GB/T 2423.13-1997 Environmental testing for electric and electronic products Part 2: Test methods Test Fdb: Broadband random vibration - Medium reproducibility
Some standard content:
GB/T 2423. 13--- 1997
This standard is equivalent to the International Electrotechnical Commission standard IEC68-2-36 "Environmental testing Part 2: Test method Test Fdb: Wide-band random vibration - Medium reproducibility" (1973 1st edition) and Amendment No. 1 (August 1983). This standard replaces GB2423.13-82 "Basic environmental testing procedures for electric and electronic products Test Fdb: Wide-band random vibration test method - Medium reproducibility".
Chapters 1 and 2 of this standard are different from Chapters 1 and 2 of GB2423.13-82. GB2423.13-82 rewrites Chapters 1 and 2 of IEC68-2-36. This revision also adds the content of the International Electrotechnical Commission's Amendment No. 1 to IEC68-2-36 in August 1983.
This standard was first issued in 1982, revised for the first time in September 1997, and implemented on October 1, 1998. From the date of implementation of this standard, it will replace GB2423.13-82. This standard was proposed by the Ministry of Electronics Industry of the People's Republic of China. This standard is under the overall jurisdiction of the Standardization Institute of the Ministry of Electronics Industry. This standard is under the jurisdiction of the National Technical Committee for Environmental Technical Standardization of Electrical and Electronic Products. The drafting units of this standard are the Standardization Institute of the Ministry of Electronics Industry and the Fifth Institute of the Ministry of Electronics Industry. The main drafters of this standard are Zhou Xincai, Wang Shurong, Ji Chunyang, Zhang Youlan, etc. 165
GB/T2423.13-1997
IEC Foreword
1. A formal resolution or agreement on a technical issue formulated by a technical committee of the International Electrotechnical Commission with the participation of all national committees that are particularly concerned about the issue, which reflects and expresses the international consensus on the issue as much as possible. 2. These resolutions or agreements are accepted by the National Committees in the form of recommended standards for international use. 3. In order to promote international unification, the International Electrotechnical Commission hopes that all member countries will adopt the contents of the International Electrotechnical Commission recommended standards as their national standards when formulating national standards, as long as the specific national conditions permit. Any differences between the International Electrotechnical Commission's recommended standards and national standards should be clearly indicated in the national standards as much as possible. This standard was prepared by the 50A Subcommittee (Shock, Vibration and Other Dynamic Tests) of the 50th Technical Committee (Environmental Testing) of the International Electrotechnical Commission. The first draft was discussed at the Stockholm Conference in 1968, and the new draft was discussed at the Tehran Conference in 1969. As a result of this meeting, the final draft 50A (Central Office) No. 133 was submitted to the National Committees for voting according to the "six-month method" in February 1971.
The following countries voted explicitly in favor of this standard: Australia
Austria
Belgium
Czechoslovakia
Hungary
Israel
Portugal
Turkey
National Standard of the People's Republic of China
Environmental testing for electric and electronic products
Part 2: Test methods
Test Fdb :Random vibration wide band--Reproducibility medium
1 Introduction
—1997
GB/T 2423.13
idtIEC68-2-36:1973
Replaces GB2423.13--82
The basic requirements for broadband random vibration tests are given in GB/T2423.11--1997 (IEC68-2-34) Test Fd: Broadband random vibration: General requirements. In addition, three possible levels of reproducibility are specified, called high, medium and low reproducibility, and are represented by tests Fda, Fdb and Fdc respectively. Each of these test methods together with its recommended verification method constitutes a separate complete standard, so that all the information required by the specification writer is included in test Fd. The information required by the test engineer is included in tests Fda, Fdb or Fde respectively.
