《纺织复合材料》课程参考文献（Composite Materials Handbook，Volume 1）CHAPTER 6 LAMINA, LAMINATE, AND SPECIAL FORM CHARACTERIZATION

MIL-HDBK-17-1F Volume 1,Chapter 6 Lamina,Laminate,and Special Form Characterization 6.4.4.2 LVDT (Linear Variable Differential Transformer)deflectometers LVDT's are electromagnetic devices designed so that as a ferromagnetic core is displaced within a transformer(consisting of three windings),a linearly varying a.c.voltage and phase shift are produced, this signal is demodulated to produce a varying d.c.output.LVDT's are available in both linear and angu- lar configurations.LVDT's are available in lengths to 10 feet(3 meters),their output linearity is about 0.1%,and their maximum resolution is 1 microinch(25 um).The accuracy of a given LVDT is commonly limited to 0.01%of total travel.An LVDT may be used directly as a deflectometer with its core contacting the specimen;it can be used with a linkage;or it can be incorporated into a contacting extensometer High temperature LVDT's may be usable up to the Curie Temperature of the core material,but are gener- ally used with extensions or linkages to avoid exposing them to hostile environments.LVDT's must be calibrated at the temperature to which they will be exposed in use. 6.4.4.3 Contacting extensometers Contacting extensometers and compressometers are devices that are used to determine the relative displacements of two points on a specimen.The contact extensometer must be clamped to the specimen surface in such a way that the contact points cannot slip,and that the extensometer does not affect the test.Extensometers are relatively complex devices which rely on integral strain gages or LVDT's to con- vert the relative displacements of their attachment points into linearly related outputs.Extensometers are available in a range of fixed gage lengths from 0.500 to 2.00 in.(12-50 mm),their output linearity is 0.1%,and they can resolve displacement to 1 microinch(25 um).This resolution does not imply accu- racy or calibration.A well-made contact extensometer is accurate to 0.01%of full scale,and can meas- ure strain up to 1.00(100%).Repeatability of contacting extensometers is dependent on their maintaining a constant initial gage length,therefore,when a zero stop is provided it should always be used when at- taching the extensometer to a specimen. Contact extensometers are available which can be used at liquid nitrogen temperatures,others can safely be exposed to temperatures of 500F(260C)for extended periods of time.Extensions and link- ages are available which allow remote use of extensometers on specimens exposed to temperatures up to 3000F(1600C).ASTM E 83 requires that extensometers be calibrated at the temperature at which they will be used.Extensometer calibration should be verified whenever the extensometer is subjected to deflection exceeding the normal range,been exposed to a hostile environment,received rough handling. and whenever the knife edges or points are replaced. 6.4.4.3.1 Contacting extensometers,applications Extensometers are chosen in preference to bondable strain gages when one or more of the following conditions exist: 1.The price of individual bonded strain gages exceeds the cost of a comparable extensometer. 2.The construction of a laminate will induce a non-uniform strain field under a bonded strain gage. 3. Strains will exceed the practical limit of bonded strain gages(0.03 or 3%). 4. The net deformation of a complex structure or assembly is required (for example a bonded or bolted joint). 5. When specimen conditioning or preconditioning will not allow proper bonding of strain gages. Extensometers are not recommended when the following circumstances apply: 1.Extensometers fitted with points or knife edges may cause premature failures in notch sensitive materials. Extensometers of large inertial mass respond unpredictably to rapid changes in strain. 3. Catastrophic failure of a specimen while an extensometer is attached will result in damage to the extensometer requiring repair and recalibration or replacement. 6-16

MIL-HDBK-17-1F Volume 1, Chapter 6 Lamina, Laminate, and Special Form Characterization 6-16 6.4.4.2 LVDT (Linear Variable Differential Transformer) deflectometers LVDT’s are electromagnetic devices designed so that as a ferromagnetic core is displaced within a transformer (consisting of three windings), a linearly varying a.c. voltage and phase shift are produced, this signal is demodulated to produce a varying d.c. output. LVDT’s are available in both linear and angu￾lar configurations. LVDT’s are available in lengths to 10 feet (3 meters), their output linearity is about 0.1%, and their maximum resolution is 1 microinch (25 µm). The accuracy of a given LVDT is commonly limited to 0.01% of total travel. An LVDT may be used directly as a deflectometer with its core contacting the specimen; it can be used with a linkage; or it can be incorporated into a contacting extensometer High temperature LVDT’s may be usable up to the Curie Temperature of the