Method of testing a tyre

Abstract

A method (99) and apparatus for testing a tire (70), the method (99) comprising: - mounting (1) the tire (70) on a rim and inflating (30) the tire (70) to a certain pressure; - rotating (2) the tire (70) about a rotation axis (101), wherein the tire (70) is subjected to thrust on a rolling surface (74); - acquiring (3) an acceleration signal representing the linear acceleration experienced by the tire (70) during the rotation; - calculating at least one spectrum (4) of the acceleration signal; - calculating at least one exponent (5) based on the at least one spectrum; - evaluating (6) the integrity status of the tire (70) based on the time trend of the at least one exponent.

Description

Technical Field

[0001] This invention relates to a method for testing tires, such as a rolling test. Background Technology

[0002] Typically, vehicle tires have a substantially annular structure around the tire's axis of rotation during operation and have an equatorial plane perpendicular to the axis of rotation. This equatorial plane is generally a (substantially) geometrically symmetrical plane (e.g., without considering any minor asymmetries such as tread patterns and / or sidewall lettering and / or structural or profile asymmetry).

[0003] The terms "radial" and "axial" refer to directions that are substantially perpendicular and substantially parallel to the axis of rotation of the tire, respectively.

[0004] The term "tangential" refers to a direction that is substantially perpendicular to both the radial and axial directions (e.g., generally oriented in accordance with the rolling direction of the tire).

[0005] In the context of industrial tire production, it is known that tires are tested or inspected during the prototype stage or production to verify whether they meet certain requirements (such as quality, reliability, safety, performance, durability, etc.).

[0006] Some of these tests are designed to measure the duration a tire remains in its intact state while rolling under certain conditions, which are typically more demanding than those expected under normal use conditions. The duration a tire remains in its intact state (i.e., without any defects) represents the performance achieved in terms of both the absolute value of the duration and the duration relative to a predetermined threshold (such as several hundred hours).

[0007] Such rolling tests, which are typically destructive or unrepeatable under any circumstances, can be performed on sample tires representing a particular category of tires (e.g., characterized by a particular model and / or size and / or manufacturing process and / or formulation of compounds used) for purposes such as certification before the relevant category of tires is put on the market, and / or to verify compliance during the production of tires in that category, for example, to ensure that the process does not cause unacceptable deviations in the finished tires.

[0008] For the rolling test, the tire is initially mounted on the rim and inflated. The rim is then fixed to a rotating (usually idling) hub, and the tire is rotated on a rolling surface (such as one made of a powered roller). The tire is subjected to a load, i.e., a thrust on the rolling surface.

[0009] An example of a rolling test is the FMVSS 109 / 135 durability test, which specifies that the tire be kept rotating at a constant angular velocity for a certain period of time (typically tens of hours) under constant internal pressure and load applied to the tire (typically, the load is greater than the load the tire must withstand under normal use conditions). The tire's rotational speed, load, and pressure are determined based on the tire's characteristics, such as size, load index, or speed index.

[0010] Every so often, such as every 12 hours, the tires are stopped for visual inspection. If a visible defect (symptom of tire damage) is found, the rolling test is stopped and the tire is thoroughly inspected to classify the damage; otherwise, the test continues. The test is considered passed if no defects are found when a predetermined duration threshold (e.g., 34 hours) is reached.

[0011] Typically, a monitoring system monitors one or more of the following parameters in real time: load, pressure, temperature, rotational speed, and rolling radius (e.g., obtained by the ratio of the rotational speed of the tire and the rolling wheel). In this way, any rapid and / or relevant changes in the monitored parameters (which indicate that the tire is about to fail or has already failed) can be detected, thus interrupting the test. Summary of the Invention

[0012] In the context of rolling tests on tires, the applicant found that, for example, visual inspection of tires at predetermined time intervals, and / or the aforementioned continuous monitoring of certain parameters (pressure, temperature, rolling radius, etc.), may not be a reliable method for detecting the exact moment when tire damage occurs and / or for identifying the exact type of damage that initially appears, under certain conditions.

[0013] In fact, real-time monitoring can only detect tire damage (such as bursting or tearing) or structural failures, and is completely unsuitable for detecting damage that does not affect, for example, tire air tightness and / or structural sealing (such as the formation of cracks or internal cracks).

[0014] In this sense, visual inspection is generally more sensitive than the continuous monitoring mentioned above.

[0015] However, firstly, visual inspection only detects visible damage at pre-planned inspection times, thus there is a potential delay compared to the time when the damage occurs, which can be as large as the time interval between two inspections.

[0016] In addition, it is possible that the tire is damaged from the inside out, but there are no visible defects, so it cannot be identified by visual inspection.

[0017] In this situation, the applicant discovers that two different scenarios may arise.

[0018] In the first scenario, as the rolling test proceeds, damage that initially appears inside the tire and is therefore difficult to detect develops into an externally visible and thus detectable defect. In this case, the rolling test, which is typically performed (i.e., a planned visual inspection at predetermined time intervals), may lead to an overestimation of the tire's durability, corresponding to the time required for internal damage to develop into a corresponding externally visible defect (this time may be added to the planned visual inspection, before the externally visible defect is detected).

[0019] Furthermore, in this situation, the damage visible during inspection may not completely correspond to the damage initially generated inside the tire, leading to inaccurate classification of the initial damage during testing.

[0020] In the second scenario, internal tire damage may not produce any externally visible defects until the longest possible time before testing, and therefore will not be detected by any visual inspection. In this case, although the tire has internal damage, it may be considered undamaged and thus pass the test ("false negative").

[0021] Furthermore, according to the applicant's findings, visual inspection is very labor-intensive and time-consuming (due to machine downtime), which affects the cost of testing.

[0022] In the context of tire rolling tests, the applicant faces the challenge of determining the onset time of potential damage.

[0023] Against this backdrop, the problem facing applicants is how to identify the existence of latent defects in order to limit the occurrence of false negative tires.

[0024] In the above context, the problem faced by the applicant is how to accurately identify the type of damage that first appears during testing.

[0025] According to the applicant's findings, one or more of the aforementioned problems were resolved by analyzing the spectrum of a signal representing the linear acceleration experienced by the tire.

[0026] According to one aspect, the present invention relates to a method for testing tires.

