《纺织复合材料》课程参考文献（Composite Materials Handbook，Volume 1）CHAPTER 7 STRUCTURAL ELEMENT CHARACTERIZATION

MIL-HDBK-17-1F Volume 1,Chapter 7 Structural Element Characterization CHAPTER 7 STRUCTURAL ELEMENT CHARACTERIZATION 7.1 INTRODUCTION The material in this chapter focuses on test methods and matrices for experimental characterization of composite structures at a laminate/element level of complexity of the building block approach de- scribed in Volume 3,Chapter 4.The test elements,discussed here,provide data on notched laminates, bolted and bonded joints,and damage tolerance behavior that is needed for analysis and design of com- posite structures.General discussion on analysis and design of bolted and bonded joints can be found in Volume 3,Chapter 6,while damage tolerance is covered in Volume 3,Chapter 7. Any joint in a composite structure is a potential failure site.Without proper design a joint can act as a failure initiation point,which can lead to a loss in structural strength and eventual failure of the compo- nent.Two types of joints are in common use:(1)mechanically-fastened joints and(2)adhesively-bonded joints.These guidelines define test types,laminates,environments,and replication that are needed for structurally sound joint design. For mechanically bolted joints,tests are described that characterize the joint for various failure modes:notched tension/compression,bearing,bearing/by-pass,shear-out,and fastener pull-thru.The tests are drawn from ASTM standards when available.Otherwise common usage tests are recom- mended.In addition,suggested test matrices are provided that characterize the joint properties for the different variables that affect those properties.The suggested matrices should be considered as the least amount of testing required to obtain design properties.The test matrices are derived from the generic laminate/structural element test matrices in Section 2.3.5.and are included here for completeness.A de- tailed analysis of the stress distribution around a fastener hole is not presented here but is available in Volume 3,Section 6.3. For bonded joints,two types of tests are described.One type determines adhesive properties that are needed in design.These tests provide adhesive stiffness and strength properties needed for analysis and design methods of Volume 3,Section 6.2.The second type is used to verify specific designs.Examples of such tests are shown. The tests in the damage tolerance section are of two types.One type characterizes the damage re- sistance of a given laminate and the second the damage tolerance of that laminate.The Compression after Impact(CAl)test,an example of the latter type,is used widely in the aerospace industry to gauge damage tolerance potential of composite materials. 7.2 SPECIMEN PREPARATION 7.2.1 Introduction The general topic of specimen preparation has been described adequately in Section 6.2 of this vol- ume and in ASTM D 5687 for standard flat specimens.This section provides specific guidance for ele- ments that represent mechanically fastened and bonded joints.Additionally,for tests where an ASTM standard exists the standard contains specific specimen preparation guidelines.Specimens for damage tolerance tests are flat plates which require no special specimen preparation procedures other than those in Section 6.2 7.2.2 Mechanically fastened joint tests The main concerns with mechanically fastened joint specimens are hole drilling and fastener installa- tion.Holes should be drilled undersized and reamed to final dimensions.Drill back-up plates should be used to prevent delaminations at the drill exit side.Hole diameters should be verified as to their confor- mity to the specimen drawing.Specimen hole preparation methods should be recorded. 7-1

MIL-HDBK-17-1F Volume 1, Chapter 7 Structural Element Characterization 7-1 CHAPTER 7 STRUCTURAL ELEMENT CHARACTERIZATION 7.1 INTRODUCTION The material in this chapter focuses on test methods and matrices for experimental characterization of composite structures at a laminate/element level of complexity of the building block approach de￾scribed in Volume 3, Chapter 4. The test elements, discussed here, provide data on notched laminates, bolted and bonded joints, and damage tolerance behavior that is needed for analysis and design of com￾posite structures. General discussion on analysis and design of bolted and bonded joints can be found in Volume 3, Chapter 6, while damage tolerance is covered in Volume 3, Chapter 7. Any joint in a composite structure is a potential failure site. Without proper design a joint can act as a failure initiation point, which can lead to a loss in structural strength and eventual failure of the compo￾nent. Two types of joints are in common use: (1) mechanically-fastened joints and (2) adhesively-bonded joints. These guidelines define test types, laminates, environments, and replication that are needed for structurally sound joint design. For mechanically bolted joints, tests are described that characterize the joint for various failure modes: notched tension/compression, bearing, bearing/by-pass, shear-out, and fastener pull-thru. The tests are drawn from ASTM standards when available. Otherwise common usage tests are recom￾mended. In addition, suggested test matrices are provided that