System and Method for Predicting Trailer Tire Blowout Using Axle-Mate Differential Pressure and Temperature with ST Tire Age and Speed Derating, Hub Vibration, and Edge Inference
Two ideas in this disclosure are already patented: comparing tire pressure across an axle, and detecting tread separation from rotation-frequency vibration. What this disclosure claims instead is a narrower combination built for towed trailers, where the federal tire pressure rule does not reach. It combines pressure and temperature differentials against the axle-mate, tightened as the tire ages, with ST speed-rating enforcement. A silent-sensor fallback still catches a soft tire when its sensor goes silent.
Abstract
A proposed system would predict blowout of trailer tires before failure, using fused pressure, temperature, vibration, speed, and tire-age signals processed by edge inference on the trailer. A sensor on each wheel would measure internal tire pressure and temperature. An accelerometer at each hub would measure vibration. A trailer-mounted module would record speed from GPS and the manufacture date of each tire from its DOT date code. Each tire would be judged against its axle-mate, not against a fixed threshold. Two tires on one axle share the same road, the same speed, and nearly the same load. A tire running hotter or softer than its axle-mate is failing even when its absolute readings look normal. Two elements of that comparison are already patented art, and this disclosure concedes them: Goodyear's axle cross-comparison patent covers pressure differentials across an axle, and Continental's tread-separation patent covers wheel-speed-proportional vibration detection. What this disclosure adds is the combination that neither covers. It uses pressure and temperature differentials together, with thresholds that tighten as the tire ages. It enforces ST speed ratings, including the pressure caps that make higher speeds unachievable on common load ranges. And it confirms a soft tire from load shift when that tire's sensor has gone silent. All scoring would run on the trailer, with no cloud connection needed to warn the driver.
Field of the Invention
This invention relates to towed-trailer tire safety. It covers a trailer-specific combination of axle-mate differential sensing, tire-age and speed-rating derating, tread-separation vibration analysis, and silent-sensor fallback, with on-trailer edge inference and driver warning.
Background
Tires kill people in numbers that look small until they are yours. NHTSA's TireWise page reports 511 people died in tire-related crashes in 2024. A NHTSA pre-crash study from 2012, using crash data from 2005 to 2007 before tire pressure monitoring was required, estimated 414 fatalities, 10,275 injuries, and 78,392 crashes per year from flat tires or blowouts. Those two figures come from different eras and are not comparable. In the same study, 9 percent of sampled crashes involved tire problems in the pre-crash phase. Tires underinflated by more than 25 percent were about three times as likely to be cited as critical pre-crash events. Underinflation is the mechanism. A soft tire flexes more than it was built to. Flexing makes heat, and heat breaks down the rubber and the bond between rubber and steel belts until the tread comes off or the sidewall lets go.
Trailers get the worst of this and the least protection. The federal tire pressure monitoring rule, FMVSS 138, covers passenger cars and light trucks. It does not cover the trailer behind them. A tow vehicle can have a factory monitoring system on all four of its own wheels and be completely blind to the tires carrying the load it is pulling. Trailer tires make the blind spot worse in three ways. First, ST tires are load-rated on the assumption of a 65 mph maximum speed, and the sidewall load number is a 65 mph figure. A driver towing at 75 mph on the interstate is already outside the rating unless cold pressure was raised 10 psi, where the tire's maximum pressure allows it, and most drivers have never heard of the rule. Second, trailer tires age out rather than wear out. A trailer stands still most of the year, often in sunlight, and the rubber oxidizes whether it rolls or not. A tire with full tread and a six-year-old date code is a blowout waiting for a hot day. Third, the driver cannot feel a trailer tire going soft. A tow vehicle's own flat announces itself through the steering wheel. A trailer flat announces itself through sway, or through the sound of the tread coming off at highway speed. By that time the options are limited to holding the wheel straight and hoping the trailer stays behind the truck.
Patented Tire-Monitoring Art
Two load-bearing ideas in this field are already patented, and this disclosure does not claim them on their own. Goodyear's EP 4112335 B1 and US 20230001752 cover a tire pressure monitoring system employing axle cross comparison. The system determines a pressure leak by monitoring the pressure difference across two tires on opposite ends of the same axle. It accumulates the difference over time and alerts when the difference crosses a threshold. That patent also distinguishes slow leaks from rapid leaks by leak rate. Any claim here to axle cross-comparison of pressure alone would read on Goodyear's art.
Continental's EP 1308320 B1 and US 20030080857 cover determining the onset of tread separation by detecting periodic vibrations proportional to wheel rotational speed, superimposed on the wheel speed signal. Michelin's US 7,233,237 covers monitoring tire condition by computing the energies of a wheel-rotation signal at harmonics of the wheel round, from the 1st to the 64th harmonic. It compares those energies against a threshold to detect tread separation. GB 2406170A covers a tire-mounted accelerometer measuring tread acceleration to detect the vibration of a separating tread. Any claim here to rotation-frequency vibration detection of tread separation alone would read on that art.
