JP5828271B2 - Bearing preload judgment method - Google Patents

Description

  The present invention relates to a bearing preload determination method for determining the preload of an assembled bearing using FFT (Fast Fourier Transform).

Conventionally, this type of technology has been described in Patent Document 1.
This is because vibration is applied to the bearing from the outside with a vibration exciter, and the vibration is measured by an acceleration sensor on the inner and outer ring sides of the bearing and then analyzed by FFT and calculated from the resonance frequency (natural frequency). This is to determine the preload of the bearing.

Japanese Patent Laid-Open No. 2000-074788

  However, when a large preload is applied to the assembly of the bearing, as in a bearing such as a differential (differential) bearing of an automobile, the spring stiffness of the bearing increases and the amplitude of the natural vibration becomes extremely small. Accordingly, the calculation of the natural frequency by measuring the vibration of the bearing becomes difficult due to the influence of noise components in an environment where a large preload is required for the assembly of the bearing. For this reason, depending on the prior art in which the bearing preload is determined from the natural frequency, determination of the bearing preload such as the differential bearing (determination of whether the preload is within a predetermined range such as a preload standard) It is difficult.

  An object of the present invention is to provide a bearing preload determination method that can easily determine a large bearing preload as in the assembly of a bearing such as a differential bearing.

In order to achieve the above object, the invention according to claim 1 is a bearing preload determination method for determining bearing preload using fast Fourier transform, wherein the bearings are first and second bearings that support a rotating shaft. And the rotational vibration information from the first and second acceleration sensors arranged in the vicinity of the first bearing and in the vicinity of the second bearing capable of transmitting the vibrations of the first and second bearings . after a fast Fourier transform, the size of the absolute value of the magnitude and the frequency response function including the real and imaginary parts of the absolute value of the real part of the converted into each frequency response function, the frequency response function obtained The natural angular frequency of the bearing roller of the bearing is calculated from the calculated total ratio , and the bearing preload is estimated from the calculated natural angular frequency of the bearing roller. Whether the preload is within a predetermined range. Characterized in that it constant.
The invention according to claim 2 is characterized in that the bearing is a differential bearing.

  According to the present invention, after performing a fast Fourier transform on each rotational vibration information from the acceleration sensor and converting it to a frequency response function, the magnitude of the absolute value of each frequency response function, the real part, and the imaginary part And the sum ratio with the magnitude of the absolute value of the frequency response function. Since the bearing preload determination is made based on the natural angular frequency of the bearing roller based on this summing ratio, the preload determination can be performed while suppressing the influence of noise components. Therefore, it is possible to easily determine the preload of a large bearing such as a differential bearing.

It is a flowchart which shows one Embodiment of the preload determination method of the bearing by this invention. It is a figure which shows typically an example of the equipment structure to which this invention method is applied. FIG. 3 is a signal waveform diagram obtained by subjecting each piece of rotational vibration information from the first and second acceleration sensors in FIG. 2 to FFT. The magnitude G _nR (ω) of the real part of each frequency response function obtained by converting the FFT output signal shown in FIG. 3 and the magnitude G _n (ω of the frequency response function including the real part and the imaginary part. ) In an orthogonal coordinate system. Same as above for explaining the calculation of the natural angular frequency Ω of the bearing roller by the sum ratio f (ω) of the real part of each frequency response function and the frequency response function including the real part and imaginary part It is a graph. It is a graph for demonstrating the preload determination of a bearing based on a natural angular frequency same as the above.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, the same code | symbol shows the same or an equivalent part between each figure.
The bearing preload determination method according to the present embodiment shown in the flowchart of FIG. 1 is executed using, for example, the equipment configuration shown in FIG.
In FIG. 2, 1 a and 1 b are first and second bearings provided in the bearing 1 such as a differential bearing to which a relatively large preload is applied to the assembly, and supports the rotating shaft 2.
The acceleration sensor 11 is a sensor that measures rotational vibration information (acceleration: speed change rate) of the bearing 1. A plurality of acceleration sensors 11 a and 11 b in the illustrated example are arranged at different locations. Here, the first acceleration sensor 11a is disposed in the vicinity of the first bearing 1a, and the second acceleration sensor 11b is disposed in the vicinity of the second bearing 1b.
The first and second acceleration sensors 11 a and 11 b are each connected to a fast Fourier transform circuit (hereinafter abbreviated as FFT) 12, and the FFT 12 is connected to an analysis device 13.

As shown in FIG. 1, in the bearing preload determination method according to the present embodiment, first, in step 101, each rotational vibration information from the first and second acceleration sensors 11a and 11b is applied to the FFT 12 (with respect to the rotational vibration information). Fast Fourier transform).
FIG. 3A shows an example of a signal waveform after fast Fourier transform of rotational vibration information from the first acceleration sensor 11a, and FIG. 3B shows an example of a signal waveform after fast Fourier transform of rotational vibration information from the second acceleration sensor 11b. An example of a signal waveform is shown.
The rotational vibration information from the first and second acceleration sensors 11a and 11b is prerotated, and the rotation of the bearing 1 (first and second bearings 1a and 1b) in a state where the rotating shaft 2 to be supported is rotating. Vibration information, specifically acceleration signals.