[0027] The method includes:

[0028] - Mount the tire onto the rim and inflate the tire to a certain pressure;

[0029] - Rotate the tire about an axis of rotation, wherein the tire is subjected to thrust on a rolling surface;

[0030] - Obtain an acceleration signal representing the linear acceleration experienced by the tire during the rotation;

[0031] - Calculate at least one spectrum of the acceleration signal;

[0032] - Calculate at least one index based on the at least one spectrum;

[0033] - Assess the integrity status of the tire based on the time trend of the at least one index.

[0034] According to another aspect, the present invention relates to an apparatus for testing tires.

[0035] The device includes:

[0036] - A rotor, rotatably fixed to a stator so as to be able to rotate about a rotation axis, wherein the rotor includes an attachment portion configured to mount a rim on which the tire is mounted;

[0037] - A rolling wheel having a corresponding axis of rotation parallel to the axis of rotation;

[0038] - An acceleration sensor configured to detect the linear acceleration experienced by the rotor during rotation and generate an acceleration signal representing the linear acceleration;

[0039] - A processing unit that communicates with the acceleration sensor, the processing unit being used to acquire the acceleration signal.

[0040] Preferably, the processing unit is programmed and configured as follows:

[0041] - Calculate at least one spectrum of the acceleration signal;

[0042] - Calculate at least one index based on the at least one spectrum;

[0043] - Assess the integrity status of the tire based on the time trend of the at least one index.

[0044] The applicant discovered that the spectrum of the linear acceleration signal experienced by a rolling tire depends on the tire's physical state. For example, when one or more defects occur, the acceleration stress experienced by the tire changes, and this change is particularly noticeable and / or significant in the frequency domain rather than the time domain. According to the applicant's findings, this is due to the periodicity of rolling.

[0045] Therefore, the applicant has discovered that a temporal trend of at least one exponential value calculated from at least one spectrum of an acceleration signal, for example, allows for the determination of the onset of damage (including visible and invisible) at the end of a test or in real time. In fact, the applicant has demonstrated that, at least for certain types of damage and / or under certain test conditions and / or for certain types of tires, the temporal trend of the exponential value exhibits characteristic behavior at the instant of damage occurrence, due to potential variations in the spectrum of the acceleration signal. This allows for a more accurate determination of the moment of damage occurrence, thereby providing more accurate information for the actual durability of the tire.

[0046] The applicant discovered that the time trend of at least one exponent calculated from at least one spectrum allows for the identification, for example, of damage not visible during visual inspection of the tire at the end of the test or in real time, such as internal damage that does not extend to the outer surface of the tire. In fact, the time trend of the exponent can be used to assess whether there are significant changes in the spectrum of the acceleration signal, which are signs of defect occurrence. In this way, the method is more reliable, for example, by limiting the testing to tires with internal damage (because they do not have visible defects).

[0047] Furthermore, the applicant discovered that assessing tire damage status based on the time trend of at least one index (which is calculated based on at least one spectrum) allows, in some cases, the classification of the type of damage occurring in the tire by the time trend itself. Indeed, the spectrum used to calculate the index may depend on, for example, the structural elements of the damaged tire; therefore, the type of damage may correspond to the characteristic time trend of the index.

[0048] The present invention may have one or more of the following preferred features in one or more of the above aspects.

[0049] Preferably, the processing unit is programmed and configured to perform one or more of the following operations provided in the method of the present invention.

[0050] Preferably, the acceleration signal includes one or more, more preferably, all of the following different acceleration sub-signals: a first acceleration sub-signal representing the radial component of the linear acceleration, a second acceleration sub-signal representing the axial component of the linear acceleration, and a third acceleration sub-signal representing the tangential component of the linear acceleration.

[0051] Preferably, the acceleration sensor is configured to detect one or more, more preferably all, of the axial, radial and tangential components of the linear acceleration.

[0052] Preferably, calculating the at least one spectrum of the acceleration signal includes calculating the corresponding spectrum of each of the acceleration sub-signals.

[0053] Preferably, calculating the at least one index based on the at least one spectrum includes calculating the corresponding index based on each corresponding spectrum.

[0054] In this way, stress information is obtained and explained about the tire's stress in at least one of three spatial directions, preferably all of them, that are reasonably chosen relative to rotation.

[0055] In the following text, any references to acceleration signals, spectra, and exponents refer to the general acceleration signal, its corresponding spectrum and exponent, one or more acceleration sub-signals, and their corresponding spectra and exponents.

[0056] Preferably, assessing the tire integrity status based on the time trend of the at least one index includes assessing the tire integrity status based on the time trends of one or more of the corresponding indices (e.g., individually or in combination). In this way, the tire integrity status can be assessed based on the time trend of the corresponding index for each spatial direction, thereby obtaining characteristic information about the tire integrity status.

[0057] Preferably, the tire is determined to be damaged when the time trend of the at least one index (preferably, one or more of the corresponding indices) verifies (preferably, the corresponding) damage state.

[0058] Preferably, the damage state is verified when the value of the at least one index (preferably, a corresponding index) is greater than or equal to the threshold of the (preferably, corresponding) index; more preferably, the verification is performed continuously within a fixed time interval (e.g., greater than or equal to thirty minutes, or greater than or equal to sixty minutes).

[0059] The applicant has in fact verified that the value of the index tends to increase over time during the occurrence and development of damage.

[0060] Preferably, the assessment of the tire's integrity status based on the time trend of the at least one index (preferably one or more of the corresponding indices) is performed in real time (i.e., during tire rotation).

[0061] In this way, the integrity of the tire can be monitored in real time without necessarily stopping the test. For example, potential damage can be detected at the same moment (or close to the moment) it occurs, and the test can be intervened.

[0062] Preferably, the rotation of the tire is interrupted when the time trend of at least one index (preferably one or more of the corresponding indices) verifies (preferably, the corresponding) damage condition. In this way, the test can be stopped in real time (regardless of any predetermined visual inspection), and a visual inspection of the tire can be performed for a thorough evaluation. The test can be stopped (thus saving time compared to predetermined inspections) or resumed.

[0063] In one implementation, rotation is periodically interrupted (preferably at intervals of 15 or 20 hours) and a predetermined visual inspection is performed on the tire to assess its integrity. Thus, the predetermined visual inspection (regardless of the index) is incorporated into and integrated with real-time monitoring of the integrity status via the time trend of the index, providing a dual inspection. For example, specific damage (e.g., aesthetic damage to the surface) can be detected, which in some cases might not be detectable by the time trend of the inspection index (or the corresponding index). According to the invention, continuous monitoring based on the index also allows for an extension of the time interval between two consecutive predetermined inspections relative to the normally used time interval (e.g., 12 hours).

