This standard specifies the bending resonance method for determining the damping performance of acoustic materials. This standard is applicable to uniform and layered uniform acoustic materials. Its measurement results can provide a basis for acoustic structure design and noise vibration control design, and can also be used to evaluate the vibration damping effect of various composite structure specimens. GB/T 16406-1996 Acoustics Bending resonance test method for damping performance of acoustic materials GB/T16406-1996 Standard download decompression password: www.bzxz.net
This standard specifies the bending resonance method for determining the damping performance of acoustic materials. This standard is applicable to uniform and layered uniform acoustic materials. Its measurement results can provide a basis for acoustic structure design and noise vibration control design, and can also be used to evaluate the vibration damping effect of various composite structure specimens.

GB/T 16406—1996
This standard is formulated with reference to the international standard ISO6721-3;1994 "Plastics--Measurement of dynamic mechanical properties--Part 3: Flexural vibration-resonance curve method" and the American Society for Testing and Materials standard ASTME756-83 "Standard method for measuring vibration damping properties of materials". The flexural resonance test method of acoustic materials has been widely used in my country's industrial and scientific research departments for many years. In order to timely convert international standards into Chinese standards and facilitate independent use and adapt to my country's national conditions, when compiling this standard, the cantilever beam and free beam test methods for two types of test samples, uniformly hooked specimens and composite specimens, are compiled into this standard at the same time. In this way, the needs of material testing and damping effect evaluation are taken into account. Appendix A and Appendix B of this standard are indicative appendices. This standard is proposed and managed by the National Technical Committee for Acoustics Standardization. The drafting units of this standard are: Institute of Acoustics, Chinese Academy of Sciences, 715th Institute of China Shipbuilding Industry Corporation, Tianjin Rubber Industry Research Institute, and 71st Institute of China Shipbuilding Industry Corporation. The main contributors to this standard are Zhang Tonggen, Lu Jinbao, Chen Yaohui and Shao Hanlin. 1 Scope
National Standard of the People's Republic of China
Acoustics-Flexural resonance testing methodfor damplng properties of acoustic materials
Acoustics-Flexural resonance testing methodfor damplng properties of acoustic materials This standard specifies the flexural resonance method for determining the damping properties of acoustic materials: GB/T 16406—1996
This standard is applicable to homogeneous and layered homogeneous acoustic materials. The measurement results can provide a basis for the design of acoustic structures and the design of noise and vibration control, and can also be used to evaluate the vibration damping effect of various composite structure specimens. The test method specified in this standard has a lower limit of 0.5MPa for the measurement of storage modulus and a measurement range of 10 to 101 for the loss factor in the frequency range of 10Hz to 1000Hz. 2 Referenced standards
The clauses contained in the following standards constitute the clauses of this standard through reference in this standard. When this standard is published, the versions shown are valid. All standards will be revised. The parties using this standard should explore the possibility of using the latest versions of the following standards: GB/T3974:1996 Acoustic terminology
GB4472-84 General rules for determination of density and relative density of chemical products GB 9870-88 General requirements for dynamic testing of elastomers IS0 10112-1991 Damping materials-Graphical method of complex modulus 3 Definitions and symbols
This standard adopts the following definitions, which are applicable to the condition of linear deformation. The acoustic terms used in this standard follow the rules of GB/T3917. 3.1 Complex flexural modulus (Er) flextlral camplex motlulus The complex ratio between bending stress and bending strain, unit: Pa, Pa. Er = E -E
Note: Complex bending modulus and complex tensile modulus are often also called complex Young's modulus, but the results of bending test and bell test are comparable only when the stress-strain relationship is linear and the sample structure is uniform. 3.2 Storage flexural modulus (Ei) flexural storage modulus The real part of the complex flexural modulus, unit: Pa. 3.3 Loss modulus () lexural loss modulus The imaginary part of the complex flexural modulus, unit: Pa. 3.4 (Material) loss factor (tane) loss factor (of a material) The ratio of the loss modulus to the storage modulus.