characterize the joint properties for the different variables that affect those properties. The suggested matrices should be considered as the least amount of testing required to obtain design properties. The test matrices are derived from the generic laminate/structural element test matrices in Section 2.3.5, and are included here for completeness. A de￾tailed analysis of the stress distribution around a fastener hole is not presented here but is available in Volume 3, Section 6.3. For bonded joints, two types of tests are described. One type determines adhesive properties that are needed in design. These tests provide adhesive stiffness and strength properties needed for analysis and design methods of Volume 3, Section 6.2. The second type is used to verify specific designs. Examples of such tests are shown. The tests in the damage tolerance section are of two types. One type characterizes the damage re￾sistance of a given laminate and the second the damage tolerance of that laminate. The Compression after Impact (CAI) test, an example of the latter type, is used widely in the aerospace industry to gauge damage tolerance potential of composite materials. 7.2 SPECIMEN PREPARATION 7.2.1 Introduction The general topic of specimen preparation has been described adequately in Section 6.2 of this vol￾ume and in ASTM D 5687 for standard flat specimens. This section provides specific guidance for ele￾ments that represent mechanically fastened and bonded joints. Additionally, for tests where an ASTM standard exists the standard contains specific specimen preparation guidelines. Specimens for damage tolerance tests are flat plates which require no special specimen preparation procedures other than those in Section 6.2. 7.2.2 Mechanically fastened joint tests The main concerns with mechanically fastened joint specimens are hole drilling and fastener installa￾tion. Holes should be drilled undersized and reamed to final dimensions. Drill back-up plates should be used to prevent delaminations at the drill exit side. Hole diameters should be verified as to their confor￾mity to the specimen drawing. Specimen hole preparation methods should be recorded

MIL-HDBK-17-1F Volume 1,Chapter 7 Structural Element Characterization Proper fastener installation procedures are critical for determination of mechanical joint properties. These are specific to each type of bolt tested and are provided either by the bolt manufacturer or part fab- ricator.Unless finger tight bolt torque is specified,test specimens containing fasteners must be installed per company specification for the data to be meaningful for a given application.Correct grip sizes must also be selected based on the thickness of the mating parts.All bolt installations must be inspected for proper seating and fit. 7.2.3 Bonded joint tests Test specimens for bonded joint characterization must be fabricated using processing specifications for bonding surface preparation and cure.This requirement is reiterated in the ASTM standards for bonded joint tests described in this chapter(7.6).For the bonded joint data to have any practical use,the specimens must be fabricated to strict processing controls which are the same as for fabrication of actual parts 7.3 CONDITIONING AND ENVIRONMENTAL EXPOSURE 7.3.1 Introduction The objective of testing environmentally conditioned specimens is to quantify property changes caused by exposure to humidity,liquid water,or other fluids(gaseous or liquid)under controlled(or at least defined)conditions.In general,the considerations and procedures presented in Section 6.3 of this volume apply to structural elements as well as to the simpler laminate specimens.However,there are some additional issues associated with environmental exposure of structural elements.These special considerations are discussed in the following sections,and cover general specimen preparation(strain gaging,notched laminates,and mechanically fastened joints),bonded joints,damage characterization, and sandwich structure.For the purposes of these discussions,the term "moisture"refers to any ab- sorbed medium(water vapor.liquid water.or other fluid). 7.3.2 General specimen preparation 7.3.2.1 Strain gaging Structural element tests may involve the use of more strain gages than for small specimens.These gages are frequently applied after exposure to the conditioning medium to prevent the gages from inter- fering with the conditioning process or to preclude environmental degradation of the gage adhesive lead- ing to premature gage failure.When multiple gages are applied,the test articles are likely to be at ambi- ent conditions for a considerable period of time during the gage bonding process,increasing the risk of significant moisture loss.To minimize this risk,gages should be applied as quickly as possible,and arti- cles should be returned to the conditioning environment or suitable storage container as soon as gaging is complete.If all gages cannot be applied in a single,short session,articles should be returned to the environment or storage between gaging sessions.It is also possible to bag all or portions of the article together with moist towels.Small areas can then be exposed to allow local gaging while minimizing mois- ture loss of the overall article. In instances where an elevated temperature cure is required for the gage bonding adhesive,it may be possible to accomplish the cure by returning the specimens to the elevated temperature conditioning en- vironment rather than curing in dry air and risking moisture loss.However,it must be determined if the conditioning environment will have a detrimental effect on the cure reaction. In some cases it may be necessary to bond gages prior to exposure(for example,if a conditioning fluid like oil would render the specimen surface unsuitable for adhesive bonding).Judgment must be used in determining whether to condition before or after gage bonding.Strain gage and/or gage adhesive manufacturers can often provide valuable advice in making this decision. 7-2