[0064] In one implementation, tire rotation is not interrupted for periodic, predetermined visual inspections (except for those that may be performed at the end of the test), and the tire rotation is only interrupted when the time trend of the at least one index (preferably, one or more of the corresponding indices) verifies (preferably, the corresponding) damage condition. This saves downtime required to perform the predetermined visual inspections and / or significantly reduces the labor costs for performing the tests, since only one visual inspection is performed at the end of the rolling test to confirm whether damage detected (or undetected) by monitoring the time trend of the indices (and / or one or more corresponding indices) actually exists on the tire (or correspondingly, does not exist on the tire).

[0065] Preferably, an alarm signal is generated based on the determination of the tire damage. In this way, potential damage can be reported, for example, during real-time monitoring, in which case a visual inspection by the operator is necessary to assess the tire's condition and / or terminate the test.

[0066] Typically, the rotation is characterized by a rotational speed (i.e., the number of revolutions per unit of time), which defines the fundamental frequency.

[0067] Typically, the rotation is characterized by the rotation cycle (i.e., the time required for the tire to complete one rotation).

[0068] Preferably, the at least one spectrum, more preferably, each of the corresponding spectra includes a fundamental wave with a frequency equal to the fundamental frequency.

[0069] Preferably, a speed signal representing the rotational speed of the tire is acquired.

[0070] Preferably, the device includes a speed sensor configured to detect the rotational speed of the tire and generate a speed signal representing the rotational speed. Preferably, the speed sensor is applied partly to the stator and partly to the rotor. Preferably, the processing unit communicates with the speed sensor to acquire the speed signal.

[0071] This allows for spectral analysis to be performed taking into account the obtained fundamental frequency, and / or sampling synchronized with rotation, as described below.

[0072] Preferably, the rotational speed is constant (basically, for example, with a maximum variation of 0.5-1 km / h). This ensures that the fundamental frequency and its higher harmonics do not change during testing, and the exponent can be calculated based on a reference spectrum obtained across the entire tire.

[0073] Preferably, the velocity signal and / or the acceleration signal (preferably each of the acceleration sub-signals) are acquired over a time sequence of acquisition intervals, wherein preferably, each acquisition interval has a duration, which is preferably continuous and greater than or equal to 1 minute and / or less than or equal to 5 minutes. Preferably, the acquisition intervals have the same duration. Typically, each acquisition interval comprises multiple (e.g., hundreds) rotational cycles. Preferably, the time interval between the corresponding initial moments of two subsequent acquisition intervals is constant, more preferably greater than or equal to 5 minutes and / or less than or equal to 15 minutes.

[0074] Preferably, the time trend of the at least one index (preferably, each corresponding index) is determined by: (preferably, respectively) performing the calculation on at least one spectrum (preferably, calculating each corresponding spectrum) for each of the acquisition intervals, and / or performing the calculation on at least one index (preferably, each corresponding index). In this way, the time trend of the index can be obtained efficiently.

[0075] Preferably, the velocity signal and / or the acceleration signal (preferably each of the acceleration sub-signals) are acquired at a sampling frequency of greater than or equal to 4000 Hz and / or less than or equal to 7000 Hz (for acquiring signals of sufficient quality).

[0076] Preferably, calculating the at least one spectrum (preferably each corresponding spectrum) includes (preliminarily) calculating the variance (RMS) of the acceleration signal (preferably each of the acceleration sub-signals) and accepting the acceleration signal (preferably each of the acceleration sub-signals) as a function of the variance. In this way, it is possible to assess whether the values ​​of the obtained acceleration signals (and / or acceleration sub-signals) are acceptable and unaffected by factors unrelated to the damage phenomenon, and / or to evaluate the reliability and / or consistency of the obtained acceleration signals (and / or acceleration sub-signals).

[0077] Preferably, calculating the at least one spectrum (preferably each corresponding spectrum) includes filtering the acceleration signal (preferably each of the acceleration sub-signals) with a low-pass filter, more preferably a fourth-order Butterworth filter (preferably after the variance calculation). In this way, obfuscation, i.e., undersampling of the considered acceleration signal, can be eliminated.

[0078] Preferably, calculating the at least one spectrum (preferably each corresponding spectrum) includes (preferably, after the filtering) interpolating the acceleration signal (preferably each acceleration sub-signal) using splines for each acquisition interval, and sampling the interpolated acceleration signal (preferably each acceleration sub-signal) at a given number of points (preferably greater than or equal to one hundred, and / or less than or equal to five hundred) for each cycle of the rotation. Preferably, the sampling is performed synchronously with the rotation (i.e., for each rotation cycle, the sampling points fall at the same angular position of the tire).

[0079] In this way, the first and last points of the given number of points can always be within the rotation period (e.g., obtained from the velocity signal) and fall precisely at the start and end moments.

[0080] Preferably, the calculation for the at least one spectrum (preferably, each corresponding spectrum) includes, more preferably, calculating a Fourier transform, and more preferably a Fast Fourier Transform (FFT), for the acceleration signal (preferably, for each acceleration sub-signal) after the sampling. In this way, the spectrum of the acceleration signal can be obtained quickly and efficiently.

[0081] Preferably, the Fourier transform is calculated on multiple (e.g., dozens, preferably, a predetermined number) portions of the acceleration signal (preferably, each of the acceleration sub-signals), each portion corresponding to an integer number (preferably greater than or equal to 20 and / or less than or equal to 100) of consecutive rotation cycles.

[0082] Preferably, the calculation for the at least one spectrum (preferably each corresponding spectrum) includes averaging the Fourier transform in complex form, the Fourier transform being calculated over multiple portions of the acceleration signal (preferably each of the acceleration sub-signals).

[0083] In this way, noise in the acceleration signal (and / or sub-signals of the acceleration signal) obtained synchronously can be eliminated. Furthermore, the weight of the average allowable limit harmonic phase difference (related to the imaginary part of the complex number), in complex form, provides an indication of the defect location, and considering only the harmonic amplitude (related to the real part of the complex number) when calculating the exponent, it provides an indication of the presence (and / or weight) of new defects.

[0084] Preferably, the calculation for the at least one spectrum (more preferably, each corresponding spectrum) includes determining a set of M harmonics, wherein the M harmonics are integer multiples of the fundamental frequency, and more preferably, the M harmonics are consecutive, starting from and including the fundamental frequency. Preferably, M is an integer greater than or equal to 20 and / or less than or equal to 50. This way, only harmonics directly related to the overall dynamics of the tire are considered, eliminating harmonics related to phenomena unrelated to damage.