tanp, = Ee/E
State Administration of Technical Supervision1996-05-27Approved1996-12-01Verified
CB/T 164061996
3.5 Composite (specimen) loss factor (n) compositelossfactor A measure of the ability of a composite specimen to dissipate energy. Its value is proportional to the ratio of the damping energy of the specimen to the strain energy. 4 Principle
This standard stipulates the use of rectangular strip specimens, and the measurement principle diagram is shown in Figure 1. There are two test methods. Method A is to install the specimen vertically, with the upper end rigidly clamped and the lower end free, referred to as the cantilever beam method + Method B is to install the specimen horizontally and suspend it at the vibration node of the specimen with two thin wires, referred to as the free beam method. The cantilever method is suitable for most types of materials, including softer materials. The free beam method is suitable for testing rigid and straight specimens. For softer materials, they should be attached to metal plates to make composite specimens for testing. The test system consists of two parts: excitation and detection. The signal generator excites the magnetic transducer to apply a simple harmonic excitation force to the sample. The detection transducer detects the movement signal of the sample, which is amplified and sent to the indicating and recording instrument. The constant excitation force is maintained, the frequency is continuously changed, and the velocity bending resonance curve of the sample is measured. According to the bending resonance frequency and the peak width of the resonance, the storage bending modulus and loss factor can be calculated. Constant overflow index
Thinking line,
To amplifier
To value generator
A) Inductive arm beam method
Monitoring instrument
New frequency meter
Transducer
Constant camera
One sample
Transducer
To excitation transducer
Signal generator
Oscilloscope
Excitation transducer|| tt||Moving node
Detection transducer
B) Free beam method
To detection transducer
Amplifier
(\) Block diagram of measuring instrument
Figure 1 Schematic diagram of measurement
Recording indication
This standard stipulates that the resonance peak width refers to the frequency difference at the position where the amplitude is 0.707 of the resonance amplitude (i.e. 3dB down) on both sides of the common grip frequency. Since the energy is proportional to the square of the amplitude, the resonance peak width is also often called the half-power bandwidth or -3dB bandwidth. Note: In the two test methods specified in this standard, the sample is in a state of vibration, and its displacement, velocity and acceleration responses have different common frequencies, among which the velocity resonance frequency coincides with the natural frequency of the sample, and the displacement and acceleration resonance frequencies both deviate from the surrounding frequency. The greater the damping of the sample, the greater the difference. Although this difference can be ignored when tan<0.1, the force method for measuring displacement response can be used for testing. However, the relationship between the common peak width of the displacement resonance curve and the loss factor becomes quite complicated when the damping is high. Therefore, this standard recommends the use of a test device with a non-contact velocity transducer.
5 Test device
5.1 Most
GB/T 16406—1996
5.1.1 Vernier caliper: used to measure the length of the sample, the minimum division should not be greater than 0.05mm. 5.1.2 Micrometer: used to measure the width and thickness of the sample + the minimum division should not be greater than 0.002 mm. 5.1.3 Balance, the sensitivity should not be greater than 0.001g. 5.1.4 Thermometer; the resolution should reach 0.1℃. 5.2 Measuring bracket
The cantilever beam or free beam measuring bracket should avoid external mechanical vibration and meet the following requirements. 5.2.1 The inherent frequency of the measuring plate should be far away from the test frequency range. The measuring bracket should have a heavy base. 5.2.2 The clamp of the cantilever beam measuring bracket should have sufficient holding force to prevent additional friction damping (see Appendix A). 5.2.3 The suspension wire of the free beam measuring bracket should be soft and thin, and the length should not be shorter than 30cm. Silk or cotton thread should be used for the suspension wire. 5.3 Electromechanical transducer
The excitation transducer should be an electromagnetic transducer. It is recommended to use a non-contact velocity transducer (such as an electromagnetic transducer) as the detection transducer. In the case of tan<0.1, a displacement transducer (such as a capacitive transducer) can also be used for detection. The fluctuation of the transducer's recording sensitivity within the test frequency range of each resonant mode should be less than or equal to 0.05B. 5.4 Measuring instruments
Within the test frequency range, the measuring instruments shall meet the following requirements: 5.4.1 The frequency stability of the signal generator shall not be less than 1×104/h, the output power shall ensure that the signal-to-noise ratio is greater than 30 dB during detection, and the output amplitude stability shall not be less than 1×10 /h. 5.4.2 The time base stability of the frequency meter shall not be less than ±2×10-1/d, and the resolution shall be equal to or better than 0.1Hz. 5.4.3 The frequency response fluctuation of the amplifier shall be less than 0.1dB. 5.4.4 The nonlinearity of the indicating instrument and recorder shall not be greater than 1%. 5.4.5 This standard recommends the use of a digital storage oscilloscope, which can be used as both a display and a recording instrument. 5.4.6 The temperature non-uniformity along the length of the sample in the constant temperature box should not exceed ±1. During each measurement, the temperature should remain stable and its change should not exceed ±0.5. The gas in the constant temperature box can be air or inert gas. When measuring the temperature, the heating rate should be stable and controllable. The heating rate should not exceed 5/min.