MIL-HDBK-17-1F Volume 1, Chapter 7 Structural Element Characterization 7-2 Proper fastener installation procedures are critical for determination of mechanical joint properties. These are specific to each type of bolt tested and are provided either by the bolt manufacturer or part fab￾ricator. Unless finger tight bolt torque is specified, test specimens containing fasteners must be installed per company specification for the data to be meaningful for a given application. Correct grip sizes must also be selected based on the thickness of the mating parts. All bolt installations must be inspected for proper seating and fit. 7.2.3 Bonded joint tests Test specimens for bonded joint characterization must be fabricated using processing specifications for bonding surface preparation and cure. This requirement is reiterated in the ASTM standards for bonded joint tests described in this chapter (7.6). For the bonded joint data to have any practical use, the specimens must be fabricated to strict processing controls which are the same as for fabrication of actual parts. 7.3 CONDITIONING AND ENVIRONMENTAL EXPOSURE 7.3.1 Introduction The objective of testing environmentally conditioned specimens is to quantify property changes caused by exposure to humidity, liquid water, or other fluids (gaseous or liquid) under controlled (or at least defined) conditions. In general, the considerations and procedures presented in Section 6.3 of this volume apply to structural elements as well as to the simpler laminate specimens. However, there are some additional issues associated with environmental exposure of structural elements. These special considerations are discussed in the following sections, and cover general specimen preparation (strain gaging, notched laminates, and mechanically fastened joints), bonded joints, damage characterization, and sandwich structure. For the purposes of these discussions, the term “moisture” refers to any ab￾sorbed medium (water vapor, liquid water, or other fluid). 7.3.2 General specimen preparation 7.3.2.1 Strain gaging Structural element tests may involve the use of more strain gages than for small specimens. These gages are frequently applied after exposure to the conditioning medium to prevent the gages from inter￾fering with the conditioning process or to preclude environmental degradation of the gage adhesive lead￾ing to premature gage failure. When multiple gages are applied, the test articles are likely to be at ambi￾ent conditions for a considerable period of time during the gage bonding process, increasing the risk of significant moisture loss. To minimize this risk, gages should be applied as quickly as possible, and arti￾cles should be returned to the conditioning environment or suitable storage container as soon as gaging is complete. If all gages cannot be applied in a single, short session, articles should be returned to the environment or storage between gaging sessions. It is also possible to bag all or portions of the article together with moist towels. Small areas can then be exposed to allow local gaging while minimizing mois￾ture loss of the overall article. In instances where an elevated temperature cure is required for the gage bonding adhesive, it may be possible to accomplish the cure by returning the specimens to the elevated temperature conditioning en￾vironment rather than curing in dry air and risking moisture loss. However, it must be determined if the conditioning environment will have a detrimental effect on the cure reaction. In some cases it may be necessary to bond gages prior to exposure (for example, if a conditioning fluid like oil would render the specimen surface unsuitable for adhesive bonding). Judgment must be used in determining whether to condition before or after gage bonding. Strain gage and/or gage adhesive manufacturers can often provide valuable advice in making this decision