The users of this standard are strongly advised to read this standard in conjunction with GB/T2423.11--1997 (IEC68-2-34). It must be noted that throughout the text of the standard, two terms of particular importance in the context of random vibration testing are frequently mentioned. In order to better understand the content of this standard, the following definitions are given:
acceleration spectral density accelerationspectraldensity(ASD) Spectral density of acceleration variation, expressed as square of acceleration units per unit frequency. The frequency spectrum of the acceleration spectral density ASDspectrum The way in which the acceleration spectral density varies over a frequency range. 2 Purpose
To determine the ability of components and equipment to withstand random vibrations of specified severity levels. This random vibration test is applicable to components and equipment that may be subjected to random vibration conditions during use. The purpose of the test is to determine mechanical weaknesses and/or degradation of specified performance, and to use this information in conjunction with relevant specifications to determine whether the test specimen is acceptable. When applying the environmental stresses (conditioning tests) specified in this test, the test specimen is subjected to a given level of random vibration tests over a wide frequency band. Because the test specimen and its fixture will produce complex responses, this test requires special attention to the preparation, conduct and verification of its specified requirements.
3 Installation and Control
3.1 Installation
The test sample shall be installed on the test equipment in accordance with the requirements of GB/T2423.43-1995 (IEC68-2-47) "Installation requirements and guidelines for components, equipment and other products in dynamic tests such as impact (Ea), collision (Eb), vibration (Fc and Fd) and steady-state acceleration (Ga).
Approved by the State Administration of Technical Supervision on September 1, 1997 and implemented on October 1, 1998
3.2 Reference points and control points
GB/T 2423.13-1997
The test requirements are verified by measurements made with reference points or, in some cases, at all control points related to the fixed points of the test sample. When an assumed reference point is specified, measurements need only be made at that point. If many small test specimens are mounted on a fixture, when the lowest resonant frequency of the loading fixture exceeds the upper test frequency, the reference point and/or control point can be considered to be related to the fixture and not to the test specimen fixing point. 3.2.1 Fixing point
The fixing point is defined as the part of the test specimen that contacts the fixture or the vibration table. It is usually the place where the test specimen is normally tightened in use. If a part of the actual mounting structure is used as a fixture, the fixing points of these mounting structures should be taken as the fixing points instead of the fixing points of the test specimen.
3.2.2 Control point
The control point is usually the fixing point, or as close to the fixing point as possible. In any case, its connection with the fixing point should be rigid. If the test specimen has four or fewer fixing points, then each fixing point is used as a control point. If there are more than four points, the relevant specification should specify four representative fixing points as control points. Note
1 For large and/or complex test specimens, it is an important issue to specify control points in the relevant specification. 2 Control points refer only to points that have requirements or are representative of vibration. 3.2.3 Reference point
The reference point is a single point used to obtain a reference signal to verify the test requirements and represent the movement of the test sample. It can be a control point or an assumed point established by manually or automatically processing the signals of each control point. If an assumed point is used, the spectrum of the reference signal is defined as the arithmetic mean of the acceleration spectrum density values of all control signals at each frequency. In this case, the total root mean square value of the reference signal is equal to the root mean square of the root mean square value of the signal at each control point. The relevant provisions should state the reference point used, or how to select the reference point. It is recommended to use an assumed reference point for large and (or) complex test samples.
4 Frequency response measurement and resonance check
In the following sinusoidal vibration test stage, the tolerance shall be in accordance with the provisions of GB/T2423.10-1995 (IEC68-2-6) Test Fc, Vibration (Sinusoidal).
4.1 Sine amplitude
Unless otherwise specified in the relevant specifications, the sine amplitude used for frequency response measurement and resonance inspection shall be determined by Table 1 according to the acceleration spectrum density level. This amplitude shall be applied to the reference point. If the random vibration condition test uses the assumed point, then this sine amplitude shall be applied to the control point.
Acceleration spectrum density level
(m/s*)°/Hz
4. 8-19.2
4.2 Frequency response measurement method
(0.05—0.2)
Sine amplitude (peak)
In the verification method of acceleration spectrum density, it is often required to measure the frequency characteristics of the test sample in the predetermined direction of the sample reference point.