5.4.7 In the detection system, if a filter is selected, when measuring the resonance curve, a tracking filter should be used, and the filter gain change should not exceed 0.1d. When measuring by attenuation method, it is not appropriate to use a one-third octave or octave filter. It is recommended to use a high-pass and low-pass combination filter. The passband width should be much larger than the bandwidth of the sample resonance curve. 6 Samples
6.1 Sample preparation
6.1.1 The physical properties of the sample should be uniform, and can refer to the requirements of 7.1 and 8.1 in GB 9870. If there are other standard provisions for the material, the sample can be prepared according to the requirements of the corresponding standard.
6.1.2 The composite specimen should be composite in the thickness direction. Usually, the composite specimen composed of the metal layer and the damping layer is constructed with different thickness ratios for testing. When using adhesives, the modulus of the adhesive after curing should be higher than the modulus of the damping material, and the thickness of the adhesive layer should not exceed 0.05 mm.
6.1.3 When comparing different materials and evaluating their damping effects, the composite method with a mass ratio of 1:5 between the damping material (including free damping beam damping) and the metal bottom layer should be preferred. When the weight factor is not taken into account, a composite form with a thickness ratio of 2:1 can be used. The sample preparation process should be carried out according to the technical requirements of the actual application of the product. 6.1.4 For non-magnetic samples, a piece of ferromagnetic thin film can be glued to each end of the sample. The attached mass should be less than 1% of the sample mass. In order to avoid introducing additional stiffness, the distance between the gluing position and the end point should not exceed 2% of the sample length. 6.2 Sample size
GB/T16406-1996
6.2.1 The length of the sample is related to the frequency to be measured. The thickness of the sample should be selected appropriately to ensure that it is straight and has a certain bending stiffness. In general, the ratio of the length to the thickness of the sample should not be less than 50, and the width of the sample should be less than half a wavelength. The sample size is measured at laboratory temperature, and the influence of thermal expansion and contraction is not considered. According to the above principles, the sample size can be selected in the following range: Length: 150mm~300mm
Width: 10mm~20mm
Thickness: 1 mm~3 mm
For materials with good uniformity, in order to make a unified comparison, the recommended sample size is: free length of cantilever beam sample 180mm, width 10mm; free beam sample length 150mm, width 10mm. 6.2.2 When measuring with a composite sample, the thick substrate can be steel plate or aluminum plate, and a 1mm thick cold-rolled steel strip is recommended. For composite specimens in cantilever beam mode, the root of 2mm to 25mm should be kept free of the material to be tested for clamping. It can also be processed into a thickened metal root, and the original thickness of the root should not be less than the thickness of the composite layer. 6.3 Number of specimens
When testing at only one temperature, there should be no less than 3 specimens of the same material and size. When variable temperature measurement is required, one of them can be selected for testing.
6.4 Sample environment adjustment
The storage of the specimen and the temperature and temperature adjustment before the test can refer to the requirements of Chapter 8 of GB9870. 7 Test procedure
7.1 Measurement of material density
According to the provisions of GB 1172, the density measurement accuracy should not be less than 0.5%. 7.2 Measurement of the cross-sectional dimensions of the specimen
The measurement accuracy of the width and thickness of the specimen should not be less than 0.5%. When measuring the thickness of the specimen along the length of the specimen, the average value should be obtained by measuring five points. If the thickness of any measuring point exceeds ±3% of the average value, the specimen cannot be used to accurately measure the storage modulus value, but can be used to measure the loss factor.
