure also builds heat. Every turn of the wheel flexes the sidewall and the belt edges, and rubber turns that flexing into heat. That heat weakens the bond between the steel belts and the rubber around them. NHTSA’s engineering analysis of the Firestone recall called the belt edge the highest-strain region in a steel-belted radial.
A tire that goes this way sheds its tread. It is a belt separation, and the sudden burst most drivers picture is a different failure. The same analysis found vehicle load matters more to it than pressure does, so low pressure alone rarely takes a sound tire that far. It does stack the odds. In one NHTSA crash study, tires more than 25 percent low sat on vehicles with a pre-crash tire problem 10 percent of the time. Correctly inflated tires did so 3.4 percent of the time.
What low pressure costs in tread
There is one published figure for what low pressure costs in tread, and almost nobody uses it. In its 2002 rulemaking on tire pressure monitoring, NHTSA adopted unpublished data from Goodyear estimating that tread life falls 1.78 percent for every 1 psi a tire runs under. Goodyear assumed the relationship was linear. On the 45,000-mile tire life in the same data, that is about 800 miles of tread per psi. NHTSA used the rate exactly that way. It multiplied 800 miles by a 6.8 psi average deficit and reported 5,440 miles lost.
Run that arithmetic on the seasonal drop. A car set correctly in September and left alone through a 30 degree cooling is 3 to 6 psi light all winter, which is 2,400 to 4,800 miles of tread. Run it on the warning threshold instead and a tire sitting where the light is finally required gives up around 7,000 miles. Both are Goodyear’s linear assumption carried to a different deficit, so treat them as orders of magnitude.
Michelin publishes a steeper number. A tire run 20 percent low loses about a fifth of its tread life, which it works through as 5,000 miles off a 25,000-mile tire. Twenty percent under a 35 psi placard is 7 psi, which the Goodyear rate would put nearer 12 percent. Two manufacturers, the same order of magnitude, a factor of less than two between them. The 50 percent figures that circulate on tire blogs sit outside both, and none of them name a study.
Fuel moves less than tread does. DOE and EPA put the savings from correct pressure at 0.6 percent on average. They give 3 percent as a worst case, which their own footnote defines as 25 percent under-inflation.
Why a standard tread check can miss it
Here is the part nobody writes down. The federal in-use tread inspection, 49 CFR 570.9, asks an inspector to look for wear indicators in any two adjacent major grooves. It puts those looks at three locations spaced approximately equally around the outside of the tire.
All three locations run around the circumference. Nothing in the procedure asks anyone to look across the width. On a four-groove tire, the two inner grooves are an adjacent pair, and so are the two outer ones. An inspection can comply in full while sampling one half of the tread and never touching the other.
Inflation wear is a difference between the middle and the edges. The check that decides whether a tire is legal is not built to see a difference, and 570.9 sets no pressure criterion for the tire either.
The eye is not a reliable backstop. No manufacturer, regulator or peer-reviewed source publishes how large a center-to-shoulder difference has to be before the pattern becomes visible. One across-width figure circulates, a 2/32 inch rule of thumb from inspection trade content, and it describes a measured difference with no stated basis. It says nothing about the eye. So the standard advice to watch for uneven wear rests on a detection point nobody has set.
Where TireBuddy fits
That gap is the reason Anyline built TireBuddy the way it did.
It reads the whole tread face
TireBuddy reads tread depth from the camera on a phone. It builds a depth model of the whole tread face, sampling roughly 100,000 points per tire at a precision down to 0.5 mm. A hand gauge reaches three to six points on that same tire. Before Discount Tire moved to digital tire inspections, its technicians measured as many as 12 spots per tire by hand. Twelve is still a thin sample of a gradient running across eight inches of rubber, and a differential is the first thing a thin sample rounds away.
It runs on the phone already in the bay
It runs on hardware the bay already owns. Anyline recommends the iPhone SE third generation and iPhone 11 upward, the Samsung Galaxy A53, S10, S21 and S23, and the Pixel 5 and above. For tablets, it recommends the iPad Pro third generation, the iPad Mini sixth generation, or the Galaxy Tab S8. There is a test in that worth putting to any vendor: ask which of the devices already in your workshop their scanner reads on. A quote for a handheld unit is a different answer to the one you asked for.
Every reading is dated, and it stays
Every reading is dated, attached to a vehicle, and kept. That turns a measurement into a record. You can show it to a customer, set it against the last visit, and explain why a tire that passed in spring will not pass in autumn. Coombes Johnston BMW runs Anyline inside its digital inspection workflow and now surfaces around 90 percent of its tire opportunities while the vehicle is still on site.
The technician still decides
Anyline does not decide that a tire is underinflated, and it does not diagnose the car. The technician does that. What changes is the evidence underneath. A profile across the whole tread face, with a date on it, instead of an impression formed at a glance. The judgment stays where it belongs, and the evidence under it stops being six looks in two grooves.
Frequently asked questions
Should tire pressure be 32 or 35?
Neither figure is automatically right. Tire pressure is set by the vehicle manufacturer, and the number is printed on the placard inside the driver’s door. Most passenger cars land in the low to mid 30s psi, so 32 and 35 are both common placard values. Only one of them applies to your car, and front and rear can differ. Set it cold, after the vehicle has been parked for at least three hours.
Will it ruin tires if you inflate them to 40 PSI instead of 35 PSI when the recommended PSI is 35?
Inflating to 40 psi on a car placarded at 35 will not ruin the tires, as long as 40 psi cold sits at or below the maximum molded into the sidewall. What changes is how the car behaves. The contact patch narrows, the ride firms up, and the center ribs wear ahead of the shoulders. Federal rules set a warning threshold for low pressure and none for high, which says where regulators put the risk.
What is the 4 PSI rule on tires?
The 4 PSI rule says a warm tire should read about 4 psi above its cold pressure, and that a larger rise means the tire was low. Cooper publishes it as a sealed-road guideline, with a 6 psi variant for four-wheel drive and light-truck tires. No standards body or regulator endorses it, and nobody names its origin. The rise depends on speed, load, road surface and outside temperature, none of which the rule accounts for.
What are the symptoms of underinflated tires?
Underinflated tires wear faster on both outer shoulders while the center stays deeper. They run hot, they take more fuel to turn, and the car can feel vague in corners. None of those signs is reliable on its own. NHTSA is blunt about the most obvious one: with radial tires, it is usually not possible to determine underinflation by visual inspection. Only a gauge will tell you.
Can I fill my own tires with air?
Yes, and a gas station air line or a home compressor will both do it. Check the pressure cold first, then inflate to the placard figure inside the driver’s door. Verify the result with your own gauge instead of the pump readout. A warm tire reads above its cold target, so do not set pressure from a warm reading. NHTSA says to top up if you suspect a low tire. Recheck and adjust once you can get a cold reading.