When measuring, the forward and reverse sine sweeps shall be performed in the entire test frequency range (f1-f.), and the sweep rate shall not exceed octaves per minute. During the frequency sweep, the sine amplitude at the reference point shall be kept constant as specified in 4.1, and the AC input voltage of the power amplifier shall be measured. 168
GB/T 2423. 13--- 1997
The measured voltage is a function of frequency and is approximately inversely proportional to the frequency response. Taking into account the displacement limit of the shaker, the amplitude of the sinusoidal acceleration can be reduced at the low frequency end, but this should be taken into account when calculating the frequency response. The acceleration amplitude should be measured at all control points and the lateral measurement should be carried out as described in 5.3. When measuring the frequency response, an excitation equalizer (an instrument used to correct the general response of an unloaded shaker), a low-pass filter (greater than, cutoff), a high-pass filter (less than fcutoff) and other broadband filters can be used. Narrow-band equalizers, such as peak-valley filters, should not be inserted during the measurement. The bee-valley amplitude ratio A./A. (see Figure 1) is the ratio of the maximum value to the minimum value on the frequency response curve. This measurement does not require the use of a precision frequency meter.
Figure 1 Determination of the peak-to-valley amplitude ratio
The peak-to-valley frequency ratio (Bm) (see Figures 1 and 2) is calculated according to the following equation: Bn = Ife- fal
Where: f, peak frequency,
f. valley frequency.
This measurement requires only a precision frequency meter. fu
The most severe pair of peaks and valleys is indicated in the verification method for the acceleration spectral density shown in Appendix A (Standard Appendix) and Appendix B (Standard Appendix). If the valley frequency ratio is to be used, A,/A and Bm should be measured for several pairs of peaks and valleys (four pairs in Figure 2), and the analytical error and residual ripple of each pair should be estimated in order to obtain the most severe pair of peaks and valleys. Figure 2 Determination of several pairs of valleys and frequency ratio
4.3 Resonance check method
If the relevant specification requires a resonance check, an initial resonance check can sometimes be performed at the same time as the frequency response measurement. During the inspection, a sinusoidal frequency sweep shall be made in both the forward and reverse directions over the entire frequency range. During the resonance check, the test specimen shall be checked to determine the frequencies of the following phenomena: 169
GB/T 2423.13--1997
a) malfunction and/or degradation of the test specimen due to vibration; b) mechanical resonance of the test specimen.
The frequency sweep may be interrupted in order to study these effects more closely and find the exact frequencies. During the initial resonance check, all frequencies and amplitudes at which these phenomena occur shall be recorded for comparison with the frequencies and amplitudes obtained during the final resonance check. The relevant specification shall specify the measures to be taken in the event of any change in the resonant frequency. During the resonance check, the test specimen shall be operated, if applicable. If the mechanical vibration characteristics of the test specimen cannot be determined because it is in operation, an additional resonance check shall be made with the test specimen in a non-operating state. Any arrangements made to detect internal effects of the test specimen shall not significantly change the overall dynamic characteristics of the test specimen. A recovery time must be specified after the conditioning test to allow the test specimen to recover to the same conditions as at the beginning of the resonance check, e.g. temperature effects.
5 Vibration Movement Requirements
5.1 Basic Movement
The basic movement of the fixed points of the test specimen shall be linear and its instantaneous acceleration values shall have the random nature of a normal (Gaussian) distribution, with the points having essentially the same movement. 5.2 Distribution
The distribution of the instantaneous acceleration values at the reference points shall normally be within the tolerance band shown in Figure 3. If assumed points are used, this distribution also applies to the control points.