7.3 Installing the specimen
7.3.1 Cantilever beam method
After tightening the specimen with the cantilever beam measuring bracket, measure the free length of the specimen, and the accuracy should be not less than 0.5%. 7.3.2 Free beam method
The following steps should be followed:
a) Measure the specimen length, and the accuracy should be not less than 0.2%. b) Draw the node position marking line. The distance from each node to the end of the specimen can be calculated according to the following formula: LL/1 = 0. 224
L./=0.660/(2i+1)
Where: 2——specimen length, mm
iResonance order.
c) When hanging the sample, pay attention to the fact that when testing in different resonance modes, it should be suspended horizontally at the node position of the corresponding order. 7.4 Transducer position adjustment
·(2)
The distance between the transducer and the sample should be far enough so that the static magnetic attraction does not affect the test results. In general, when measuring the first-order vibration, the recommended distance is greater than 3mm. When measuring the highest-order vibration, the spacing can be reduced to 1mm. 7.5 Temperature adjustment
GB/T 16406--1996
Adjust the temperature in the constant temperature box according to the test purpose. In general, it should be measured from low to high in a heating sequence. The heating rate is 1℃/min to 2C/min, and the temperature increment is 10℃. In the transition area, the temperature increment can be reduced to 2℃ to 5℃. At each temperature point, it should be kept warm for 1.0 min before measurement.
7.6 Measurement and Recording
Adjust the signal generator and measuring amplifier to measure the resonance frequency and resonance peak width position of the sample. Set the sweep frequency range of the signal generator and record the resonance curve with a recorder. When measuring and recording the resonance curve, the amplitude measurement accuracy should be no less than 0.5%, the resonance frequency measurement accuracy should be no less than 1%, and the resonance peak width measurement resolution should be at least 1% of the resonance peak width. 7.7 Notes
7.7.1 In the cantilever beam test mode, the second to fourth order vibration mode is usually used for testing. In the free beam test mode, the first three order vibration modes are usually used for testing. When using a composite specimen to evaluate the vibration damping effect, under the condition of the same specimen structure, the test results of the cantilever beam second order vibration mode and the free beam first order vibration mode are equivalent. 7.7.2 During the test process, if abnormal phenomena are found (such as asymmetric resonance curve), in addition to checking whether the node position and transducer installation position are appropriate, nonlinear testing can be further performed (see Appendix A). 7.8 Composite specimen test
7.8.1 The complex bending modulus of the damping material measured by the composite specimen method should be carried out in two steps: a) measuring the resonant frequency and storage bending modulus of the metal substrate (because the loss constant of the metal beam is about 0.001 or lower, it is assumed to be a piano when calculating),
b) measuring the resonant frequency and resonant peak width after making the composite specimen. The complex bending modulus of the damping material can be calculated according to formulas (7) to (23) from the data obtained in the two measurements. 7.8.2 In order to avoid the influence of the moment of inertia and shear deformation, the thickness ratio of the composite specimen should not be greater than 4. and meet the following requirements: (f,/.)(1 + DT) 2 1. 1
Where: fa is the first-order resonance frequency of the composite specimen, Hz; f is the first-order resonance frequency of the metal substrate, Hz; D is the ratio of the density of the damping material to the density of the metal material; T is the ratio of the thickness of the damping layer to the thickness of the metal layer. -+(3)
This method is suitable for measuring damping materials with higher modulus, that is, materials with E>100MPa in the transition zone from glassy state to glassy state to rubbery state
7.9Alternative method
For materials with very small damping, the resonance peak width is very small, and the water standard allows the attenuation method to be used for measurement instead. When the sample resonates, the excitation signal is disconnected, and the sample enters a free decay vibration state. The waveform of the vibration decaying with time is measured, and the loss factor tⅡ is calculated from the logarithmic reduction rate 4:4 -In(X./X.+1)
tan,=A/yuan
Where X, and X,+1 are two adjacent amplitude values ​​of vibration velocity or vibration displacement in the same direction. In order to improve the accuracy of egg measurement, the 9th amplitude and the 10th amplitude can be measured, and then calculated as follows: 11n(Xg/Xn)
tand,-
wherein the amplitude ratio K,/X,+, takes a value not greater than 3. 8 Result calculation
8.1 Uniform sample
The bending modulus and loss factor of the uniform sample are calculated by the following formula: E: =[4π(30)1/398/h(f/)3
wherein, E
energy bending modulus, Pa
loss bending modulus, Pal
tanar——bending loss factor
——sample material density, kg/m
GB/T 164061996
tandr=Af/f
E - E(tano.