core material, but are gener￾ally used with extensions or linkages to avoid exposing them to hostile environments. LVDT’s must be calibrated at the temperature to which they will be exposed in use. 6.4.4.3 Contacting extensometers Contacting extensometers and compressometers are devices that are used to determine the relative displacements of two points on a specimen. The contact extensometer must be clamped to the specimen surface in such a way that the contact points cannot slip, and that the extensometer does not affect the test. Extensometers are relatively complex devices which rely on integral strain gages or LVDT’s to con￾vert the relative displacements of their attachment points into linearly related outputs. Extensometers are available in a range of fixed gage lengths from 0.500 to 2.00 in. (12 - 50 mm), their output linearity is 0.1%, and they can resolve displacement to 1 microinch (25 µm). This resolution does not imply accu￾racy or calibration. A well-made contact extensometer is accurate to 0.01% of full scale, and can meas￾ure strain up to 1.00 (100%). Repeatability of contacting extensometers is dependent on their maintaining a constant initial gage length, therefore, when a zero stop is provided it should always be used when at￾taching the extensometer to a specimen. Contact extensometers are available which can be used at liquid nitrogen temperatures, others can safely be exposed to temperatures of 500°F (260°C) for extended periods of time. Extensions and link￾ages are available which allow remote use of extensometers on specimens exposed to temperatures up to 3000°F (1600°C). ASTM E 83 requires that extensometers be calibrated at the temperature at which they will be used. Extensometer calibration should be verified whenever the extensometer is subjected to deflection exceeding the normal range, been exposed to a hostile environment, received rough handling, and whenever the knife edges or points are replaced. 6.4.4.3.1 Contacting extensometers, applications Extensometers are chosen in preference to bondable strain gages when one or more of the following conditions exist: 1. The price of individual bonded strain gages exceeds the cost of a comparable extensometer. 2. The construction of a laminate will induce a non-uniform strain field under a bonded strain gage. 3. Strains will exceed the practical limit of bonded strain gages (0.03 or 3%). 4. The net deformation of a complex structure or assembly is required (for example a bonded or bolted joint). 5. When specimen conditioning or preconditioning will not allow proper bonding of strain gages. Extensometers are not recommended when the following circumstances apply: 1. Extensometers fitted with points or knife edges may cause premature failures in notch sensitive materials. 2. Extensometers of large inertial mass respond unpredictably to rapid changes in strain. 3. Catastrophic failure of a specimen while an extensometer is attached will result in damage to the extensometer requiring repair and recalibration or replacement

MIL-HDBK-17-1F Volume 1,Chapter 6 Lamina,Laminate,and Special Form Characterization 6.4.4.4 Bondable resistance strain gages Strain gages are structures of precisely etched metal foil or wire (usually on a polyimide film sub- strate)which are permanently bonded to a specimen surface so that the strain field of that surface is im- mediately transmitted to the gage.In use,the strain gage forms part of a Wheatstone Bridge circuit, which allows strain to be accurately measured as a function of the change in resistance of the grid.Strain gages are made from alloys(Constantan,Karma Alloy)which show relatively small changes in strain sen- sitivity(ratio of change in resistance to change in length)when they are deformed beyond their propor- tional limits(Reference 6.4.4.4). Strain gages have inherently infinite resolution(limited by the accuracy of the gage factor calibration); their ability to indicate small changes in strain accurately is limited only by the instrumentation used. Strain gages are versatile: 1. Strain gages can be applied directly to a specimen,or can be used to construct extensometers or beam bending deflectometers. 2.Several strain gages can be applied to a single specimen in different orientations to measure si- multaneous multiaxial properties. 3.Several strain gages can be applied to a single specimen in various places in similar orientations to identify stress concentrations. 6.4.4.4.1 Strain gage selection Strain gages are available in a wide range of styles.The selection of the proper strain gage is critical if accurate and repeatable results are to be obtained.Polymeric matrix composites are relatively poor thermal conductors,therefore,3502 or higher resistance gages are usually chosen in preference to 1202 gages,higher resistance gages operate at lower currents for a given strain and are less likely to produce errors due to self-heating (Reference 6.4.4.4.1(a)). Since stresses in woven composites are transmitted by the interaction of relatively large repeating units,the gage must be large enough to integrate any strain gradient associated with the weave.The grid size chosen for a composite specimen will generally be larger than that for a similar metal specimen.Grid sizes of 0.125,0.250 and 0.500 in.