[0085] Preferably, the M amplitudes of the M harmonics are determined; more preferably, this determination is performed in an average spectrum obtained by averaging the Fourier transform in complex form, wherein the Fourier transform is calculated over the plurality of portions of the acceleration signal. For example, the M amplitudes are determined by integrating the amplitudes of the harmonics, which fall in the neighborhood of the nominal frequency of the corresponding harmonic, which is an integer multiple of the fundamental frequency.

[0086] Preferably, calculating the at least one index (preferably, calculating each corresponding index) includes calculating a reference spectrum (preferably, the corresponding reference spectrum) of the acceleration signal (preferably, each of the acceleration sub-signals) under reference conditions. Preferably, the reference conditions correspond to the complete state of the tire. Preferably, the reference conditions are predetermined in time. Preferably, the reference conditions correspond to a reference interval, which is preferably after the fifth hour and / or before the fifteenth hour after the tire begins to rotate on the rolling surface under thrust.

[0087] In this way, the spectrum of the signal (and / or each sub-signal) relative to the state of the undamaged tire is obtained. Specifically, during the period between the fifth and fifteenth hours after the start of the test, the tire is considered to have reached a stable state while maintaining its initial integrity, or more generally, maintaining its initial state.

[0088] Preferably, calculating the (preferably, corresponding) reference spectrum involves averaging a series of the (preferably, corresponding) spectra calculated for each of a series of (preferably, more than or equal to two, and / or less than or equal to ten) consecutive acquisition intervals under the reference conditions (e.g., within the reference interval). In this way, a (preferably, corresponding) reference spectrum of the acceleration (sub)signal can be obtained, which is unaffected by noise or other irrelevant factors and is stable.

[0089] Preferably, the calculation of the at least one index (preferably, the calculation of each corresponding index) includes, more preferably, comparing the at least one spectrum (preferably, each corresponding spectrum) with the reference spectrum (preferably, the corresponding reference spectrum) for each of the acquisition intervals.

[0090] Preferably, the comparison includes, more preferably, for each acquisition interval:

[0091] - Arrange the M harmonics in the set of the M harmonics in descending order of amplitude;

[0092] - Select a subset from the sorted set of the M harmonics, the subset comprising P consecutive harmonics starting from the first, wherein P is preferably greater than or equal to six (more preferably, eight) and / or less than or equal to thirty (more preferably, twenty);

[0093] - Compare the amplitudes of the P harmonics of the subset with the corresponding amplitudes of the corresponding harmonics in the (preferably, corresponding) reference spectrum.

[0094] Thus, only the first P harmonics with large amplitudes among the first M harmonics that are integer multiples of the fundamental frequency are considered, which, according to the applicant, is of utmost importance for calculating the exponent.

[0095] Preferably, the comparison includes, for each acquisition interval:

[0096] - Selecting possible further P' harmonics of the at least one spectrum in each acquisition interval (preferably in each of the corresponding spectra), these harmonics correspond to harmonics that belong to a corresponding subset of the P harmonics in the spectrum of the acquisition interval preceding each acquisition interval (preferably in the corresponding spectrum) and do not belong to the subset of the P harmonics in the at least one spectrum (preferably in each of the corresponding spectra), wherein P' is preferably greater than or equal to 0 and / or less than or equal to MP;

[0097] - Compare the P' amplitudes of the P' possible further harmonics with the corresponding amplitudes of the corresponding harmonics in the (preferably, corresponding) reference spectrum.

[0098] Preferably, the comparison is performed using the following formula, and more preferably, the following formula is applied to each acquisition interval:

[0099]

[0100] Wherein, IDP is the at least one index, or preferably, the corresponding index.

[0101] N is a number that is greater than or equal to the harmonic number P of the subset and less than or equal to the harmonic number M of the subset (preferably, N is equal to the sum of the harmonic number P belonging to the subset and the possible further harmonic numbers P').

[0102] A i It is the amplitude of the i-th harmonic among the P harmonics of the subset of harmonics (preferably, also among the P' possible further harmonics);

[0103] It is the amplitude of the i-th harmonic corresponding to the (preferably, the corresponding) reference spectrum of the i-th harmonic.

[0104] In this way, the amplitude of the i-th harmonic of the acceleration signal's spectrum can be compared with the corresponding amplitude of the i-th harmonic of the reference spectrum of each of the said acceleration sub-signals. Therefore, the degree of variation of the acceleration signal's spectrum relative to the reference spectrum of the acceleration signal representing the tire's undamaged state in each spatial direction in which the acceleration signal was acquired can be evaluated.

[0105] Preferably, the time trend of the at least one index (preferably, the time trend of each corresponding index) is obtained by the following method:

[0106] - The acquisition intervals are grouped into groups of consecutive acquisition intervals in a time series (preferably, each group includes a certain number, more preferably, two or more and / or ten or less acquisition intervals).

[0107] - Average the indices (preferably, the respective indices that are separate from each other) for each group, the indices being calculated for each collection interval of the group.

[0108] In this way, noise in the exponent value can be reduced.

[0109] Preferably:

[0110] - Provides a rotor rotatably fixed to the stator so as to be able to rotate about the axis of rotation;

[0111] - The rim is mounted on the attachment portion of the rotor;

[0112] The acceleration signal is obtained on the stator.

[0113] Preferably, the acceleration sensor is mounted on the stator; more preferably, it is mounted at the end of the stator near the attachment portion of the rotor.

[0114] In this way, the detection of acceleration and / or the frequency interpretation of acceleration signals become simple and / or reliable.

[0115] Preferably, a rolling wheel is provided, the corresponding axis of rotation of which is parallel to the axis of rotation, wherein the rolling surface belongs to the rolling wheel.

[0116] Typically, the axis of rotation is static (and so is the corresponding axis of rotation of the roller).

[0117] Preferably, the device includes a thrust device for applying a thrust to the rotor in the direction of the rolling wheel.

[0118] Preferably, the force is constant. Preferably, the force is selected as a function of the load index of the tire, for example, the force is greater than or equal to 400 kg.

[0119] Optionally, the device includes a pressure sensor for detecting the pressure of the tire and generating a pressure signal representing the pressure. Preferably, the pressure sensor is mounted on the wheel rim. Preferably, the processing unit communicates with the pressure sensor to acquire the pressure signal.

[0120] Preferably, the pressure can be kept constant by adjusting the system. Attached Figure Description

[0121] Figure 1 A partial schematic diagram of a tire testing apparatus according to an embodiment of the present invention is shown, some of which are transparent.