Note: For most random vibration tests, this distribution falls within the tolerance band and therefore only needs to be verified in exceptional cases. However, if possible, it is recommended to measure the acceleration waveform (e.g. visually) to ensure that the peak value present is at least 2.5 times the RMS value of the signal. 99.9
(X+3a)
-30)）
-Total RMS acceleration
Figure 3 Tolerance band of instantaneous acceleration distribution
GB/T 2423.13—1997
5.3 Spectrum of acceleration spectral density and total RMS acceleration The relevant specifications shall specify the acceleration spectral density level and frequency range. The spectrum of acceleration spectral density shall be as shown in Figure 4. With these values, the nominal value of the total RMS acceleration can be determined at the same time, which can also be obtained by looking up Table 3a and Table 3b. +6dB+
7777727
Specified frequency range
(logarithmic scale)
Mi-upper tolerance limit, medium reproducibility;M-lower tolerance limit, medium reproducibility;N-nominal value of specified acceleration spectral density Figure 4 Frequency spectrum and tolerance range of acceleration spectral density The tolerances of acceleration spectral density and true value of total RMS acceleration are shown in Table 2. It can be seen from this table that the tolerance of true value of total RMS acceleration is much narrower than the tolerance of true value of acceleration spectral density. In order to verify the motion requirements, only acceleration measurements need to be made in the predetermined direction of the reference point (see 3.2.3). The RMS acceleration in the frequency band from f2 to 10f. or to 10kHz (whichever is narrower) shall not exceed 70% (-3dB) of the required total RMS acceleration within the specified frequency range. Table 2
Tolerance range
True value of acceleration spectral density
True value of total root mean square acceleration
(fi~fa)
Verification of the acceleration spectral density tolerance can be carried out by any method that meets the given tolerance. When such verification has great technical difficulties, it is recommended to select a verification method from Appendix A to Appendix C. "Guidelines to assist in making such a selection are given in Chapter 6 below. Note
In the special case of a specified shaped spectrum, the verification methods shown in Appendix A to Appendix C can still be used. 1
2 It must be noted that in order to verify the acceleration spectral density level, if the instrument error sources, such as analyzer bandwidth, sampling time, etc., are not corrected. It is not allowed to use scanning technology to automatically process the signals from each control point and establish an assumed point. Total root mean square acceleration value within the specified frequency range 5.4
Within the specified frequency range, the The required total rms acceleration values are given in Tables 3a and 3b. To verify these values, a low-pass filter is used with a cut-off frequency (3 dB) of f2. If the 3 dB bandwidth differs by more than 2 % from the equivalent noise bandwidth obtained with a white noise input signal at the filter output, this bandwidth should be taken into account when using the rms values calculated in the table. Note: To verify the total rms acceleration, it is permitted to use a scanning technique to automatically process the signals from the control points to establish a hypothetical point. 5.5 Displacement Limits
All shakers are There is a displacement limit. In order to limit the peak displacement, a high-pass filter must be inserted in front of the power amplifier. Note: If the acceleration spectrum density value must be reduced in the low-frequency region due to the displacement limit of the vibration table, the reduced value must be indicated and the agreement of both parties must be obtained.
6 Selection of verification method
GB/T2423.13--1997
In Appendix A to Appendix C, three verification methods for acceleration spectrum density spectra are specified as recommended methods. 6.1 Selection of inference
When selecting When verifying the method, the following factors must be considered: a) the frequency range specified for the test, b) special requirements of relevant specifications; c) mechanical response characteristics of the test sample; d) the thrust of the vibration table; e) the size, stiffness and mass of the vibration table motion unit; f) the stiffness and mass of the fixture; g) the type of instrument used; h) the characteristics of the instrument used (for example: filter bandwidth, dynamic range, frequency range, applicable sweep rate, hum and noise). 6.2 Applicability of the recommended verification methods
If the basic errors included in the recommended verification methods do not invalidate the test, the recommended verification methods shown in Annexes A to C may be used.
The verification method using the scanning filter technique shown in Annex A is generally applicable, but is more time-consuming than the other methods. The acceleration variation with time is recorded with a tape recorder during the conditioning test and then analyzed. If the velocity range is wide and the conditioning test time is short, then the spectrum of the acceleration spectral density needs to be verified.