What exists today in products measures around the problem rather than at it. Aftermarket trailer monitoring kits such as the TST 507 put a pressure and temperature sensor on each trailer wheel and show the readings on a dash display. They warn on fixed thresholds, typically a low-pressure limit and a high-temperature limit. They do not compare tires against each other, do not analyze vibration, do not know the tire's age, and do not know the speed rating. Factory tow-vehicle systems under FMVSS 138 monitor the tow vehicle's own tires and warn at 25 percent below placard pressure. They cover no trailer tire, and the 25 percent threshold is a late warning: a trailer tire 25 percent low has already been running hot for miles. Electronic sway control detects trailer oscillation through the tow vehicle's yaw sensors and brakes individual wheels to damp it. It reacts to sway after it starts. A blowout causes sway; sway control does not predict the blowout that caused it. Manual inspection, checking cold pressure before every trip and reading the DOT date code once a year, works when it happens. It does not happen on most trips, it cannot catch a leak that starts an hour into a drive, and it says nothing about what is happening inside the tire at speed.
A search of US and European patent art and of commercial trailer monitoring products found no system combining these elements. The combination has six elements:
- pressure and temperature differentials together against the axle-mate, with thresholds that tighten as the tire ages;
- ST speed-rating enforcement that accounts for the maximum-pressure caps;
- a load-shift confirmation that identifies a probable soft tire from its axle-mate's temperature rise when the soft tire's own sensor has gone silent;
- scoring rejection rules for bad data;
- on-trailer edge inference with no cloud connection; and
- a sway precursor from the trailer module's own accelerometer.
One concession is needed on scope. Michelin's US 7,233,237 expressly contemplates trailers and caravans, so the trailer distinction here rests on the FMVSS 138 gap and the ST-specific packaging, not on trailer application alone. That combination, packaged as a valve-stem-sensor kit for towed trailers with no tow-vehicle modification, is the gap this disclosure addresses.
Detailed Description
This section describes a proposed system. No prototype exists. Every threshold, rate, and cost figure in it is a design target, not a measured result. Conditional language is used deliberately throughout.
1. System Overview
Four parts would make up the system. A sensor on each trailer wheel would measure internal pressure and temperature: either a valve-stem cap sensor that screws onto the existing stem, or an internal sensor strapped to the wheel inside the tire. An accelerometer at each hub would measure vibration at that wheel. A trailer-mounted module, powered by the trailer's 12 volt system with a small solar panel and battery backup, would collect the sensor data over a short-range radio link. It would read speed from its own GPS receiver and ambient temperature from a sensor on the module, and run the scoring. A phone application in the tow vehicle, configured to communicate with the module over Bluetooth Low Energy and to announce warnings by voice through the phone's speaker, would show each tire's state and issue the warning. No part of the tow vehicle would be modified, no cloud connection would be needed to score a tire, and no sensor would be required on the tow vehicle itself. A second accelerometer inside the trailer module would measure low-frequency lateral motion of the trailer body for the sway precursor described in section 7.
2. Axle-Mate Differential: The Axle-Mate as Control
Each tire would be judged against its axle-mate. Two tires on one axle roll over the same road at the same speed, carry nearly the same load, and sit in nearly the same air. A failing tire diverges fast and far. For each axle pair, the system would compute the differential pressure and differential temperature: the difference between the two tires after correcting each reading to a common reference temperature.
Temperature correction matters because pressure follows temperature. A tire at 65 psi cold at 80 degrees Fahrenheit reads about 77 psi when its internal air reaches 160 degrees on a hot highway run. The arithmetic: absolute pressure 79.7 psi times the ratio of absolute temperatures, 620 over 540 Rankine, gives 91.5 psi absolute, or about 77 psi gauge. That is a rise of about 12 psi from heat alone. A threshold system set to warn at 75 psi would flag a healthy hot tire and miss a failing one. The differential system would correct each tire's pressure to 80 degrees using the ideal gas law, then compare the corrected pressures.
Here is a worked example with numbers that actually produce a warning. Take an ST225/75R15 Load Range D tire, rated 2,540 pounds at 65 psi. Suppose the failing tire reads 62 psi at 150 degrees Fahrenheit, while its axle-mate reads 65 psi at 110 degrees. Correcting both to 80 degrees gives about 53 psi for the failing tire and about 61 psi for its axle-mate. The corrected differential is about 8 psi, above the 5 psi first-warning threshold. The raw readings alone would not show it: 62 versus 65 psi looks like a 3 psi difference, and both temperatures look plausible on a hot day. Only the temperature-corrected comparison reveals the leak.
First-warning thresholds, as design targets, would be a corrected differential above 5 psi, or a temperature differential above 15 degrees Fahrenheit sustained for 10 minutes. Pull-over thresholds would be a corrected differential above 10 psi, or a temperature differential above 30 degrees. Age would tighten those thresholds, which is the concrete interaction between the age and differential elements. A tire past five years from its DOT date would get first-warning thresholds of 3 psi and 10 degrees, because aged rubber fails with less warning than new rubber. Those numbers are design targets, pending measurement.