[0122] Figure 2 A flowchart of a method for testing tires according to an embodiment of the present invention is shown.

[0123] Figure 3 , Figure 4 and Figure 5 Examples of time trends for indices of three different types of tire damage, calculated according to embodiments of the method of the present invention, are shown. Detailed Implementation

[0124] The features and advantages of the present invention will be further illustrated by the following detailed description of some embodiments, which are presented by way of non-limiting examples of the invention and with reference to the accompanying figures.

[0125] Figure 1 An example of a device 200 for testing a tire 70 according to the present invention is shown. Figure 1 A top view can be shown as an example.

[0126] Tire 70 is mounted on a rim (not shown) and inflated to a typically predetermined pressure to achieve the characteristics of a car wheel.

[0127] exist Figure 1 In the image, some transparent parts visible through the wheels are represented by dashed lines.

[0128] The device 200 includes a rotor 71 rotatably fixed to a stator 72 so as to be able to rotate about a rotation axis 101. The rotor 71 includes an attachment portion 73 configured for mounting a wheel rim.

[0129] The device 200 includes a roller 75, the corresponding axis of rotation 102 of which is parallel to the axis of rotation 101 of the rotor 71. Normally, the axes of rotation 101 and 102 are static.

[0130] The device 200 includes an accelerometer 76. For example, the accelerometer 76 is triaxial and is configured to detect accelerations at different speeds. Figure 1 The tangential, axial, and radial components of the linear acceleration oriented along the X, Y, and Z axes, respectively.

[0131] For example, the acceleration sensor 76 is mounted on the stator 72 (e.g., on the outer surface of the stator 72), for example, at the end of the stator 72 near the attachment portion 73 of the rotor 71.

[0132] In an alternative embodiment (not shown), the acceleration sensor may be mounted directly on the rotor 71, on the rim, or directly on the tire 70.

[0133] The device 200 includes a processing unit 77, which communicates with the accelerometer 76, for example via a communication line A (wireless or non-wireless), to receive acceleration signals generated by the accelerometer 76.

[0134] For example, processing unit 77 is programmed and configured to perform the operations described below.

[0135] For example, device 200 includes a thrust device 80 for applying a thrust F directed toward the roller 75 to the rotor 71. For example, the thrust device may include one or more cylinders that act on the stator 72 during use, which in turn transmits the thrust to the rotor 71 and thus to the tire 70, so that the tire maintains the thrust on the roller 75.

[0136] Exemplarily, the device 200 includes a pressure sensor 79 configured to detect the pressure of the tire 70 and generate a pressure signal representing that pressure. Exemplarily, the pressure sensor 79 communicates with the processing unit 77, for example, via a communication line P (wireless or non-wireless), and it is applied to the inflation valve of the tire 70, which is mounted on the rim.

[0137] Exemplarily, the device 200 includes a speed sensor 78 configured to detect the rotational speed of the tire 70 and generate a speed signal representing that rotational speed. Exemplarily, the speed sensor is applied partly to the rotor 71 and partly to the stator 72. Exemplarily, the speed sensor 78 communicates with the processing unit, for example, via a communication line V (wireless or non-wireless).

[0138] Figure 2 An operational flowchart illustrating an example of a method 99 for testing a tire 70 according to the present invention is shown, which can be implemented using the apparatus 200 described above.

[0139] Preferably, the method 99 includes mounting the tire 70 1 onto the rim and inflating the tire 70 30 to a certain pressure. For example, the pressure of the tire 70 is equal to the working pressure of a particular tire, such as about 200 kPa, and remains constant during the test.

[0140] Preferably, the method 99 includes rotating the tire 70 about the axis of rotation 101, wherein the tire 70 is under the thrust of a force F on the rolling surface 74 of the roller 75. For example, the force F applied to the tire remains constant, for example, equal to about 1000 kg.

[0141] For example, the rotation is characterized by a rotational speed of 10 revolutions per second, which defines a fundamental frequency of 10 Hz. For example, the rotational speed remains constant, with a maximum variation of 0.5-1 km / h.

[0142] In one possible implementation, the rotational speed is varied during the test, for example, from 8 revolutions / second to 14 revolutions / second, incremented (e.g., 1 revolution / second) at predetermined intervals (e.g., every 200 minutes). In this implementation, each interval corresponds to a different rotational speed that remains constant during the interval and corresponds to a different fundamental frequency (e.g., varying from 8 Hz to 14 Hz). In this case, the method described below can be implemented exemplary for each interval (e.g., with a corresponding reference spectrum for each interval).

[0143] For example, the rotation is characterized by a rotation period of approximately 0.1 seconds.

[0144] Preferably, the method 99 includes acquiring an acceleration signal 3 representing the linear acceleration experienced by the tire 70 during rotation. For example, the acceleration signal includes all of the following different acceleration sub-signals: a first acceleration sub-signal representing the radial component of the linear acceleration, a second acceleration sub-signal representing the axial component of the linear acceleration, and a third acceleration sub-signal representing the tangential component of the linear acceleration.

[0145] In other implementations, for example, if you want to analyze specific damage to tire 70 (which is more easily detected on one component of acceleration), only one and / or two components of the linear acceleration experienced by tire 70 can be detected.

[0146] For example, the following operations are performed on each obtained sub-signal. Therefore, in the following text, each reference to the index (IDP) refers to each corresponding index (IDP). X IDP Y IDP Z Furthermore, each reference to a spectrum (such as a spectrum or reference spectrum) refers to the corresponding spectrum.

[0147] For example, the method 99 includes acquiring a speed signal 31 representing the rotational speed of the tire 70 (e.g., one pulse signal per revolution).

[0148] For example, the velocity signal and each acceleration sub-signal are acquired over a time series of acquisition intervals, wherein, exemplaryly, each acquisition interval has the same continuous duration of three minutes. Exemplarily, each acquisition interval comprises approximately 1800 rotation cycles. For example, the time interval between the initial moments of two subsequent acquisition intervals is constant and equal to 10 minutes.

[0149] In other words, for example, data recordings are acquired every ten minutes for a duration of three minutes, each data recording targeting the velocity signal and each acceleration sub-signal. In this description and claims, operations performed with respect to the acquisition interval (such as calculating the spectrum or calculating the index) are intended to operate on the corresponding data records acquired within that acquisition interval.

[0150] For example, the velocity signal and each acceleration sub-signal are acquired at a sampling frequency of approximately 5000 Hz.

