Measuring and testing – Vibration – Vibrator
Reexamination Certificate
2000-03-06
2001-11-13
Larkin, Daniel S. (Department: 2856)
Measuring and testing
Vibration
Vibrator
C073S011050, C073S662000, C073S579000
Reexamination Certificate
active
06314813
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of The Invention
The present invention relates in general to measuring a part, such as a brake rotor, to determine its ability to damp vibrations, and particularly relates to a method for locating vibration antinodes on such parts and calculating vibration parameters around the antinodes. The invention is an improvement of the method described in U.S. Pat. No. 6,014,899.
2. Description of Prior Developments
Although the method of determining and quantifying vibration and noise suppression as described in U.S. Pat. No. 6,014,899, is most effective, it is somewhat time consuming to complete. That is, when the method of U.S. Pat. No. 6,014,899 is carried out, measurements of the slope of the vibration decay curve of a part are taken at close intervals around the circumference of the part. Each measurement can take a significant amount of time to complete. Moreover, this prior method is subject to certain inaccuracies created by modulated, nonlinear vibration decay curves. This modulation is created by two separate modes of vibration which are present in the excited part and which can lead to significant differences in vibration damping measurements and Q-factor calculations, depending on the location on the part at which the slope of the decay curve is measured.
In the method of U.S. Pat. No. 6,014,899, it is suggested to simply eliminate those calculated values of vibration damping, known as Q-factors, which significantly exceed or differ from the average Q-factor calculated for the part. (It has now been learned that such values typically occur around vibration nodes on the part where modulated vibration decay curves are common.) Although this prior technique of simply ignoring and excluding large variations in Q-factors from the Q-factor curve fit improves the accuracy of the calculation of the average Q-factor of the part, it does not reduce the time required to calculate the average Q-factor, nor does it eliminate the need to make Q-factor calculations at each test point.
Accordingly, a need exists for a method and apparatus for reducing the time required to calculate an average or other single value representative Q-factor of a part.
A further need exists for increasing the accuracy of Q-factor measurement.
A further need exists for avoiding the use of modulated decay curves in Q-factor calculations which previously reduced the accuracy of Q-factor calculations.
Yet a further need exists for a method of determining Q-factors on a part, such that modulated nonlinear decay curves are excluded from the Q-factor calculations.
SUMMARY OF THE INVENTION
The improved method described herein has been developed to fulfill the needs noted above and therefore has as an object the provision of a method and apparatus for measuring the average Q-factor of a part in a reduced amount of time and with an increase in accuracy over prior methods. A further object is to eliminate the calculation of Q-factors which are based on modulated vibration decay curves.
These and other objects are achieved in accordance with the present invention which is directed to a method and apparatus for expediting the determination of the average or representative Q-factor of a part and for increasing the accuracy of the associated Q-factor calculation by avoiding areas on a test part likely to produce modulated vibration decay curves. The invention is based on the realization that the most accurate Q-factor calculations are determined from vibration decay slope values taken at the vibrational antinodes of the part, and that the most inaccurate Q-factor calculations are derived from vibration decay slope values taken at the vibrational nodes of the part.
Since it is now known that the value of the Q-factor varies sinusoidally around the circumference of a part, such as a circular brake rotor or brake drum, it is possible to use only a portion of such a sinusoidal curve to define the entire curve. The present invention is based on the use of only selected portions of such a sinusoidal curve to reduce the number of test points required for accurate curve fitting and to avoid testing those points on a part which are subject to modulated vibration decay curves. Instead of testing a part at closely spaced arbitrary intervals, the present invention tests a part only around one or more vibrational antinodes.
By measuring vibration decay slopes only around vibrational antinodes, fewer test points are required, and the accuracy of the test data is greatly improved. That is, the most inaccurate vibration decay slope data is typically measured around vibration nodes where the greatest sharing of vibrational energy takes place between the two “twin pair” vibration modes, as discussed in U.S. Pat. No. 6,014,899.
The twin vibration modes interact around vibrational nodes to produce non-linear modulated vibration decay curves, which are modulated by the difference between the two natural frequencies, typically a difference of about 0 to 5 Hz. This modulation is greatest at vibrational nodes and produces Q-factor values, which when plotted, can and often do define spiked portions on an otherwise sinusoidal plot. These spiked portions of the plot adversely affect the accuracy of the curve fitting and resulting overall Q-factor value obtained from the curve fit.
By locating the positions of the vibrational antinodes in advance and calculating Q-factor values only around the antinodes, the spiked portions of the sinusoidal plot of Q-factor values can be eliminated and ignored. This approach also eliminates much of the variance in Q-factor calculations due to the effects of other variables which affect decay slope measurements taken between antinode locations.
Because of part sample nonlinearity and a lack of homogeneity due to material discontinuities, the measurement of Q-factor values between vibrational antinode locations on the part sample can produce, when plotted, either maximum or minimum values. Other factors determining the production of a maximum or minimum Q-factor values include the decay range (in decibels) over which the vibration decay curve is measured, the difference in frequencies between the twin mode frequency pairs, the number of points measured on the sample part, and the modulation effects from mode pair frequency interaction during decay measurements.
The present invention takes this nonlinearity into account, and through an appropriate correlation or superposition of Q-factor curve fit with the natural frequency plot of the twin vibration modes, a consistent and accurate determination of Q-factor can be obtained.
The improved method described below determines the angular location of maximum response for each of the two frequencies of the two nodal diameter vibration modes as well as the frequencies of each. Q-factor measurements are taken only around antinodes while exciting the sample at a pre-determined frequency of the dominant one (largest response) of the two twin vibration modes at each selected antinode. A sine wave function is then fit to this abbreviated data similar to the original method of U.S. Pat. No. 6,014,899, however, the phase and amplitude information of the plot are utilized in conjunction with the zero crossing value (“C
2
”) to determine a single average or representative Q-factor at the antinodes (rather than simply at “C
2
”). This new value is a value that is more repeatable and describes the sample characteristics better than earlier methods which used only C
2
.
The improved method deals with the fact that each of the two twin mode frequencies can have a different Q-factor associated with it, and that the points of maximum response are not necessarily 45 degrees apart and not necessarily coincident with the Q-factor maximum or minimum. What used to take 45 minutes to an hour (using a semi-automated PC driven system to measure every point's frequency and damping) with the prior method now takes 15 to 20 minutes with the improved method, and achieves accuracy within about 1% versus 3% with the original method of U.S. Pat. No. 6,014,899.
The
Calcaterra Mark P.
DaimlerChrysler Corporation
Larkin Daniel S.
Saint-Surin Jacques
LandOfFree
Method and apparatus for measuring vibration damping does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for measuring vibration damping, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for measuring vibration damping will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2578660