Sun side and road crown would complicate the comparison, and the design accounts for them rather than pretending they cancel out. On a long westbound afternoon drive, the sun-side tires run hotter than the shaded side for hours. A crowned road loads the curbside tires differently from the roadside tires. Neither offset is measured in any published source this disclosure relies on, so the system would estimate the sun-side and crown offsets during the first 30 minutes of each trip and subtract them from the differential. The estimate would be the median temperature differential over minutes 10 to 30 of the trip, after the 10-minute scoring suspension ends. The sun side would be identified from GPS heading and time of day. Sun offset and crown offset would not be separated; the combined offset would be subtracted. When the estimate is unstable, changing faster than 5 degrees per 10 minutes as a design target, the temperature differential would be suppressed and scoring would rely on the corrected pressure differential and vibration alone. The system would be degraded on long sun-side stretches, not inoperable, and the app would say which channels are active. That fallback is an untested expectation. No data shows pressure-plus-vibration scoring suffices when temperature is suppressed.
Single-axle trailers have no axle-mate, so the system would fall back to a weaker comparison. The tire would be scored against its own rolling history on the same trip. It would also be scored against a model of expected temperature given speed, ambient temperature from a sensor on the module, and load. Differential thresholds would widen for single-axle trailers, and the app would say so, because the comparison is weaker. Tandem and triple-axle trailers would get the full differential treatment, with each axle pair scored independently and the axle pairs cross-checked against each other.
3. Tread Separation Precursor: Once-Per-Revolution Vibration
A separating tread would not fail silently. Before the tread leaves the tire, the belt package loosens and the tire stops being round. Each revolution would then produce a pulse as the loose section passes the contact patch. The system would look for that pulse at the wheel rotation frequency, which falls out of the tire geometry and the speed. Hub mounting is the distinction from the tire-mounted accelerometer art: the sensor measures the wheel assembly's vibration through the hub, needs no tire dismount, and works alongside a valve-stem pressure sensor.
An ST205/75R15 tire has a diameter of 15 inches plus twice the sidewall height, 205 millimeters times 0.75. That is 381 plus 307.5 millimeters, or 0.689 meters. Its circumference is about 2.16 meters. At 65 mph, 29.06 meters per second, the tire turns 29.06 divided by 2.16, about 13.4 times per second. A once-per-revolution fault would therefore appear as a vibration line at about 13.4 Hz at 65 mph. That figure is nominal, from unloaded geometry. Loaded rolling radius differs from nominal by a few percent, so the search band would be widened to plus or minus 5 percent of the nominal frequency, and the band would track speed proportionally as speed changes: about 11.4 Hz at 55 mph, about 15.5 Hz at 75 mph. The module would compute the expected rotation frequency from GPS speed and the tire size entered at setup, then measure vibration energy in that band and its second harmonic. The second harmonic is the vibration at twice the rotation frequency, which a once-per-revolution fault also produces.
A healthy tire shows little energy at its rotation frequency. A tire with a separating tread, a slipped belt, or a flat spot would show a rising line at that frequency. The line would grow over minutes to tens of minutes before failure, and a temperature rise in the same tire would accompany it. That combination is what would separate a tread fault from a merely rough road. A rough road shakes every tire at once, while a separating tread shakes one tire at its own rotation rate and heats it at the same time. Whether the line actually grows over minutes is a hypothesis with no data behind it; no prototype has measured it. The accelerometer would sample at 500 Hz, which is enough to resolve the rotation line and its harmonics up to the speeds a trailer should be traveling. Frames would be rejected when the trailer is turning sharply or crossing a bridge expansion joint, both of which inject broadband vibration that is not a tire fault.
4. Pressure Loss Modes and Load Shift
Not all pressure loss is the same, and the response would depend on which kind it is. A slow leak, from a valve stem or a nail, would show as a steady corrected-pressure decline of 1 to 3 psi per hour in one tire with no temperature anomaly at first. Those rates are design targets. The correct action would be to plan a stop at the next exit. A rapid leak, from a puncture or a valve failure, would show as a drop of more than 5 psi in one minute. The correct action would be to slow down and pull over now. A blowout itself would show as a pressure collapse to near atmospheric plus a burst of broadband vibration, and the system would log it as a failure event for the tire's record even though the warning had come too late to prevent it.
Load shift is the secondary signal that would confirm a soft tire and catch the case where the sensor itself has failed. On a tandem-axle trailer, when one tire loses pressure, its share of the load moves to its axle-mate and to the other axle. The axle-mate's temperature rises measurably as it carries more than its share, and its raw pressure rises slightly as it is squeezed harder. Corrected pressure cannot rise from load; only temperature and raw pressure can, and the disclosure corrects that point explicitly. The system would watch for the paired pattern: one tire's corrected pressure falling while its axle-mate's temperature rises more than 10 degrees Fahrenheit above its own trip baseline, as a design target. The trip baseline is the axle-mate's median temperature during minutes 10 to 30 of the trip. That paired pattern would be treated as a probable soft tire even if the falling tire's pressure and temperature sensor has gone silent, with its hub accelerometer still reporting. A sensor that stops reporting is itself a fault. The system would flag it, fall back to the axle-mate's behavior and the vibration channel, and tell the driver which wheel is unmonitored rather than silently showing a stale green light.