[0151] For example, the following operation is performed on the signals acquired along each acquisition interval (in other words, the operation is performed separately on each record of each sub-signal acquired within three minutes). In other words, as... Figure 2 As shown by the arrow from the output of Operation 5 to the input of Operation 4, once the calculation of the spectrum and relative exponent for the current acquisition interval is completed, in addition to updating the time trend of the exponent (Operation 6), the same operation is performed on the signal recording of the subsequent acquisition intervals.

[0152] Preferably, the method 99 includes calculating a spectrum 4 for each acceleration sub-signal. For example, each spectrum includes a fundamental wave with a frequency equal to the fundamental frequency (preferably obtained from the velocity signal).

[0153] For example, the method 99 first includes calculating the variance 10 of each accelerometer sub-signal, and accepting each accelerometer sub-signal if the variance is less than a given percentage value.

[0154] For example, the method 99 then includes filtering each accelerometer signal 11 with a fourth-order Butterworth filter.

[0155] For example, each acceleration sub-signal is interpolated using splines 12, and each interpolated acceleration sub-signal is sampled 13 in a manner synchronized with the rotation. For example, the sampling 13 performed in a synchronized manner is performed at 300 points per rotation cycle, such that the first and last sampling points always fall at the beginning and end of the rotation cycle (e.g., obtained from the velocity signal).

[0156] For example, calculating each spectrum 4 includes subsequently sampling in a synchronous manner 13, and calculating a Fast Fourier Transform (FFT) 14 for each resampled record of the accelerometer sub-signal. For example, the record of each accelerometer sub-signal is divided into a predetermined number (e.g., 30) parts, and a Fast Fourier Transform 14 is calculated for each part. For example, each part corresponds to 50 consecutive rotation cycles.

[0157] For example, calculating each spectrum 4 involves averaging a Fast Fourier Transform calculated for each sub-signal (and each acquisition interval) over a predetermined number of portions of each acceleration sub-signal, in complex form 15. This averaging in complex form results in an averaged spectrum characterized by amplitude values ​​distributed according to frequency.

[0158] For example, in each average spectrum 16, a set of M amplitudes (typically stored in the processing unit) associated with M(or more) integer multiples of the fundamental frequency, starting from and including the fundamental frequency, is determined. In other words, exemplary, only the amplitudes of the first M(or more) integer multiples of the fundamental frequency are determined and used to calculate the exponent. For example, M equals 30. For example, each amplitude is calculated by integrating the amplitudes of harmonics falling within approximately 10% of the fundamental frequency value, centered on the values ​​of the corresponding nominal frequencies of the corresponding integer multiples of the fundamental frequency.

[0159] In other words, for each acquisition interval and each sub-signal, the calculated spectrum can, for example, include M-tuples of real values, representing the amplitudes of the first M (e.g., 30) integer multiples of the fundamental harmonics.

[0160] For example, method 99 includes calculating a reference spectrum 17 for each acceleration sub-signal under reference conditions. For example, these reference conditions provide for the realization of structural and / or dimensional stability of the tire during the test, while maintaining its integrity. For example, these reference conditions can be predetermined in time; for instance, they can correspond to a reference time interval from the start to the end of the tenth hour from when the tire 70 begins to rotate.

[0161] For example, the reference spectrum is obtained by averaging the spectrum calculated over six consecutive acquisition intervals (e.g., by averaging the M-tuples of the amplitudes of the corresponding first M harmonics), for example, the acquisition intervals being between the ninth and tenth hours from the start of the rotation.

[0162] In other words, the reference spectrum may, for example, include M-tuples of real values, representing the average amplitude of the first M harmonics of the spectrum calculated for each acquisition interval, for example, from the beginning to the end of the tenth hour of the rolling test.

[0163] Preferably, method 99 includes, for each acquisition interval, calculating an index 5 based on the spectrum, and comparing the spectrum with a reference spectrum 18.

[0164] For example, the comparison 18 includes, for each acquisition interval:

[0165] - Select a subset of P harmonics (e.g., P equals 12) from the set of M harmonics, where the P harmonics have the largest amplitude among the M harmonics;

[0166] - Preferably, a number of possible further harmonics P' are selected in the spectrum of the current acquisition interval, which correspond to harmonics that belong to a corresponding subset of P harmonics in the corresponding spectrum of the acquisition interval before the current acquisition interval, and do not belong to a subset of P harmonics in the current spectrum (P' is an integer greater than or equal to zero and less than or equal to MP).

[0167] - Compare the P amplitudes of the P harmonics and the P' amplitudes of the possible further P' harmonics with the amplitudes of the corresponding harmonics in the reference spectrum.

[0168] In other words, for the acquisition interval immediately following the calculation of the reference spectrum, only the P harmonics with larger amplitudes are used. For each subsequent acquisition interval, in addition to the P harmonics with the largest amplitude in the spectrum of the current acquisition interval, P' possible further harmonics that exist between the P harmonics with larger amplitudes in the previous acquisition intervals but do not exist between the P harmonics in the current acquisition interval are also used.

[0169] For example, the formula for calculating the index (IDP) for each collection interval is:

[0170]

[0171] Where N is equal to the sum of the harmonic number P that belongs to the subset for the current spectrum and the possible further harmonic numbers P' (therefore N is greater than or equal to P and less than or equal to M).

[0172] A i It is the amplitude of the i-th harmonic among the P harmonics in the current spectrum, and also among the possible further P' harmonics.

[0173] It is the amplitude of the i-th harmonic of the reference spectrum corresponding to the i-th harmonic of the current spectrum.

[0174] Preferably, the index is calculated for each collection interval to obtain the time trend of the index.

[0175] For example, the time trend of the index is obtained by grouping the acquisition intervals into time series of consecutive acquisition interval groups of 19, with an exemplary number of acquisition interval groups equal to 6, and for each group, a 20-fold average is performed on the corresponding index calculated for each acquisition interval of that group.

[0176] In other words, for each current collection interval, an average is performed from time to time between the current index and the index calculated for the previous five exemplary collection intervals (i.e., the index is averaged over a moving window of, for example, one hour).

[0177] For example, method 99 includes basing decisions on three corresponding indices (IDP). X IDP Y IDP Z The time trend of one or more indices in ) is used to assess the integrity status of 6 tires 70 in real time.

[0178] In one implementation, the integrity status of tire 70 is assessed offline based on a report automatically generated by the processing unit at the end of the tire rolling test, which relates to the time trend of each index.

[0179] For example, the integrity status of 6 tires 70 is assessed by considering the time trend of each corresponding index.