If the filter bandwidth is very narrow and the resonance of the test sample has little effect on the system, then the verification method using the fixed filter technique shown in Annex B can be used. This verification method also requires the use of a tape recorder. If simple instruments are used, the test sample is very rigid, or the test sample has a much smaller total moving mass than rigidity, such as small electronic components mounted in a rigid box fixture, then the verification method using the sine scanning technique shown in Annex C has its advantages. The use of narrow-band equalizers is not allowed. This method does not require complex analysis equipment when the test is performed each time. 6.3 Hybrid verification method
This verification method is intended to give the same reproducibility. However, in some cases, the analysis error or residual ripple (defined in Annex A and Annex B) may become too large in certain parts of the frequency range, which is not allowed. In other cases, the analysis time may be too long. These problems often force the use of different verification methods in different parts of the frequency range. It is worth noting that even if a mixed verification method is used, the conditioning test should be carried out simultaneously over the entire frequency range. Even for spectra with more than one specified acceleration spectral density level, such tests should not be separated. Note: For the verification of the acceleration spectral density spectrum, existing methods with higher accuracy can also be used. 7 Initial test
The test sample should be subjected to electrical and mechanical tests in accordance with the requirements of the relevant specifications. If the relevant specifications require a resonance check before and after the conditioning test, the entire test procedure including the resonance check should be completed in one axis and repeated in the other axis. The resonance check method is specified in 4.3. 8 Excitation before conditioning test
When using sinusoidal vibration for frequency response measurement or resonance check, the time should be as short as possible and the amplitude applied is specified in 4.1. The entire test procedure, including frequency response measurements, any resonance checks and conditioning tests, shall be completed on one axis (without removing the test specimen from the shaker) and then repeated on each of the other axes in sequence. Before the formal (i.e. full-level) random vibration test, the test specimen must first be subjected to a lower level of random excitation for pre-conditioning (i.e. equalization and pre-analysis). The important issue is that the vibration level applied at this time should be kept to a minimum and the time should be kept to a minimum. 172
GB/T2423.13—1997
The pre-conditioning excitation time (i.e., establishment time) allowed before the formal random vibration test is: a) less than 25% of the specified level, no time limit; b) between 25% and 50% of the specified level, the time should not be more than 1.5 times the specified test time; c) between 50% and 100% of the specified level, the time should not be more than 10% of the specified test time. It must be noted that the above pre-conditioning excitation time should not be deducted from the specified test time. 9 Conditioning test
Unless otherwise specified in the relevant specification, the test sample shall be vibrated in three mutually perpendicular axes in sequence. The axes shall be selected so as to expose the sample failure most easily. The severity level is specified in the relevant specification. Unless otherwise specified in the relevant specification, if conditions permit, the sample shall be operated during the conditioning test to determine the electrical function and mechanical effect.
For components, the relevant specification shall specify whether electrical checks are to be carried out during the conditioning test and at which stage of the conditioning test these checks are to be carried out.
During the entire conditioning test, the total root mean square acceleration within the specified frequency range shall be measured and controlled. The applicable values are shown in Tables 3a and 3b, and the tolerances are specified in accordance with 5.3.
At the beginning and end of the conditioning test, the root mean square acceleration greater than f shall be measured. The root mean square acceleration value in the frequency band from 2 to 10f2 or to 10kHz (whichever is narrower) shall not exceed 70% (-3dB) of the total root mean square acceleration value required within the specified frequency range. When using verification methods (such as those shown in Annexes A and B), in order to verify the frequency spectrum of the acceleration spectral density, samples with instantaneous acceleration time history should be taken during the conditioning test. The minimum duration of each sample is twice the maximum averaging time of the analysis equipment used. For tests lasting no more than 10 minutes, 10 samples are sufficient. For longer durations, samples should be taken at the beginning and end of the conditioning test. If the vibration system device is changed during the long conditioning test, additional samples should be taken immediately after the change. For very long conditioning test durations, it is recommended to take additional samples during the conditioning test. Whether during or after the conditioning test, the acceleration spectral density should be verified according to the verification method used. 10 Final test