1…- sample length in free beam mode; in cantilever beam mode, it is sample free length, III; h- sample thickness, m
- resonance order:
f,—— ith order common frequency, Hz
Af- ith order resonance peak width, Hz
- numerical calculation factor for the ith order resonance, determined by the following formulas: For cantilever beam mode:
— 22. 0
k = (— 0. 5)3
For free beam mode;
group = ( + 0.5)3
8.2 Composite sample
(i = 1)
(i = 2)
(i = 2)
( > 2)
（8）
Pasting the damping material on one side of the metal plate is a free damping structure commonly used in engineering. The loss factor of the composite specimen and the complex bending modulus of the damping material are calculated by the following formula: Af.
Ei - E% (u - u) + Vu - w) 4T(1 - u)2r
tano, = n I+MT ×I+4MT + 6MT=AMI\ + MT3+6T+4T+2MT+MT4
u = (1 + DT).7F)3
#=4+6T+4T
M = El/Elu
（18)
(19）
( 20 )
（21）
Wherein, — loss factor of composite specimen;
GB/T 16406—1996
T =h/hy
-the ith-order resonance frequency of composite specimen, Hz
the ith-order resonance frequency of metal substrate, Hz;
Afthe width of the ith-order common peak of composite specimen, Hz:E(——storage bending modulus of damping material, PaE
storage bending modulus of metal substrate, Pa;tand-—loss factor of damping material:
D—density ratio;
Tthickness ratio.
β—density of damping material, kg/m\
P——density of metal material, kg/m,
h: thickness of damping layer, m:
h. thickness of metal plate.m.
9 Drawing
Drawing the graph of loss factor of composite specimen or complex bending modulus of material changing with temperature can be in two forms (22)
（23)
a) Temperature variation curve graph of fixed resonance mode·Due to the common frequency, the loss factor or complex bending modulus of composite specimen changes with temperature: When drawing, the data table should be listed or marked on the graph: b) Temperature variation curve graph of fixed frequency. This form of graph should be tested with temperature variation of different resonance orders, and then the graph should be drawn by interpolation method. The drawing method can refer to IS0 10112. 10 Measurement uncertainty
10.1 Uniform specimen
10.1.1 Storage flexural modulus
The test is carried out in accordance with the requirements specified in this standard. Within the range of the fourth-order vibration mode, the measurement uncertainty of the storage modulus is not more than 5%. When testing in the resonance mode above the fourth order, if the influence of shear deformation is not considered, the error will increase. For uniformly layered specimens, E represents the equivalent modulus of the multilayer system.
10.1.2 Loss factor
The measurement uncertainty of the loss factor is closely related to the numerical value of the loss factor itself and the frequency stability and resolution of the measurement and recording system. When the loss is small, the resolution requirement is much higher than when measuring the storage flexural modulus. The relationship between the measurement uncertainty of loss factor () and the frequency measurement uncertainty () is
u() = 2u(f)/tand,
For example, in the case of 4(f)-0.1%, when tn=0.1, the measurement uncertainty of loss factor is 1.4%, and when tangr=0.01, the measurement uncertainty of loss factor increases to 14%. 10.2 Composite specimens
When using composite specimens to measure the complex bending modulus of damping materials, the measurement uncertainty of the loss factor of the composite specimen is not greater than 20%, and the measurement uncertainty of the storage bending modulus of damping materials is not greater than 25%. 11 Test report
The test report shall include the following:
Material name;
GB/T 16406—1996
Production order. Sample preparation unit and inspection unit;
This standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions; d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping specimens, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B42 Composite specimen
(i = 1)
(i = 2)
(i = 2)
( > 2)
（8）
Pasting the damping material on one side of the metal plate is a free damping structure commonly used in engineering. The loss factor of the composite specimen and the complex bending modulus of the damping material are calculated by the following formula: Af.