[0180] In one implementation, the integrity status of a tire is assessed based on a time trend of a combination of corresponding indices, for example, based on the linear sum of three corresponding indices.

[0181] For example, when the time trend of one or more of the three corresponding indices verifies the corresponding damage state, such as verifying that the value of the corresponding index is greater than or equal to the corresponding threshold, it is determined that tire 70 is damaged, where each corresponding threshold may be different from the others, for example.

[0182] In a further implementation, a damaged state is verified when the values ​​of one or more indices change rapidly over time, regardless of whether the corresponding threshold is reached.

[0183] Figure 3 , 4 Figures 5 and 6 graphically illustrate the results of three exemplary rolling tests conducted by the applicant according to the above method.

[0184] In particular, the figure shows three indices (IDP). X IDP Y IDP Z Examples of the time trends of these indices are calculated from acceleration sub-signals obtained from the three components of linear acceleration, indicating three different types of damage produced in a tire. In particular, continuous line 300 shows the IDP associated with the tangential component of linear acceleration. X The time trend of the index, dotted line 301 shows the IDP related to the axial component of linear acceleration. Y The time trend of the exponent, with dashed line 302 showing the IDP as a function of the radial component of linear acceleration. Z The time trend of the index.

[0185] Each data point that makes up lines 300, 301, and 302 corresponds to the value of the corresponding index in the corresponding three-minute acquisition interval explained above.

[0186] exist Figure 3 , Figure 4 and Figure 5 In the chart shown, the values ​​on the horizontal axis represent the instant (in hours) from the moment after the reference condition (e.g., the moment after the 10th hour) to the end of the test (i.e., when one of the three indicators reaches the corresponding threshold).

[0187] The values ​​on the vertical axis represent index values ​​expressed in arbitrary units, where the unit value represents the threshold of the index that is reached during the test. As mentioned above, these indices are averaged over a one-hour moving window.

[0188] Figure 3 The time trends of three indices are shown in the case of tire band damage. As can be seen from the figure, the time trend of index 300, which is related to the tangential component of acceleration, has reached the threshold, thus verifying the damage state approximately 57 hours after the start of rotation.

[0189] Figure 4 The figure shows the time trends of three indices in the case of tire sidewall damage. As can be seen from the figure, the time trend of index 301, which is related to the axial component of acceleration, has reached the threshold, thus verifying the damage state approximately 48 hours after the start of rotation.

[0190] Figure 5 The figure shows the time trends of three indices in the case of tire band rupture. As can be seen from the figure, the time trend of index 302, which is related to the radial component of acceleration, has reached the threshold, thus verifying the damage state at approximately 129 hours from the start of rotation.

[0191] For example, the time trend of the index is monitored in real time, and an alarm signal 21 is generated when at least one damage condition is verified, so as to interrupt the test and / or temporarily stop the rotation for visual inspection.

Claims

1. A method (99) for testing a tire (70), the method (99) comprising: - The tire (70) is mounted (1) on the rim and the tire (70) is inflated (30) to a certain pressure; - Rotate the tire (70) about the axis of rotation (101) (2), wherein the tire (70) is subjected to thrust on the rolling surface (74); - Obtain (3) an acceleration signal representing the linear acceleration experienced by the tire (70) during the rotation; - Calculate at least one spectrum of the acceleration signal (4); - Calculate at least one index (5) based on the at least one spectrum; -Based on the time trend of the at least one index, assess (6) the integrity status of the tire (70). Calculating the at least one spectrum (4) includes interpolating the acceleration signal (12) and sampling the interpolated acceleration signal at a given number of points for each cycle of the rotation (13), wherein the sampling is performed in a manner synchronized with the rotation.

2. The method (99) according to claim 1, wherein, The rotation is characterized by a rotational speed defined by a fundamental frequency, wherein the rotation is characterized by a rotational period, wherein the at least one spectrum includes a fundamental wave having a frequency equal to the fundamental frequency, wherein the acceleration signal is acquired over a time series of acquisition intervals, wherein each acquisition interval includes multiple rotational periods, wherein each acquisition interval has a duration greater than or equal to 1 minute and less than or equal to 5 minutes, wherein the acquisition intervals have the same duration, wherein the time interval between corresponding initial moments of two successive acquisition intervals is constant, wherein a speed signal (31) representing the rotational speed of the tire is acquired, wherein the speed signal is acquired over a time series of the acquisition intervals, and wherein the speed signal or the acceleration signal is acquired at an acquisition frequency greater than or equal to 4000 Hz and less than or equal to 7000 Hz.

3. The method (99) according to claim 2, wherein, The duration is continuous, and wherein, The time interval between the corresponding initial moments of two successive acquisition intervals is greater than or equal to 5 minutes and less than or equal to 15 minutes.

4. The method (99) according to claim 2, wherein, The time trend of the at least one index is determined by performing the calculation (4) on the at least one spectrum and the calculation (5) on the at least one index for each of the acquisition intervals.

5. The method (99) according to any one of claims 1 to 4, wherein, Calculating the at least one spectrum (4) includes calculating the variance (10) of the acceleration signal and accepting the acceleration signal as a function of the variance, wherein calculating the at least one spectrum (4) includes filtering the acceleration signal with a low-pass filter (11).

6. The method (99) according to any one of claims 2 to 4, wherein, The calculation (4) of the at least one spectrum includes calculating the Fourier transform (14) of the acceleration signal.

7. The method (99) according to claim 6, wherein, The Fourier transform is calculated over multiple portions of the acceleration signal, each portion corresponding to an integer number of consecutive rotational periods, wherein the calculation (4) of the at least one spectrum comprises averaging the Fourier transforms calculated over the multiple portions of the acceleration signal in the form of a complex number (15), wherein calculating the at least one spectrum (4) comprises determining (16) a set of M harmonics, the M harmonics being an integer multiple of the fundamental frequency, the M harmonics being consecutive, starting from and including the fundamental frequency, wherein M is an integer greater than or equal to 20 and less than or equal to 50.

8. The method (99) according to claim 7, wherein, The M amplitudes of the M harmonics are determined by integrating the amplitudes of the harmonics in the neighborhood of the corresponding nominal frequency of the corresponding harmonic that falls into an integer multiple of the fundamental frequency.

9. The method (99) according to claim 8, wherein, The M amplitudes are determined in the average spectrum obtained by averaging the Fourier transform calculated on the plurality of portions of the acceleration signal in the form of complex number (15).