5. Tire Age and Speed-Rating Enforcement
At setup, the owner would photograph or enter the DOT date code from each tire's sidewall: four digits, week and year of manufacture, so 3223 means the 32nd week of 2023. The module would compute each tire's age and apply the replacement rule. Goodyear's guidance for RV tires is replacement at 6 years in service, or 6 years from the DOT manufacture date if the in-service date is unknown. This disclosure uses the DOT date as the proxy and flags a tire for replacement at five years from that date, with a do-not-tow advisory at six years. That advisory is advisory only; the system has no way to stop a driver from towing.
Speed enforcement would work from the same setup data. For ST tires without an explicit speed symbol, the rating is 65 mph. The module would compare GPS speed against the lowest speed rating on the trailer and warn when the driver exceeds it, with the warning escalating if the speed is held. The Tire and Rim Association guidance allows 66 to 75 mph only with cold inflation raised 10 psi above the placard, and 76 to 87 mph only with that pressure increase plus a 10 percent load reduction. Both conditions are capped: never above the maximum pressure molded on the tire or the wheel, and never above the speed symbol on the sidewall. Those caps matter because they make the higher-speed conditions unachievable on common tires. An ST225/75R15 Load Range D tire carries 2,540 pounds at 65 psi, and 65 psi is its maximum pressure. The plus 10 psi condition for 66 to 75 mph would require 75 psi cold, which that tire cannot take. On a trailer fitted with those tires, 65 mph is the effective limit, and the app would say so rather than offering a higher speed the tires cannot support.
6. Edge Classifier and Warning Levels
A gradient-boosted tree ensemble, or a small quantized neural network of equivalent size, would map the feature vector to two outputs: a tire health index from 0 to 100 for each wheel, and a fault class. The index scale is defined as 80 and above for green, 50 to 79 for amber, and below 50 for red. Those cutoffs are design targets. Features would include corrected differential pressure and temperature against the axle-mate, absolute temperature, pressure trend over the last 30 minutes, rotation-frequency vibration energy and its growth rate, speed relative to the tire's rating, tire age in years, and load-shift indicators from the axle-mate. Model size under 100 kilobytes and inference time under 5 milliseconds per scoring cycle on a Cortex-M class microcontroller are both design targets, not measured results. Scores would update every 30 seconds while rolling, also a design target.
The channels would also gate each other, which is where the combination does work no single channel can do alone. When the corrected pressure differential is above the first-warning threshold, the vibration growth-rate threshold would drop by half and the search band would narrow, because a soft tire is more likely to be separating. When the differential is clean, the vibration threshold would rise to suppress false alarms from rough roads. A red warning for tread separation would require both channels to agree: a rising rotation-frequency line plus a temperature differential in the same tire. Neither channel alone could trigger a red separation warning.
Warnings would be graded so the driver can act on them. Green means all tires within differential bounds. Amber means a tire is diverging from its axle-mate or its own history: plan a stop at the next exit, check pressure by hand, and do not increase speed. Red means pull over at the next safe shoulder: a corrected differential above the pull-over threshold, a rapid leak, a rising rotation-frequency vibration line, or a tire past its age limit running at highway speed. The phone would announce red warnings by voice, since the driver should not be looking at a phone while towing. The module would also carry a small LED and buzzer on the trailer itself, so a driver doing a walk-around at a fuel stop can see a red tire without opening the app.
7. Tow-Vehicle Coupling and Sway Precursor
A trailer tire failure often shows up first in the tow vehicle, not the trailer. As a tire goes soft, the trailer starts to wander, and the driver corrects with small steering inputs that grow over minutes. This system would not instrument the tow vehicle, so it cannot measure steering directly. It would approximate the coupling from its own sensors. A soft tire increases rolling resistance on one side, which yaws the trailer slightly, and that yaw would show up as a low-frequency lateral oscillation in the trailer module's own accelerometer, below 2 Hz as a design target, growing in amplitude as the tire worsens. The system would track that oscillation as a sway precursor feature and raise the warning level when it grows alongside a tire differential. This is a proxy, not a sway measurement, and the disclosure says so. It would never claim to replace a weight-distribution hitch, a sway control device, or the driver's judgment about when a trailer is no longer safe to tow.
8. Fleet Learning and Tire Profiles
On opt-in, anonymized feature vectors and outcomes would sync from the phone to a fleet service. The service would refine the classifier from verified outcomes: tires that owners report as blown, separated, or replaced after a warning, and tires that ran to their age limit without incident. It would also build tire profiles: the expected temperature and pressure behavior of common ST tire sizes and load ranges, so a new installation starts from a reasonable model instead of a blank one. Correlated failures across many trailers using the same tire model would indicate a bad production batch or a defective size. No location data would leave the device. Only feature vectors and outcome labels would sync, and only with explicit opt-in. That claim needs a qualifier: the fleet service would still receive timestamps and coarse trip durations, which can reveal travel patterns, so "no location data" means no GPS coordinates, not no information about movement. Fleet data would also carry selection bias, since owners who opt in and report outcomes are not a random sample of towers.