(3.17,6.35,and 12.7 mm)are commonly used,with specimen size limiting the size of the gage which can be used.The installation of gages very close to specimen edges is to be avoided,as edge effects are difficult to predict.Finally,gages are made to function optimally over a limited range of temperatures,and it is important that the manufacturers'recommendations be heeded regarding maximum operating temperatures of different gage styles(Reference 6.4.4.4.1(b)). 6.4.4.4.2 Surface preparation and bonding of strain gages Careful evaluation of surface preparation and bonding techniques for strain gages must be done if reliable data are to be obtained.Details of these techniques will be found in Section 6.2 and Reference 6.4.4.4.2.Extreme care should be used when abrading composites to minimize damage to the fibers of the surface laminae.It should be noted that the bonding of strain gages to thermoplastic materials is es- pecially difficult. 6.4.4.4.3 Strain gage circuits A strain gage or gages function as the variable element(s)in a resistance bridge;the Wheatstone bridge of four elements,shown in Figure 6.4.4.4.3,is the most usual.The diagram illustrates a 1/4 bridge,with a single active gage,3-wire configuration(the 3-wire configuration removes the effects of lead wire resistance from the circuit).P+and P-represent the excitation voltage for the bridge,S+and S-rep- resent the output signal.R1 and R3 are fixed resistors of identical value.When R2 and RG(the resis- tance of the strain gage)are identical,the bridge is said to be balanced,and no current flows between S+ and S-.A change in resistance of similar value and sign in adjacent elements (e.g.,R2,R3)is a null input 6-17

MIL-HDBK-17-1F Volume 1, Chapter 6 Lamina, Laminate, and Special Form Characterization 6-17 6.4.4.4 Bondable resistance strain gages Strain gages are structures of precisely etched metal foil or wire (usually on a polyimide film sub￾strate) which are permanently bonded to a specimen surface so that the strain field of that surface is im￾mediately transmitted to the gage. In use, the strain gage forms part of a Wheatstone Bridge circuit, which allows strain to be accurately measured as a function of the change in resistance of the grid. Strain gages are made from alloys (Constantan, Karma Alloy) which show relatively small changes in strain sen￾sitivity (ratio of change in resistance to change in length) when they are deformed beyond their propor￾tional limits (Reference 6.4.4.4). Strain gages have inherently infinite resolution (limited by the accuracy of the gage factor calibration); their ability to indicate small changes in strain accurately is limited only by the instrumentation used. Strain gages are versatile: 1. Strain gages can be applied directly to a specimen, or can be used to construct extensometers or beam bending deflectometers. 2. Several strain gages can be applied to a single specimen in different orientations to measure si￾multaneous multiaxial properties. 3. Several strain gages can be applied to a single specimen in various places in similar orientations to identify stress concentrations. 6.4.4.4.1 Strain gage selection Strain gages are available in a wide range of styles. The selection of the proper strain gage is critical if accurate and repeatable results are to be obtained. Polymeric matrix composites are relatively poor thermal conductors, therefore, 350Ω or higher resistance gages are usually chosen in preference to 120Ω gages, higher resistance gages operate at lower currents for a given strain and are less likely to produce errors due to self-heating (Reference 6.4.4.4.1(a)). Since stresses in woven composites are transmitted by the interaction of relatively large repeating units, the gage must be large enough to integrate any strain gradient associated with the weave. The grid size chosen for a composite specimen will generally be larger than that for a similar metal specimen. Grid sizes of 0.125, 0.250 and 0.500 in. (3.17, 6.35, and 12.7 mm) are commonly used, with specimen size limiting the size of the gage which can be used. The installation of gages very close to specimen edges is to be avoided, as edge effects are difficult to predict. Finally, gages are made to function optimally over a limited range of temperatures, and it is important that the manufacturers' recommendations be heeded regarding maximum operating temperatures of different gage styles (Reference 6.4.4.4.1(b)). 6.4.4.4.2 Surface preparation and bonding of strain gages Careful evaluation of surface preparation and bonding techniques for strain gages must be done if reliable data are to be obtained. Details of these techniques will be found in Section 6.2 and Reference 6.4.4.4.2. Extreme care should be used when abrading composites to minimize damage to the fibers of the surface laminae. It should be noted that the bonding of strain gages to thermoplastic materials is es￾pecially difficult. 