Next, in step 102, the signal (output signal of FFT12) after the fast Fourier transform for each acceleration signal is converted into a frequency response function G (ω).
Now, _let G _n (ω) be the magnitude of the frequency response function at the angular frequency ω for the FFT output signal (the magnitude of the absolute value including the real part and imaginary part), and the angular vibration for the FFT output signal Let G _nR (ω) be the magnitude of the absolute value of the real part of the frequency response function at several ω (see FIG. 4).
Here, _n is a numerical value derived from the number of the acceleration sensor 11. Therefore, in this embodiment, G _n (ω) and G _nR (ω) derived from the first acceleration sensor 11a are G1 (ω) and G1R (ω), and G _n (ω) derived from the second acceleration sensor 11b. , G _nR (ω) becomes G2 (ω) and G2R (ω).

In step 103, the magnitude G _nR (ω) of the absolute value of the real part of the frequency response function G (ω) derived from the first and second acceleration sensors 11a and 11b (related to each acceleration signal) is also the frequency response. The total ratio f (ω) (= Σ | G _nR (ω) | / Σ | with the magnitude G _n (ω) of the absolute value of the frequency response function including the real part and the imaginary part of the function G (ω) G _n (ω) |) is calculated.
As can be seen from FIG. 4, in the vicinity of the natural angular frequency Ω, G _nR (ω) approaches 0, so f (ω) also approaches 0. That is, the angular frequency ω of f (ω) ≈0 becomes the natural angular frequency Ω.

Next, in step 104, the natural angular frequency Ω of the bearing roller (the roller portion of the bearing 1) is calculated from the total ratio f (ω). In this embodiment, from the sum ratio f (ω) calculated in step 103, f (ω) = threshold value within the preset angular frequency range of the bearing roller, the angular frequency ω _low , ω _{Determine high} . Then, the determined midpoints of the angular frequencies ω _low and ω _high are obtained and set as the natural angular frequencies Ω [= (ω _low + ω _high ) / 2] (see FIG. 5).

Finally, in step 105, the bearing preload is estimated from the natural angular frequency Ω of the bearing roller calculated in step 104, and the estimated preload (preload estimated value) is within a predetermined range (preload regulation range). Usually, it is determined whether it is within the preload standard (see FIG. 6).
The estimation of the bearing preload is performed by a correlation coefficient (straight line A in FIG. 6) between the bearing preload and the natural angular frequency Ω determined in advance by experiments or the like.
The determination result is output to an output unit such as a display of the analysis apparatus 13 and is notified to the outside. As an example, if the preload estimated value is within the preload standard, a message “preload (natural angular frequency Ω) OK” is displayed on the display unit of the analyzer 13. Is done.
In the present embodiment, the processing from step 102 to step 105 is performed by the analysis device 13.

According to the embodiment described above, the bearings are subjected to the following processing after performing fast Fourier transform on each rotational vibration information from the first and second acceleration sensors 11a and 11b and converting them to frequency response functions G (ω). A preload of 1 was judged. That is, the size G _nR absolute value of the real part of each of the above frequency response function G (ω) (ω), the magnitude of the absolute value of the frequency response function, including a real part and an imaginary part G _nR (omega ) And the sum ratio f (ω). Then, the preload of the bearing 1 is determined based on the natural angular frequency Ω of the bearing roller based on this summing ratio, and the preload determination can be performed while suppressing the influence of noise components.
Therefore, it is possible to easily determine whether or not the preload of a large bearing such as in the assembly of the bearing 1 such as a differential bearing is within a predetermined range (preload standard or the like).

1, 1a, 1b: bearing, 2: rotating shaft, 11: acceleration sensor, 12: fast Fourier transform circuit (FFT), 13: analyzer, G (ω): frequency response function, G _nR (ω): frequency response The size of the real part of the function G (ω), G _n (ω): the size including the real part and the imaginary part of the frequency response function G (ω), f (ω): G _nR (ω) and G Total ratio with _nR (ω), Ω: natural angular frequency.

Claims (2)

In the bearing preload determination method for determining the bearing preload using fast Fourier transform,
The bearing includes first and second bearings that support a rotating shaft, and is arranged in the vicinity of the first bearing and in the vicinity of the second bearing capable of transmitting vibrations of the first and second bearings . , After performing a fast Fourier transform on each rotational vibration information from the second acceleration sensor, convert to each frequency response function,
Calculate the sum ratio of the magnitude of the absolute value of the real part of each obtained frequency response function and the magnitude of the absolute value of the frequency response function including the real part and the imaginary part,
From the calculated total ratio , the natural angular frequency of the bearing roller of the bearing is calculated,
Estimate the bearing preload from the calculated natural angular frequency of the bearing roller,
A bearing preload determination method, comprising: determining whether the estimated preload is within a predetermined range.   The bearing preload determination method according to claim 1, wherein the bearing is a differential bearing.

Images