Ei - E% (u - u) + Vu - w) 4T(1 - u)2r
tano, = n I+MT ×I+4MT + 6MT=AMI\ + MT3+6T+4T+2MT+MT4
u = (1 + DT).7F)3
#=4+6T+4T
M = El/Elu
（18)
(19）
( 20 )
（21)
Wherein, —composite specimen loss solid number;
GB/T 16406—1996
T =h/hy
-the i-th order resonance frequency of the composite sample, Hz
the i-th order resonance frequency of the metal substrate, Hz;
Afthe width of the ith order common peak of the composite sample, Hz:E(——storage bending modulus of the damping material, PaE
storage bending modulus of the metal substrate, Pa;tand-—loss factor of the damping material:
D—density ratio;
Tthickness ratio.
β—density of the damping material, kg/m\
P——density of the metal material, kg/m,
h: thickness of damping layer, m:
h. thickness of metal plate.m.
9 Drawing
Drawing the graph of loss factor of composite specimen or complex bending modulus of material changing with temperature can be in two forms (22)
（23)
a) Temperature variation curve graph of fixed resonance mode·Due to the common frequency, the loss factor or complex bending modulus of composite specimen changes with temperature: When drawing, the data table should be listed or marked on the graph: b) Temperature variation curve graph of fixed frequency. This form of graph should be tested with temperature variation of different resonance orders, and then the graph should be drawn by interpolation method. The drawing method can refer to IS0 10112. 10 Measurement uncertainty
10.1 Uniform specimen
10.1.1 Storage flexural modulus
The test is carried out in accordance with the requirements specified in this standard. Within the range of the fourth-order vibration mode, the measurement uncertainty of the storage modulus is not more than 5%. When testing in the resonance mode above the fourth order, if the influence of shear deformation is not considered, the error will increase. For uniformly layered specimens, E represents the equivalent modulus of the multilayer system.
10.1.2 Loss factor
The measurement uncertainty of the loss factor is closely related to the numerical value of the loss factor itself and the frequency stability and resolution of the measurement and recording system. When the loss is small, the resolution requirement is much higher than when measuring the storage flexural modulus. The relationship between the measurement uncertainty of loss factor () and the frequency measurement uncertainty () is
u() = 2u(f)/tand,
For example, in the case of 4(f)-0.1%, when tn=0.1, the measurement uncertainty of loss factor is 1.4%, and when tangr=0.01, the measurement uncertainty of loss factor increases to 14%. 10.2 Composite specimens
When using composite specimens to measure the complex bending modulus of damping materials, the measurement uncertainty of the loss factor of the composite specimen is not greater than 20%, and the measurement uncertainty of the storage bending modulus of damping materials is not greater than 25%. 11 Test report
The test report shall include the following:
Material name;
GB/T 16406—1996
Production order. Sample preparation unit and inspection unit;bzxZ.net
This standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions; d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping specimens, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B42 Composite specimen
(i = 1)
(i = 2)
(i = 2)
( > 2)
（8）
Pasting the damping material on one side of the metal plate is a free damping structure commonly used in engineering. The loss factor of the composite specimen and the complex bending modulus of the damping material are calculated by the following formula: Af.
Ei - E% (u - u) + Vu - w) 4T(1 - u)2r
tano, = n I+MT ×I+4MT + 6MT=AMI\ + MT3+6T+4T+2MT+MT4
u = (1 + DT).7F)3
#=4+6T+4T
M = El/Elu
（18)
(19）
( 20 )
（21)
Wherein, —composite specimen loss solid number;
GB/T 16406—1996
T =h/hy
-the i-th order resonance frequency of the composite sample, Hz
the i-th order resonance frequency of the metal substrate, Hz;
Afthe width of the ith order common peak of the composite sample, Hz:E(——storage bending modulus of the damping material, PaE
storage bending modulus of the metal substrate, Pa;tand-—loss factor of the damping material:
D—density ratio;
Tthickness ratio.
β—density of the damping material, kg/m\
P——density of the metal material, kg/m,
h: thickness of damping layer, m:
h. thickness of metal plate.m.