6.4.4.4.3 Strain gage circuits A strain gage or gages function as the variable element(s) in a resistance bridge; the Wheatstone bridge of four elements, shown in Figure 6.4.4.4.3, is the most usual. The diagram illustrates a 1/4 bridge, with a single active gage, 3-wire configuration (the 3-wire configuration removes the effects of lead wire resistance from the circuit). P+ and P- represent the excitation voltage for the bridge, S+ and S- rep￾resent the output signal. R1 and R3 are fixed resistors of identical value. When R2 and RG (the resis￾tance of the strain gage) are identical, the bridge is said to be balanced, and no current flows between S+ and S-. A change in resistance of similar value and sign in adjacent elements (e.g., R2, R3) is a null input

MIL-HDBK-17-1F Volume 1,Chapter 6 Lamina,Laminate,and Special Form Characterization to the bridge.A change in resistance of similar value and sign in opposite elements(for example,R1,R3) is summed in magnitude.These results are useful in strain measurement:in the first case a gage can be applied to a spare piece of specimen material,and if this second gage is positioned at R1 in the circuit (therefore adjacent to RG)and then exposed to the test conditions,it will compensate for the thermal re- sponses of the specimen and the active gage.In the second case,referred to as a half bridge,a speci- men has two active gages both placed within a constant strain field,the second gage is placed at R2(op- posite to RG),then the gage outputs will be summed,and dividing by 2 will give the average strain,with a 2-fold increase in resolution.Contact extensometers are often designed using four gages in a "Full- Bridge"configuration which makes good use of the bridge by effectively summing all elements(adjacent gages are positioned so as to be exposed to strain fields of equal value and opposite sign).In all cases where passive bridge elements exist they are referred to as"Bridge Completion"and are a necessary part of the instrumentation associated with strain gages. S+ S-P+ RG M R31 R1 R2 WHEATSTONE BRIDGE CIRCUIT FIGURE 6.4.4.4.3 Wheatstone bridge. 6.4.4.4.4 Strain gage instrumentation The instrumentation used with strain gages(and extensometers utilizing strain gages as their active elements)is usually of the constant voltage type.The bridge circuit is provided with a stabilized d.c.exci- tation voltage between 2 and 10 Volts,and the output is on the scale of microvolts.High gain instrumen- tation amplifiers with low drift and excellent stability are used to scale the outputs up to Volt levels. The combination of excitation and amplification in a single instrument is called a conditioner.Condi- tioners are available with fixed or variable excitation voltages.A variable excitation conditioner can be used to achieve high resolutions at high excitation voltages(high signal to noise ratio),or extended strain ranges at low voltages.It is a good idea to avoid using excitation voltages greater than 10 Volts for 3500 gages on polymer matrix composites,which do not dissipate heat efficiently,to avoid"self-heating"of the gage (Reference 6.4.4.4.4).Conditioners with fixed excitation voltages usually offer variable amplifier gains to scale outputs.There is less possibility of overheating the gage with a fixed voltage conditioner. 6-18

MIL-HDBK-17-1F Volume 1, Chapter 6 Lamina, Laminate, and Special Form Characterization 6-18 to the bridge. A change in resistance of similar value and sign in opposite elements (for example, R1, R3) is summed in magnitude. These results are useful in strain measurement: in the first case a gage can be applied to a spare piece of specimen material, and if this second gage is positioned at R1 in the circuit (therefore adjacent to RG) and then exposed to the test conditions, it will compensate for the thermal re￾sponses of the specimen and the active gage. In the second case, referred to as a half bridge, a speci￾men has two active gages both placed within a constant strain field, the second gage is placed at R2 (op￾posite to RG), then the gage outputs will be summed, and dividing by 2 will give the average strain, with a 2-fold increase in resolution. Contact extensometers are often designed using four gages in a “Full￾Bridge” configuration which makes good use of the bridge by effectively summing all elements (adjacent gages are positioned so as to be exposed to strain fields of equal value and opposite sign). In all cases where passive bridge elements exist they are referred to as “Bridge Completion” and are a necessary part of the instrumentation associated with strain gages. FIGURE 6.4.4.4.3 Wheatstone bridge. 