9. Scoring Rejection Rules
Scoring would be suspended when the data cannot support it. The module would reject scoring during the first 10 minutes of a trip, while tire temperatures are still rising toward equilibrium. It would reject frames during turns with lateral acceleration above 0.15 g and on bridge expansion joints, which inject broadband vibration. It would reject scoring on sustained grades, defined as a grade above 4 percent held for more than 2 minutes, where load distribution across axles shifts and the differential baselines move. Those turn and grade thresholds are design targets. It would reject a calibration or baseline update when feature variance exceeds a bound. Each rejection would be logged with its reason, and the app would show scoring as paused rather than green. These rejection rules are part of the claimed system, because a score produced from bad data is worse than no score.
10. Figures Description
- Figure 1: System architecture showing the per-wheel pressure and temperature sensors, hub accelerometers, trailer-mounted module with GPS and its own accelerometer, and the Bluetooth link to the phone application in the tow vehicle.
- Figure 2: Axle-mate differential diagram for a tandem-axle trailer, showing the same-axle pair comparison, the temperature correction to a common reference, the worked example with a failing tire at 62 psi and 150 degrees against an axle-mate at 65 psi and 110 degrees, and the warning thresholds on the corrected differential.
- Figure 3: Pressure-temperature calculation showing how a healthy tire gains about 12 psi from heating alone on a highway run, and why absolute pressure thresholds cannot separate a hot healthy tire from a failing one.
- Figure 4: Vibration spectrum for a healthy tire and a tire with a separating tread, showing the rotation-frequency line at a nominal 13.4 Hz at 65 mph for an ST205/75R15 tire, the plus or minus 5 percent search band, and its growth over time before failure.
- Figure 5: Load-shift diagram showing how a soft tire's load moves to its axle-mate and the other axle, with the axle-mate's temperature rise confirming the soft tire even when its sensor is silent.
- Figure 6: Tire health index decay over a trip for a single tire, with the amber plan-a-stop and red pull-over thresholds marked, plus the age-dependent tightening of the differential thresholds and the speed-rating derating applied at setup.
In patent language, that system is:
Claims
- A system for predicting blowout of towed-trailer tires, comprising: a pressure and temperature sensor on each wheel of a towed trailer; a vibration sensor at each hub of the trailer; a trailer-mounted module containing a GPS receiver, a module accelerometer, an ambient temperature sensor, and a processor; and a phone application configured to communicate with the module over Bluetooth Low Energy and to announce warnings by voice through the phone's speaker; wherein a scoring unit in the module computes, for each tire, a corrected pressure differential relative to the tire's axle-mate on the same axle, each tire's pressure having been corrected to 80 degrees Fahrenheit using the ideal gas law, and a temperature differential relative to the axle-mate, the temperature differential having been adjusted for sun-side and crown offsets estimated during the trip; applies a first-warning differential threshold of 5 psi corrected pressure or 15 degrees Fahrenheit and a pull-over differential threshold of 10 psi corrected pressure or 30 degrees Fahrenheit, the thresholds being tightened to 3 psi and 10 degrees Fahrenheit when the tire is past five years of age from its DOT date code; combines the differentials, rotation-frequency vibration energy, the tire's age from its DOT date code, and the tire's speed rating into a tire health index from 0 to 100, the index being 80 and above for green, 50 to 79 for amber, and below 50 for red; and issues a graded warning through the phone application when the index crosses a threshold.
- A system for predicting blowout of towed-trailer tires, comprising: a pressure and temperature sensor on each wheel of a towed trailer; a trailer-mounted module containing a GPS receiver and a processor; a load-shift detector; and a speed-rating unit; wherein a scoring unit computes, for each tire, a corrected pressure differential and a corrected temperature differential relative to the tire's axle-mate on the same axle, each tire's pressure having been corrected to a common reference temperature using the ideal gas law, and issues a graded warning when either corrected differential crosses a threshold that tightens when the tire is past five years of age from its DOT date code; wherein the load-shift detector identifies a probable soft tire from a paired pattern of one tire's corrected pressure falling while its axle-mate's temperature rises; and wherein the speed-rating unit compares GPS speed against the lowest speed rating on the trailer and warns the driver when the speed exceeds the rating, subject to the maximum pressure molded on the tire or wheel and the speed symbol on the sidewall.
- A system for predicting tread separation in a towed-trailer tire, comprising: a vibration sensor at a hub of the trailer; a temperature sensor on the wheel; a setup input that records the tire size; a trailer-mounted module containing a GPS receiver and a processor; wherein a scoring unit computes an expected wheel rotation frequency from GPS speed and tire circumference, measures vibration energy in a band of plus or minus 5 percent around the rotation frequency and its second harmonic, and classifies a rising rotation-frequency vibration line accompanied by a temperature rise in the same tire as a tread separation precursor.
- The system of claim 1, wherein the scoring unit corrects each tire's pressure to 80 degrees Fahrenheit using the ideal gas law before computing the corrected pressure differential, so that pressure changes caused by heating alone do not produce a false warning.
- The system of claim 3, wherein the expected wheel rotation frequency is computed as GPS speed divided by tire circumference, the circumference being derived from the tire size recorded by the setup input, and wherein the scoring unit rejects vibration frames recorded during turns with lateral acceleration above 0.15 g and on bridge expansion joints.