9 Drawing
Drawing the graph of loss factor of composite specimen or complex bending modulus of material changing with temperature can be in two forms (22)
（23)
a) Temperature variation curve graph of fixed resonance mode·Due to the common frequency, the loss factor or complex bending modulus of composite specimen changes with temperature: When drawing, the data table should be listed or marked on the graph: b) Temperature variation curve graph of fixed frequency. This form of graph should be tested with temperature variation of different resonance orders, and then the graph should be drawn by interpolation method. The drawing method can refer to IS0 10112. 10 Measurement uncertainty
10.1 Uniform specimen
10.1.1 Storage flexural modulus
The test is carried out in accordance with the requirements specified in this standard. Within the range of the fourth-order vibration mode, the measurement uncertainty of the storage modulus is not more than 5%. When testing in the resonance mode above the fourth order, if the influence of shear deformation is not considered, the error will increase. For uniformly layered specimens, E represents the equivalent modulus of the multilayer system.
10.1.2 Loss factor
The measurement uncertainty of the loss factor is closely related to the numerical value of the loss factor itself and the frequency stability and resolution of the measurement and recording system. When the loss is small, the resolution requirement is much higher than when measuring the storage flexural modulus. The relationship between the measurement uncertainty of loss factor () and the frequency measurement uncertainty () is
u() = 2u(f)/tand,
For example, in the case of 4(f)-0.1%, when tn=0.1, the measurement uncertainty of loss factor is 1.4%, and when tangr=0.01, the measurement uncertainty of loss factor increases to 14%. 10.2 Composite specimens
When using composite specimens to measure the complex bending modulus of damping materials, the measurement uncertainty of the loss factor of the composite specimen is not greater than 20%, and the measurement uncertainty of the storage bending modulus of damping materials is not greater than 25%. 11 Test report
The test report shall include the following:
Material name;
GB/T 16406—1996
Production order. Sample preparation unit and inspection unit;
This standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions; d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping specimens, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B4m.
9 Drawing
Drawing the graph of the loss factor of the composite sample or the complex bending modulus of the material with temperature can be in two forms (22)
（23)
a) Temperature-varying curve graph of fixed resonance mode Due to the common frequency, the loss factor or complex bending modulus of the composite sample varies with temperature: When drawing, the data table should be listed or marked on the graph: b) Temperature-varying curve graph of fixed frequency. This form of graph should be tested with different resonance orders and then drawn by interpolation method. The drawing method can refer to IS0 10112. 10 Measurement uncertainty
10.1 Uniform sample
10.1.1 Storage bending modulus
The test shall be carried out in accordance with the requirements specified in this standard. Within the range of not more than the fourth-order vibration mode, the measurement uncertainty of the storage modulus shall not exceed 5%. When testing in resonance mode above the fourth order, if the influence of shear deformation is not considered, the error will increase. For uniformly layered specimens, E represents the equivalent modulus of the multilayer system.
10.1.2 Loss factor
The measurement uncertainty of the loss factor is closely related to the value of the loss factor itself and the frequency stability and resolution of the measurement and recording system. When the loss is small, the resolution requirement is much higher than when measuring the storage bending modulus. The relationship between the measurement uncertainty of the loss factor () and the frequency measurement uncertainty () is
u() = 2u(f)/tand,
For example, in the case of 4(f)-0.1%, when tn=0.1, the measurement uncertainty of the loss factor is 1.4%, and when tangr=0.01, the measurement uncertainty of the loss factor increases to 14%. 10.2 Composite specimens
When using composite specimens to measure the complex bending modulus of damping materials, the measurement uncertainty of the loss factor of the composite specimen is not more than 20%. The measurement uncertainty of the storage bending modulus of the damping material is not more than 25%. 11 Test report
The test report shall include the following:
Material name;
GB/T 16406—1996
Production order. Sample preparation unit and inspection unit;
This standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions; d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping specimens, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B4m.
9 Drawing
Drawing the graph of the loss factor of the composite sample or the complex bending modulus of the material with temperature can be in two forms (22)
（23)
a) Temperature-varying curve graph of fixed resonance mode Due to the common frequency, the loss factor or complex bending modulus of the composite sample varies with temperature: When drawing, the data table should be listed or marked on the graph: b) Temperature-varying curve graph of fixed frequency. This form of graph should be tested with different resonance orders and then drawn by interpolation method. The drawing method can refer to IS0 10112. 10 Measurement uncertainty
10.1 Uniform sample
10.1.1 Storage bending modulus
The test shall be carried out in accordance with the requirements specified in this standard. Within the range of not more than the fourth-order vibration mode, the measurement uncertainty of the storage modulus shall not exceed 5%. When testing in resonance mode above the fourth order, if the influence of shear deformation is not considered, the error will increase. For uniformly layered specimens, E represents the equivalent modulus of the multilayer system.