6.4.4.4.4 Strain gage instrumentation The instrumentation used with strain gages (and extensometers utilizing strain gages as their active elements) is usually of the constant voltage type. The bridge circuit is provided with a stabilized d.c. exci￾tation voltage between 2 and 10 Volts, and the output is on the scale of microvolts. High gain instrumen￾tation amplifiers with low drift and excellent stability are used to scale the outputs up to Volt levels. The combination of excitation and amplification in a single instrument is called a conditioner. Condi￾tioners are available with fixed or variable excitation voltages. A variable excitation conditioner can be used to achieve high resolutions at high excitation voltages (high signal to noise ratio), or extended strain ranges at low voltages. It is a good idea to avoid using excitation voltages greater than 10 Volts for 350Ω gages on polymer matrix composites, which do not dissipate heat efficiently, to avoid “self-heating” of the gage (Reference 6.4.4.4.4). Conditioners with fixed excitation voltages usually offer variable amplifier gains to scale outputs. There is less possibility of overheating the gage with a fixed voltage conditioner

MIL-HDBK-17-1F Volume 1,Chapter 6 Lamina,Laminate,and Special Form Characterization 6.4.4.4.5 Strain gage instrumentation calibration Strain conditioner linearity is verified by the use of strain simulation.With 3500 taken as the balance point or zero,strain values can be simulated by using a high accuracy decade resistance box with ranges from 0.012 to 1002 in place of the active gage,and using the following equation to simulate strain val- ues: 2=0.0007esim+350 6.4.4.4.5 where decade resistance box setting to simulate target strain (ohms) Esim target strain to be simulated(microstrain) When fixed excitation conditioners have been verified in this way and found acceptable,no further calibra- tion is necessary before testing.The output of the conditioner is simply multiplied by 2/K,where K is the gage factor reported by the gage manufacturer. When conditioners offer variable excitation,shunt calibration is required. 6.4.4.4.5.1 Shunt calibration (for 1/4 bridge) When a variable excitation conditioner is used,the excitation voltage is generally chosen to scale the conditioner output(span)to the expected maximum strain level expected in the test.This provides the maximum resolution over the range of the test.With an active gage in the circuit (usually an actual specimen with no load applied),the conditioner output is zeroed.A precision resistor is placed in the cir- cuit parallel with a bridge resistor.The value of the resistor is chosen so that when it is wired parallel to the gage,the combined resistance is exactly that necessary to simulate a known strain,called the shunt value.The excitation voltage is then adjusted so that the conditioner readout shows a value equal to 2/K multiplied by the shunt value.After instrument scaling,the indicated strain will be correct at the magni- tude of the calibration strain,but slightly in error at other strain levels.The corrected strain at any differ- ent strain level can be calculated from Reference 6.4.4.4.1(b): e=2e/(2+K(es-e) 6.4.4.4.5.1 where corrected strain (microstrain) indicated strain(microstrain) es shunt cal value(microstrain) K gage factor of strain gage The topic of shunt calibration of Wheatstone bridges is treated simply here,but is actually a matter of great complexity,and it is recommended that the serious researcher carefully study Reference 6.4.4.4.5.1. 6.4.4.5 Other methods A number of extensometric methods exist which see limited use in the determination of polymer ma- trix composite properties due either to unreliability or difficulty of use.However,under appropriate cir- cumstances these techniques yield valuable data which could otherwise not be obtained,therefore,they are described here in limited detail. 6.4.4.5.1 Optical methods of extensometry A number of methods of strain measurement based on optical phenomena exist:photoelasticity, Moire interferometry,and laser extensometry.Photoelastic methods and Moire may be used to verify the results of finite element calculations,and to investigate stress distributions on test specimens or struc- 6-19

MIL-HDBK-17-1F Volume 1, Chapter 6 Lamina, Laminate, and Special Form Characterization 6-19 6.4.4.4.5 Strain gage instrumentation calibration Strain conditioner linearity is verified by the use of strain simulation. With 350Ω taken as the balance point or zero, strain values can be simulated by using a high accuracy decade resistance box with ranges from 0.01Ω to 100Ω in place of the active gage, and using the following equation to simulate strain val￾ues: Ω = 0.0007 + 350 εsim 6.4.4.4.5 where Ω = decade resistance box setting to simulate target strain (ohms) εsim = target strain to be simulated (microstrain) When fixed excitation conditioners have been verified in this way and found acceptable, no further calibra￾tion is necessary before testing. The output of the conditioner is simply multiplied by 2/K, where K is the gage factor reported by the gage manufacturer. When conditioners offer variable excitation, shunt calibration is required. 