- The system of claim 1, further comprising a load-shift detector that identifies a probable soft tire from a paired pattern of one tire's corrected pressure falling while its axle-mate's temperature rises more than 10 degrees Fahrenheit above the axle-mate's own trip baseline, the trip baseline being the axle-mate's median temperature during minutes 10 to 30 of the trip, and that, when the falling tire's pressure and temperature sensor has stopped reporting, the hub accelerometer still reporting, identifies the probable soft tire from the axle-mate's temperature rise together with a rising once-per-revolution vibration signal from the falling tire's hub vibration sensor.
- The system of claim 1, wherein the scoring unit applies leak-rate bands to the corrected pressure differential, a slow leak being a corrected-pressure decline of 1 to 3 psi per hour and a rapid leak being a corrected-pressure decline of more than 5 psi in one minute, and issues a plan-a-stop warning for a slow leak and a pull-over warning with a voice announcement for a rapid leak.
- The system of claim 1, further comprising a sway precursor detector that measures low-frequency lateral oscillation below 2 Hz in the module accelerometer and raises the warning level when the oscillation grows in amplitude above its trip baseline alongside a tire differential.
- The system of claim 1, further comprising a tire age and speed unit that stores the manufacture date of each tire from its DOT date code, flags a tire for replacement at five years from the DOT date, issues a do-not-tow advisory at six years from the DOT date, stores the speed rating of each tire, compares GPS speed against the lowest speed rating on the trailer, and warns the driver when the speed exceeds the rating, reporting whether the current setup meets the conditions for higher speeds: cold inflation raised 10 psi above the placard for 66 to 75 mph, and that pressure increase plus a 10 percent load reduction for 76 to 87 mph, in each case subject to the maximum pressure molded on the tire or wheel as a pressure cap and the speed symbol on the sidewall as a speed cap.
- The system of claim 1, further comprising a single-axle fallback in which a tire with no axle-mate is scored against its own rolling history on the same trip, the rolling history being the median corrected pressure and temperature recorded during minutes 10 to 30 of the same trip, and a model of expected temperature given speed, ambient temperature, and load, with first-warning thresholds widened to 8 psi and 25 degrees Fahrenheit.
- The system of claim 1, further comprising a fleet learning unit that, on opt-in, syncs feature vectors and verified outcomes with direct identifiers removed to a fleet service, the fleet service refining the scoring unit and detecting correlated failures across trailers using the same tire model.
- The system of claim 1, further comprising a sensor-failure unit that flags a sensor that has stopped reporting, falls back to the axle-mate's behavior and the hub vibration sensor for that wheel, and reports the wheel as unmonitored rather than showing a stale satisfactory state.
- The system of claim 1, further comprising scoring rejection rules that suspend scoring during the first 10 minutes of a trip, during turns with lateral acceleration above 0.15 g, on bridge expansion joints, and on sustained grades above 4 percent held for more than 2 minutes, and that log each suspension with its reason to the module's memory and report scoring as paused in the phone application.
- The system of claim 1, further comprising a trailer-mounted LED and buzzer that display the warning level on the LED and sound an audible alert on the buzzer for a driver doing a walk-around without opening the phone application.
- A method for predicting towed-trailer tire blowout, comprising: measuring pressure and temperature at each wheel of a towed trailer; measuring vibration at each hub; measuring speed by GPS; recording each tire's manufacture date from its DOT date code and each tire's speed rating; correcting each tire's pressure to 80 degrees Fahrenheit using the ideal gas law; computing, for each tire, a corrected pressure differential and a corrected temperature differential relative to its axle-mate on the same axle; computing an expected wheel rotation frequency from GPS speed and tire circumference and measuring vibration energy in a band of plus or minus 5 percent around that frequency and its second harmonic; combining the differentials, the vibration energy, the tire's age, and the tire's speed rating into a tire health index from 0 to 100, the index being 80 and above for green, 50 to 79 for amber, and below 50 for red, using a first-warning differential threshold of 5 psi corrected pressure or 15 degrees Fahrenheit and a pull-over differential threshold of 10 psi corrected pressure or 30 degrees Fahrenheit, the thresholds being tightened to 3 psi and 10 degrees Fahrenheit when the tire is past five years of age; and issuing, through a phone application configured to communicate with the trailer module and announce warnings by voice through the phone's speaker, a graded pull-over warning with a voice announcement when the index falls below 50.
- The system of claim 1, wherein the scoring unit lowers the vibration growth-rate threshold by half when the corrected pressure differential is above the first-warning threshold, raises the vibration threshold when the corrected pressure differential is within bounds, and requires agreement between the hub vibration sensor and the temperature differential before issuing a pull-over warning for tread separation.
Implementation Notes
Bill of materials for the per-trailer kit targets $130 at volume: four valve-stem pressure and temperature sensors at about $15 each, four hub accelerometers at about $5 each, a trailer module with GPS, Cortex-M microcontroller, module accelerometer, and Bluetooth radio at about $30, and a solar panel with battery backup at about $20. Four sensors at $15 is $60, four accelerometers at $5 is $20, plus $30 and $20 gives $130. Valve-stem sensors are the install path that matters, because they require no tire dismounting and no shop visit. Internal sensors read temperature more accurately but need a tire machine, so they are the fleet option. The classifier fits in under 100 kilobytes and runs in under 5 milliseconds per scoring cycle, both targets. The phone application shows each tire as a single number with a color state, the fault class in plain words, and a one-line instruction. It stores per-tire history locally and syncs only on opt-in. Sensor batteries are rated for about two years and are replaceable by the owner. The module mounts on the trailer frame near the axles, inside a sealed IP65 housing, with the GPS antenna facing up and the solar panel facing the sun side of the trailer.