10. The method (99) according to claim 8, wherein, Calculating the at least one index (5) includes calculating a reference spectrum (17) of the acceleration signal under reference conditions, wherein the reference conditions correspond to the complete state of the tire (70), wherein the reference conditions are predetermined in time, and wherein the reference conditions correspond to a reference interval between the fifth hour after the fifth hour and the fifteenth hour after the tire (70) begins to rotate under thrust on the rolling surface (74).

11. The method (99) according to claim 10, wherein, Calculating the reference spectrum (17) involves averaging a series of spectra calculated for each of a series of consecutive acquisition intervals under the reference conditions.

12. The method (99) according to claim 10, wherein, The calculation (5) of the at least one index includes, for each acquisition interval, comparing (18) the at least one spectrum and the reference spectrum, wherein the comparison (18) includes, for each acquisition interval, sorting the M harmonics of the set of M harmonics in descending order of amplitude, selecting a subset of the sorted set of M harmonics, the subset comprising P consecutive harmonics starting from the first, wherein P is greater than or equal to 6 and less than or equal to 30, and selecting P' possible further harmonics in the at least one spectrum of each acquisition interval, these possible further harmonics corresponding to the respective subsets of the P harmonics in the spectrum of the acquisition interval preceding each acquisition interval, and not belonging to the subset of the P harmonics in the at least one spectrum, wherein P' is greater than or equal to 0 and less than or equal to MP, and The P amplitudes of the P harmonics of the subset and the P' amplitudes of the P' possible further harmonics are compared with the corresponding amplitudes of the corresponding harmonics in the reference spectrum.

13. The method (99) according to claim 12, wherein, The comparison (18) is performed by applying the following formula to each acquisition interval: Wherein, IDP is at least one index of the collection interval; N is a number that is greater than or equal to the number P of the harmonics in the subset and less than or equal to the number M of the harmonics in the subset; A i It is the amplitude of the i-th harmonic among the P harmonics of the subset of harmonics, and also among the P' possible further harmonics; A irif It is the amplitude of the i-th harmonic corresponding to the i-th harmonic in the reference spectrum.

14. The method (99) according to any one of claims 2 to 4, wherein, The time trend of the at least one index is obtained by the following method: The acquisition intervals are grouped (19) into groups of time series of consecutive acquisition intervals, wherein each group includes a given number of acquisition intervals. And perform (20) averaging of the index for each group, wherein the index is calculated for each collection interval of the group.

15. The method (99) according to any one of claims 1 to 4, wherein, The condition for determining that the tire (70) is damaged is that the time trend of the at least one index verifies the damage state, wherein the damage state is verified when the value of the at least one index becomes greater than or equal to a threshold of the index.

16. The method (99) according to claim 15, wherein, The assessment (6) of the integrity status of the tire (70) based on the time trend of the at least one index is performed in real time, wherein the rotation of the tire (70) is interrupted when the time trend of the at least one index verifies the damage status, and wherein an alarm signal is generated based on the determination that the tire is damaged.

17. The method (99) according to any one of claims 1 to 4, wherein, A rotor (71) is provided rotatably fixed to a stator (72) to be able to rotate about the rotation axis (101), and the rim is mounted on an attachment portion (73) of the rotor (71), wherein the acceleration signal is obtained on the stator (72), wherein a rolling wheel (75) is provided having a corresponding rotation axis (102) parallel to the rotation axis (101), wherein the rolling surface (74) belongs to the rolling wheel (75).

18. The method (99) according to any one of claims 1 to 4, wherein, The acceleration signal includes one or more of the following different acceleration sub-signals: a first acceleration sub-signal representing the radial component of the linear acceleration, a second acceleration sub-signal representing the axial component of the linear acceleration, and a third acceleration sub-signal representing the tangential component of the linear acceleration, and wherein calculating the at least one spectrum (4) of the acceleration signal includes calculating a corresponding spectrum of each of the acceleration sub-signals, wherein calculating the at least one index (5) based on the at least one spectrum includes calculating a corresponding index based on each corresponding spectrum, and wherein evaluating the integrity status of the tire (70) based on the time trend of the at least one index (6) includes evaluating the integrity status of the tire (70) based on the time trend of one or more of the corresponding indices.

19. An apparatus (200) for testing a tire (70), the apparatus comprising: - A rotor (71) rotatably fixed to a stator (72) so as to be able to rotate about a rotation axis (101), wherein the rotor (71) includes an attachment portion (73) configured for mounting a rim on which the tire (70) is mounted; - A rolling wheel (75) having a corresponding rotation axis (102) parallel to the rotation axis (101); - An acceleration sensor (76) configured to detect the linear acceleration experienced by the rotor (71) during rotation and generate an acceleration signal representing the linear acceleration; - A processing unit (77) communicating with the acceleration sensor (76), the processing unit being configured to acquire the acceleration signal, wherein the processing unit (77) is programmed and configured to: - Calculate at least one spectrum of the acceleration signal (4); - Calculate at least one index (5) based on the at least one spectrum; -Assess the integrity status of the tire (70) based on the time trend of the at least one index (6). Calculating the at least one spectrum (4) includes interpolating the acceleration signal (12) and sampling the interpolated acceleration signal at a given number of points for each cycle of the rotation (13), wherein the sampling is performed in a manner synchronized with the rotation.

20. The apparatus (200) according to claim 19, wherein, The processing unit (77) is programmed and configured to perform one or more operations provided in the method (99) according to any one of claims 2 to 18.

21. The apparatus (200) according to claim 19 or 20, wherein, The acceleration sensor (76) is configured to detect one or more of the axial, radial, and tangential components of the linear acceleration, respectively, wherein the acceleration sensor (76) is mounted on the stator (72), wherein the device (200) includes a speed sensor (78) configured to detect the rotational speed of the tire (70) and generate a speed signal representing the rotational speed, wherein the speed sensor (78) is partially applied to the stator (72) and partially applied to the rotor (71), and wherein the processing unit (77) communicates with the speed sensor (78) to obtain the speed signal.

22. The apparatus according to claim 19 or 20, wherein, The device (200) includes a thrust device (80) for applying a thrust (F) directed toward the rolling wheel (75) on the rotor (71), wherein the thrust (F) is constant and selected according to the load index of the tire, wherein the device (200) includes a pressure sensor (79) for detecting the pressure of the tire (70) and generating a pressure signal representing the pressure, wherein the processing unit (77) communicates with the pressure sensor (79) to acquire the pressure signal and maintain the pressure constant.

Images