6.4.4.4.5.1 Shunt calibration (for 1/4 bridge) When a variable excitation conditioner is used, the excitation voltage is generally chosen to scale the conditioner output (span) to the expected maximum strain level expected in the test. This provides the maximum resolution over the range of the test. With an active gage in the circuit (usually an actual specimen with no load applied), the conditioner output is zeroed. A precision resistor is placed in the cir￾cuit parallel with a bridge resistor. The value of the resistor is chosen so that when it is wired parallel to the gage, the combined resistance is exactly that necessary to simulate a known strain, called the shunt value. The excitation voltage is then adjusted so that the conditioner readout shows a value equal to 2/K multiplied by the shunt value. After instrument scaling, the indicated strain will be correct at the magni￾tude of the calibration strain, but slightly in error at other strain levels. The corrected strain at any differ￾ent strain level can be calculated from Reference 6.4.4.4.1(b) : ε = 2 / (2 + K( - )) ε εε i si 6.4.4.4.5.1 where ε = corrected strain (microstrain) εi = indicated strain (microstrain) εs = shunt cal value (microstrain) K = gage factor of strain gage The topic of shunt calibration of Wheatstone bridges is treated simply here, but is actually a matter of great complexity, and it is recommended that the serious researcher carefully study Reference 6.4.4.4.5.1. 6.4.4.5 Other methods A number of extensometric methods exist which see limited use in the determination of polymer ma￾trix composite properties due either to unreliability or difficulty of use. However, under appropriate cir￾cumstances these techniques yield valuable data which could otherwise not be obtained, therefore, they are described here in limited detail. 6.4.4.5.1 Optical methods of extensometry A number of methods of strain measurement based on optical phenomena exist: photoelasticity, Moiré interferometry, and laser extensometry. Photoelastic methods and Moiré may be used to verify the results of finite element calculations, and to investigate stress distributions on test specimens or struc-

MIL-HDBK-17-1F Volume 1,Chapter 6 Lamina,Laminate,and Special Form Characterization tures.The application of these techniques to the design of test specimens and fixturing is an important stage in optimization of test geometry. The non-contact nature of laser extensometry makes it particularly attractive in circumstances where strain gages would be unreliable-at high temperatures,on small radii,and on rough surfaces. 6.4.4.5.2 Capacitative extensometers Contact extensometers are available which utilize the capacitance of an air gap between two probes fixed to the specimen surface to determine strain.These probes are accurate only for very small gage lengths,and cannot be used to record strain to failure as they are easily destroyed.They are used to de- termine modulus of materials at very high temperatures(>1000F or 500C).Capacitative extensometers can be difficult to calibrate and require complicated conditioning instrumentation.They cannot be cali- brated better than ASTM E 83 Class B-2. 6.4.4.6 Special considerations for textile composites The inhomogenity of textile composites requires that strains and displacements are measured over sufficient gage lengths to be representative of bulk(average)specimen response.Results of a study on composites made of 2D triaxial braids,3D weaves,and stitched uniwoven laminates to determine the ef- fect of strain gage size on strain measurements are given in Reference 6.4.4.6(a). In general,strain gages should be longer than the unit cell length of the textile and gage width should be no less than half the length.For specific standards in selecting strain gages use Reference 6.4.4.6(b). The gage lengths of extensometers should also be larger than the unit cell size to obtain an average or macroscopic displacement.These recommendations for minimum gage length should apply for thermal loads as well as mechanical loads. Although not addressed in References 6.4.4.6(a)and (b),several gages might be arranged end to end to avoid more costly special order gages for unit cells longer than.5 in.(12.7 mm). 6.4.5 Temperature measurement devices 6.4.5.1 Introduction Many of the properties that characterize a composite lamina or laminate are temperature dependent. Thus,temperature is one of the variables that must be measured to fully characterize a material.Many tools and techniques exist to measure temperature,but not all will provide the desired results or function in the required environment for the duration of the test.Temperature measurement devices can be divided into two categories:contact and noncontact.Five types of contact temperature measuring devices are commonly encountered:thermocouples,resistive temperature devices(metallic RTDs and thermistors). bimetallic devices,liquid expansion devices,and change of state devices.A noncontact temperature measuring device commonly used is an infrared detector. 