Setup discipline matters more than sensor quality. The tire size, load range, placard pressure, and DOT date code for each wheel must be entered correctly, because the rotation frequency, the expected temperature, and the age flag all derive from those entries. The app enforces this by refusing to score a wheel whose date code is missing and by asking the owner to confirm the size from the sidewall, not from memory. Cold pressure must be set with a gauge before the first scored trip, because the differential system needs a correct starting point. A tire that starts 10 psi low and stays 10 psi low looks like a stable tire to a trend detector, and only the axle-mate comparison catches it. The module must also reject scoring during the first 10 minutes of a trip, while tire temperatures are still rising toward equilibrium, and during sustained grades, where the load distribution across axles shifts and the differential baselines move. These rejection rules are what keep the score honest.
Limitations
This disclosure describes an unbuilt system. No prototype has been constructed, no accuracy figure has been measured, and every threshold in it is a proposed value. The 5 psi and 15 degree first-warning differentials, the 10 psi and 30 degree pull-over differentials, the tightened 3 psi and 10 degree thresholds for tires past five years, the 30 second scoring interval, the 100 kilobyte model size, the 5 millisecond inference time, and the $130 bill of materials are design targets, not measured results.
No false-positive or false-negative rate is known, because no system has been built and no tires have been driven to failure under observation. Abrupt failures have no precursor and will be missed: a tire that hits debris, suffers a valve stem failure, or takes a curb strike can fail with no warning from any channel in this system. False red warnings carry their own hazard, since they require a driver to pull onto a highway shoulder, and repeated false warnings would train drivers to ignore the system. The labeled training dataset the classifier needs does not exist yet at scale. Collecting it requires instrumenting trailers, driving them until tires fail, and labeling the failures by teardown. Blowouts are rare events, so the dataset would suffer severe class imbalance. As an illustrative order-of-magnitude estimate, thousands of healthy trips would stand behind every failure. That imbalance makes classifier training hard and validation harder. That collection effort is the largest barrier to building this system.
Sun side is the hardest confound, and this disclosure does not resolve it. The sun-side temperature offset is unmeasured in any source this disclosure relies on, and the per-trip learning rule in section 2 is a proposed mitigation, not a tested one. Road crown is a second confound: a crowned road loads the curbside tires differently from the roadside tires, which shifts the differential baseline on every road. Single-axle trailers have no axle-mate at all, so their warnings rest on a weaker comparison and will false-alarm more often. Per-wheel vibration isolation on a shared axle and frame is unproven. Cross-talk between wheels, unbalanced wheels, road texture, GPS speed error, and a wrong tire-size entry can all shift or mimic the rotation line. The "minutes to tens of minutes" growth of the separation line is a hypothesis with no data.
Sensor error is unbudgeted. A valve-stem sensor measures temperature at the stem, not the internal air temperature, and it lags behind the air temperature during rapid heating. No accuracy, drift, or dropout figures exist for the sensors in this design. Bluetooth dependence and phone dependence are single points of failure: if the phone dies or the link drops, the driver loses the voice warning, and the trailer LED and buzzer are the only remaining channel. Setup error is a failure mode of its own: a mistyped DOT date code, a wrong tire size, mixed brands or ages on one axle, a wheel swap that is never entered, or unequal tandem loading all corrupt the baselines. The do-not-tow advisory is advisory only. Finally, the system predicts tire failure, not trailer accidents. Tongue weight, hitch setup, and driver speed cause more trailer accidents than tires do, and no tire gauge fixes a trailer that is loaded tail-heavy.
Strongest Case Against
The honest objection is that a pressure gauge and a calendar already solve this problem, and a sensor network is an expensive way to replace a two-minute walk-around. Cold pressure checked before every trip, the DOT date code read once a year, and the speed held to 65 mph would prevent many trailer blowouts, on the assumption that most failures come from the conditions those checks catch. That assumption is untested. A system that needs four sensors, four hub accelerometers, a GPS module, and a phone app to tell a driver what a $15 gauge already says is a solution in search of a problem, and the drivers who need it most, the owners who tow twice a year and never check anything, are the least likely to buy it, install it, or enter their date codes correctly. No test of this system exists, so that objection stands as a reason not to build it. The argument for building it is a hypothesis: that most blowouts develop during the trip, when a walk-around cannot see them. No data in this disclosure tests that.
A second objection is the environment. A trailer lives outside, in rain, dust, and road salt. Valve-stem sensors get knocked off, corrode, and report stale data, while the phone in the tow vehicle loses its Bluetooth link exactly when the driver needs the warning. The differential design is designed to survive a dead sensor by falling back to the axle-mate's behavior, but that fallback has never been tested either. A third objection is behavioral: a green light could give false confidence, and false confidence at 70 mph with a six-year-old tire is worse than no gauge at all, because the driver who trusts the gauge stops doing the walk-around. That objection also stands. If a gauge caught some blowouts early, it would beat a gauge that only works in the driveway. But that is conditional on the gauge working, which is unproven, and on nobody mistaking the gauge for a safety device, which nothing in this disclosure can guarantee.