10.1.2 Loss factor
The measurement uncertainty of the loss factor is closely related to the value of the loss factor itself and the frequency stability and resolution of the measurement and recording system. When the loss is small, the resolution requirement is much higher than when measuring the storage bending modulus. The relationship between the measurement uncertainty of the loss factor () and the frequency measurement uncertainty () is
u() = 2u(f)/tand,
For example, in the case of 4(f)-0.1%, when tn=0.1, the measurement uncertainty of the loss factor is 1.4%, and when tangr=0.01, the measurement uncertainty of the loss factor increases to 14%. 10.2 Composite specimens
When using composite specimens to measure the complex bending modulus of damping materials, the measurement uncertainty of the loss factor of the composite specimen is not more than 20%. The measurement uncertainty of the storage bending modulus of the damping material is not more than 25%. 11 Test report
The test report shall include the following:
Material name;
GB/T 16406—1996
Production order. Sample preparation unit and inspection unit;
This standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions; d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping specimens, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow resonance peak half width tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B4Sample preparation unit and inspection unit;
Standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions: d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
-Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping samples, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B4Sample preparation unit and inspection unit;
Standard number and test method type:
Specimen number, specimen size, preparation process, damping structure type and storage conditions: d)
Testing equipment:
Test nature, divided into one type:
Plastic test (finalization or identification)
Conventional test
-Sampling test
Test results:
Indicate whether the test is carried out according to the conditions required by this standard, the data processing method for high damping samples, whether nonlinear test is carried out, what type of correction formula is used to process the data, etc. (see Appendix A, Appendix B); i) Test personnel and test period.
GE/T 164061996
Appendix A
(Suggested Appendix)
Test for deviation from linear viscoelastic behavior
Figure A1 shows the curve of amplitude X decaying with time in the form of a single logarithmic graph. Curve 1 is a typical linear viscoelastic state. Since the amplitude decays exponentially with time, it is a straight line on the single logarithmic graph; Curve 2 is a nonlinear viscoelastic state. The amplitude decays with time in a high-order exponential law, so the initial part of the curve is curved in the graph; Curve 3 is a friction damping state. The amplitude decays with time in an indirect linear relationship, so it is a curve with the opposite curvature to the above. The looseness between the fixture and the sample, the induction line is not at the vibration node position of the sample, and the friction between the vibrating sample and the air medium will produce this form of additional damping. When the damping of the sample is very small, the influence of friction damping is significantly increased.
Because in the scope of this standard, the sample is subjected to vibration, only a very weak non-harmonic deformation is produced. If the vibration waveform is observed, it is difficult to see + it can only be tested by vibration free decay. Figure A1 Curve of amplitude free decay over time Appendix B
(Suggested Appendix)
High damping data processing method
The test principle of the common report curve method is based on the theory of linear small damping. When the ta force is 0.1, the influence of the high-order terms of the loss factor cannot be ignored. The loss factor can be calculated using the following correction formula: tand,=4
When measuring the resonance curve, the measurement result is reliable only when the relative height of the resonance peak is not less than 10dB (that is, on the resonance curve, the amplitude at the resonance peak is more than 10dB higher than the amplitude at the frequency far away from the resonance frequency). In the case of large damping, the relative height of the resonance peak may be less than 10dB, and sometimes even less than 3dB. The resonance curve may also appear asymmetric. At this time, you should first pay attention to eliminating interference. If possible, it is best to reduce the thickness ratio, remake the sample, and test it. If it is not possible to make another sample, the following formula can be used for calculation: a) The resonance peak width f. is taken as the frequency width when the resonance amplitude drops by 2dB, then the loss factor is: tand: = 1,308A//J
b) The resonance peak width △f. is taken as the frequency width when the resonance amplitude drops by 1dI3, then the loss factor is: tand: = 1.965A//1,
·B3)
GB/T 16406—1996
·When calculating the frequency width when the edge drops by 3dB, the loss factor is: tand =
When the resonance curve is asymmetric, take the narrow half-width of the resonance peak tand=
When using one or more of the above approximate formulas to process data, it should be stated in the test report. B4
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.