6.4.5.2 Thermocouples A thermocouple consists essentially of two strips or wires made of different metal alloys and joined at one end.Referring to Figure 6.4.5.2.,changes in the temperature T1 at that junction induce a change in electromotive force (emf)Vab between the other ends.As temperature goes up,this output emf of the thermocouple rises,though not necessarily linearly.The open-end emf is a function of not only the closed-end temperature T1 (i.e.,the temperature at the point of measurement),but also the temperature T2 at the open end.Only by holding T2 at a standard temperature can the measured emf be considered a direct function of the change in T1.The industrially accepted standard for T2 is 32F(0C);therefore, most tables and charts assume that 12 is at that level.In industrial instrumentation,the difference be- tween actual temperature at T2 and 32F(0C)is usually corrected for electronically,within the instru- mentation.This emf adjustment is referred to as the cold-junction,or CJ,correction.Temperature 6-20

MIL-HDBK-17-1F Volume 1, Chapter 6 Lamina, Laminate, and Special Form Characterization 6-20 tures. The application of these techniques to the design of test specimens and fixturing is an important stage in optimization of test geometry. The non-contact nature of laser extensometry makes it particularly attractive in circumstances where strain gages would be unreliable - at high temperatures, on small radii, and on rough surfaces. 6.4.4.5.2 Capacitative extensometers Contact extensometers are available which utilize the capacitance of an air gap between two probes fixed to the specimen surface to determine strain. These probes are accurate only for very small gage lengths, and cannot be used to record strain to failure as they are easily destroyed. They are used to de￾termine modulus of materials at very high temperatures (>1000°F or 500°C). Capacitative extensometers can be difficult to calibrate and require complicated conditioning instrumentation. They cannot be cali￾brated better than ASTM E 83 Class B-2. 6.4.4.6 Special considerations for textile composites The inhomogenity of textile composites requires that strains and displacements are measured over sufficient gage lengths to be representative of bulk (average) specimen response. Results of a study on composites made of 2D triaxial braids, 3D weaves, and stitched uniwoven laminates to determine the ef￾fect of strain gage size on strain measurements are given in Reference 6.4.4.6(a). In general, strain gages should be longer than the unit cell length of the textile and gage width should be no less than half the length. For specific standards in selecting strain gages use Reference 6.4.4.6(b). The gage lengths of extensometers should also be larger than the unit cell size to obtain an average or macroscopic displacement. These recommendations for minimum gage length should apply for thermal loads as well as mechanical loads. Although not addressed in References 6.4.4.6(a) and (b), several gages might be arranged end to end to avoid more costly special order gages for unit cells longer than .5 in. (12.7 mm). 6.4.5 Temperature measurement devices 6.4.5.1 Introduction Many of the properties that characterize a composite lamina or laminate are temperature dependent. Thus, temperature is one of the variables that must be measured to fully characterize a material. Many tools and techniques exist to measure temperature, but not all will provide the desired results or function in the required environment for the duration of the test. Temperature measurement devices can be divided into two categories: contact and noncontact. Five types of contact temperature measuring devices are commonly encountered: thermocouples, resistive temperature devices (metallic RTDs and thermistors), bimetallic devices, liquid expansion devices, and change of state devices. A noncontact temperature measuring device commonly used is an infrared detector. 6.4.5.2 Thermocouples A thermocouple consists essentially of two strips or wires made of different metal alloys and joined at one end. Referring to Figure 6.4.5.2., changes in the temperature T1 at that junction induce a change in electromotive force (emf) Vab between the other ends. As temperature goes up, this output emf of the thermocouple rises, though not necessarily linearly. The open-end emf is a function of not only the closed-end temperature T1 (i.e., the temperature at the point of measurement), but also the temperature T2 at the open end. Only by holding T2 at a standard temperature can the measured emf be considered a direct function of the change in T1. The industrially accepted standard for T2 is 32°F (0°C); therefore, most tables and charts assume that T2 is at that level. In industrial instrumentation, the difference be￾tween actual temperature at T2 and 32°F (0°C) is usually corrected for electronically, within the instru￾mentation. This emf adjustment is referred to as the cold-junction, or CJ, correction. Temperature