The test sample should be subjected to electrical and mechanical tests in accordance with the requirements of the relevant specifications. If a resonance check is required, the final resonance check should be carried out in accordance with the method described in 4.3. Table 3a Total RMS acceleration value
Total RMS acceleration for each frequency range for each acceleration spectral density (rectangular spectrum, unit: m/s\) Specified frequency range (fi～f)
Specified acceleration
Spectral density
(m/s*)°/Hz
5~1505~2010~15010~20020~15020~20Q20~50020~2 000|20~5 00050~50050~2 00050~5 000Total RMS acceleration
Specified acceleration
Spectral density
(m/s2)2/Hz
Specified acceleration
Spectral density
GB/T2423.13
—1997
Table 3a (end)
Specified frequency range (ff)
5~1505~20010~15010~20020~15020~20020~50020~2 00020~5 00050~50050~2 00050~5 000Total RMS acceleration
Total RMS acceleration value
Total RMS acceleration for each frequency range for each acceleration spectral density (rectangular spectrum, unit: &) Specified frequency range (～f)
5~1505~20010~15010~20020~15020~20020~50020~200020~500050~50050~200050~5000Total RMS acceleration
A1 Note
GB/T 2423.13—1997
Appendix A
(Appendix of Standard)
Verification method using scanning filter technology In order to verify whether the requirements of random vibration test have been met, this verification method requires the use of a scanning analyzer. The accuracy of the spectrum obtained by analysis depends on the characteristics of the analyzer and the spectrum to be analyzed. The curve given represents the resulting decomposition error, hereinafter referred to as the analysis error. The calculation of the analysis error curve is based on the typical response of the test sample and fixture to the vibration table system. If the analysis time is long and the conditioning test time is short, verification after the conditioning test is almost always necessary. Measurement of the A2 scanning analyzer characteristics
The 3dB, 12dB, 30dB and 50dB bandwidths (called B3, B1z, B3c and Bso respectively) should be measured, assuming that the shape of the filter is independent of the center frequency.
The bandwidth factor CB is defined as a function of the filter shape factor B12/B: C = 0.1+
with the following conditions:
B12 < 2.2: B
.B30≤3.8:B
·(Al)
If these conditions are not met, the error curves in this verification method cannot be applied and the swept analyzer is not suitable for this verification method.
The relative bandwidth of the filter, B., is defined as the ratio of the 3 dB bandwidth of the analyzer to the center frequency to which it is tuned at a given moment in the analysis. Note: This measurement should only be performed when necessary. A3
Analysis error estimation
The frequency response measurement should be performed according to 4.2.
The analysis error depends on the equivalent relative bandwidth B., B. Defined as: B = CCB,
·(A2)
, where B*m is the order ratio corresponding to the amplitude ratio, see Figure A1 and Figure A2. If you do not intend to use a very precise frequency meter to achieve frequency response measurement to determine B, then C=1. A,After finding the value of /An, select the appropriate curve in Figures A1 and A2 and then read the analytical error at the value of B. calculated above. Linear interpolation between the curves is allowed. In order to find the pair of valleys with the maximum error, several pairs of peaks and valleys must usually be studied. Note: When this verification method is combined with the method shown in Appendix B and the estimated residual ripple is lower than A./A. obtained from the frequency response measurement, the lower value can be used as A./A for the relevant part of the frequency range in Figures A1 and A2, and it is the same as Cr obtained by the method of Appendix B. 175
GB/T2423.13—1997
Analysis error in the trough area
·See example
0.2 0.30.5
Figure A2Analysis error in the peak area
25dB 2. 7%
18dB 2.0%
14dB 1.7%
10dB 1.5%
3.5dB1.3%
Au/ An B'ma
Equivalent relative bandwidth
Ap/ An B'pn
3.5dB1.3%
36dB 5.0%
Example: Using a 4% sweep analyzer, the waveform and waveform factor of the filter in the analyzer are measured as follows: B30 2. 4B
Bso 4. 0,
Biz — 1. 55Bwww.bzxz.net
Cg = 0.1 + 1.55× 2.4
=0.75,B,=4%
The peak-to-valley amplitude measured by the rate response is
=7dB, and the corresponding peak-to-valley frequency ratio is Bp-3.9%. Using Bm=3.9%, then: Bm =
× 0. 75 × 4% = 1%
This gives an estimated analysis error of 1.9dB in the trough area and 0.95dB in the peak area. If the frequency response measurement that has been carried out is not to measure B with a very precise frequency meter, then C = 1, and B. = 1X0.75×4% = 3%
Therefore, it can be seen that the analysis error is large, that is, 3.6dB in the trough area and 2.3dB in the peak area. After studying other pairs of peaks and valleys, it is found that the above errors are the largest. 176
Tip: This standard content only shows part of the intercepted content of the complete standard. If you need the complete standard, please go to the top to download the complete standard document for free.