What a Trailer Owner Can Do Now
None of this needs to be built before it helps. Read the DOT date code on every trailer tire: the last four digits of the DOT number on the sidewall are the week and year the tire was made, so 3223 means the 32nd week of 2023, and any tire past five years is on borrowed time, past six years it should not be towed on at all, no matter how good the tread looks. Set cold pressure with a gauge before every trip, to the pressure on the trailer's placard. Do not use the number on the sidewall, and do not set pressure by feel. Check pressure again at every fuel stop with the back of your hand on each tire. A tire that is noticeably hotter than its axle-mate is the one to worry about, even if it looks fine. Hold 65 mph with ST tires unless the sidewall shows a higher speed rating and the pressure has been raised 10 psi to match, and remember that a tire whose maximum pressure is 65 psi cannot meet that condition at all. Weigh the loaded trailer and check the number against the tires: a common ST205/75R14 Load Range C tire carries 1,760 pounds, and four of them carry 7,040 pounds total, which is less than many loaded travel trailers weigh. Keep a torque wrench in the kit and check the lug nuts at 50 miles after any wheel change, because a loose wheel kills a tire just as surely as a nail does. Carry a real spare, inflated and with a date code you have also checked, plus the jack and wrench that actually fit the trailer. Practice changing the tire in the driveway once, so the first time is not on a highway shoulder at dusk.
Know someone who tows a travel trailer on tires they have never checked the date code on? Send them this.
Prior Art References
- NHTSA TireWise: 511 people died in tire-related crashes in 2024; tire maintenance guidance including monthly cold pressure checks and DOT date code reading
- NHTSA, Tire-Related Factors in the Pre-Crash Phase (DOT HS 811 617, April 2012), Choi, E-H.: 2005 to 2007 NMVCCS data; 414 fatalities, 10,275 injuries, and 78,392 crashes per year from flats and blowouts before tire pressure monitoring; 9 percent of sampled crashes involved tire problems in the pre-crash phase; tires underinflated by more than 25 percent about three times as likely to be cited as critical pre-crash events
- 49 CFR 571.138 (FMVSS 138): Tire pressure monitoring systems required on passenger cars and light trucks; trailers are not covered by the mandate
- 49 CFR 393.75: Federal Motor Carrier Safety Administration tire rules for commercial vehicles, including minimum tread depths and prohibitions on tires with tread or sidewall separation or audible leaks
- Tire and Rim Association guidance for ST tires, as summarized by a trailer tire load calculator: ST load ratings calculated for a maximum of 65 mph; 66 to 75 mph requires cold inflation raised 10 psi; 76 to 87 mph requires the pressure increase plus a 10 percent load reduction; no rating above 87 mph; ST225/75R15 Load Range D rated 2,540 lb at 65 psi
- Trailer tire speed ratings and replacement intervals, manufacturer FAQ: ST tires without a speed symbol are rated 65 mph; speed symbols J, L, M, and N correspond to 62, 75, 81, and 87 mph; most trailer tires should be replaced every 5 to 6 service years even if tread remains
- RV tire safety guide summarizing Goodyear guidance: Goodyear recommends replacing any RV tire in service that was made more than 6 years earlier, based on the DOT date code; the same guide describes a tire pressure monitoring system with sensors on each wheel as one of the most useful safety add-ons for any RV
- US 20230001752 / EP 4112335 B1 (Goodyear): Tire pressure monitoring system employing axle cross comparison; determines a pressure leak by monitoring the pressure difference across two tires mounted on opposite ends of the same axle, accumulating the difference over time, and distinguishing slow leaks from rapid leaks by leak rate
- US 20030080857 / EP 1308320 B1 (Continental): Process for determining the onset of tread rubber separations by detecting periodic vibrations proportional to wheel rotational speed, superimposed on the wheel speed signal
- US 7,233,237 (Michelin): Method and system for monitoring tire condition by computing the energies of a wheel-rotation signal at harmonics of the wheel round, from the 1st to the 64th harmonic, and comparing them against a threshold to detect tread separation
- GB 2406170A: Tire monitor with an accelerometer measuring acceleration of the tire tread to detect the vibration of a separating tread
- LITF-PA-2026-018: Continuous road surface friction estimation via tire-road acoustic emission; related tire sensing, distinct target and no per-tire pressure or temperature measurement
- LITF-PA-2026-065: Non-contact tire tread depth estimation via wet-road spray plume computer vision; related tire condition sensing, distinct target and no blowout prediction
- LITF-PA-2026-237: Chainsaw chain wear detection via acoustic, vibration, and motor current signatures with per-unit baseline learning; related multimodal wear sensing and baseline method, distinct machine and failure mode
- LITF-PA-2026-208: Bicycle drivetrain chain wear detection via acoustic and cadence analysis; related acoustic wear sensing on rotating components, distinct target and signal environment