MIL-HDBK-17-1F Volume 1,Chapter 7 Structural Element Characterization 7.3.2.2 Notched laminates and mechanically fastened joint specimens Specimens with drilled holes,such as used for open hole,filled hole,and mechanically fastened joint tests,should be conditioned after drilling to avoid local dry-out around the holes due to heat generated by the drilling process. 7.3.3 Bonded joints Bonded joint configurations fall into three categories when considering environmental conditioning: articles with thin composite adherends,articles with thick composite adherends,and articles with metallic (non-absorbing)adherends.Thin adherends are defined as those capable of reaching a moisture equilib- rium condition within a reasonable period of time.Since bonding adhesives generally absorb at a faster rate than fiber-resin composites,the adhesive is usually at equilibrium when the composite adherends reach equilibrium.In such cases no modifications to the guidelines in Section 6.3 are needed. Bonded joints which employ thick composite adherends are defined as those geometries which will not reach moisture equilibrium within a time period that is practical for a test program.Indeed,some ge- ometries may require years or even decades for equilibrium to be reached throughout the bond.In such cases,the test articles must be treated in the same manner as joints with metallic(non-absorbing)adher- ends. For joints with metallic adherends (and,for practical purposes,thick composite adherends),moisture diffusion can only occur through the edges of the bond.In many cases,the bond length and width di- mensions may be such that moisture equilibrium of the adhesive cannot be achieved within a reasonable time period.Estimates of the required diffusion time can be calculated if the diffusivity of the adhesive has been previously determined from neat adhesive specimens (see Section 6.6.8 on moisture diffusiv- ity).Even if it is estimated that moisture equilibrium can be achieved within the timeframe of the test pro- gram,tracking of moisture uptake is another problem.Since the mass of non-absorbent metal adherends may be several orders of magnitude greater than the mass of the bonding adhesive,accuracy in deter- mining equilibrium from periodic weighings is poor at best.Travelers consisting of aluminum foil adher- ends bonded together with the same adhesive as the test article,and in the same bondline thickness and same bond length and width dimensions as the test article,have been used in an attempt to reduce the mass of the adherends relative to the adhesive while still limiting absorption to the bond edges.Theoreti- cally these travelers,when placed in the conditioning environment along with the test articles,offer in- creased accuracy in determining when moisture equilibrium has been reached.However,the foil and adhesive masses must be known accurately.and foil corrosion introduces another potential interference. Thus,this practice has not been widely adopted.A possible work-around for the corrosion issue is the use of stainless steel or other corrosion-resistant foil,although this has not been documented Since conditioning to equilibrium is often either impractical or inaccurate,fixed time conditioning is the only real option in many cases.Although the entire bondline does not,in general,reach a constant mois- ture content,the region near the edges of the bond will be at,or close to,the equilibrium moisture level. This is the same region where shear and peel stresses are typically highest in a bonded joint under load, and from which the failure of the test article will initiate.Therefore it can be argued that,although the en- tire bondline is not at the desired moisture level.the areas where bond failures will initiate are at the de- sired level.1000 hour exposures at 85-95%relative humidity and elevated temperature(up to 185F (85C)for 350F(177C)curing epoxies)have been used by some labs as an accelerated fixed time con- dition for relatively short overlaps.However,this approach and rationale should not be used as a general excuse for short exposure times.Since structural tests of bonded joints evaluate the joint as a system, and not just the bonding adhesive in isolation,other effects of conditioning,such as metal adherend sur- face preparation degradation,may also contribute to bond failure.Such effects should be taken into con- sideration when selecting a fixed time environmental condition. 7-3

MIL-HDBK-17-1F Volume 1, Chapter 7 Structural Element Characterization 7-3 7.3.2.2 Notched laminates and mechanically fastened joint specimens Specimens with drilled holes, such as used for open hole, filled hole, and mechanically fastened joint tests, should be conditioned after drilling to avoid local dry-out around the holes due to heat generated by the drilling process. 7.3.3 Bonded joints Bonded joint configurations fall into three categories when considering environmental conditioning: articles with thin composite adherends, articles with thick composite adherends, and articles with metallic (non-absorbing) adherends. Thin adherends are defined as those capable of reaching a moisture equilib￾rium condition within a reasonable period of time. Since bonding adhesives generally absorb at a faster rate than fiber-resin composites, the adhesive is usually at equilibrium when the composite adherends reach equilibrium. In such cases no modifications to the guidelines in Section 6.3 are needed. Bonded joints which employ thick composite adherends are defined as those geometries which will not reach moisture equilibrium within a time period that is practical for a test program. Indeed, some ge￾ometries may require years or even decades for equilibrium to be reached throughout the bond. In such cases, the test articles must be treated in the same manner as joints with metallic (non-absorbing) adher￾ends. For joints with metallic adherends (and, for practical purposes, thick composite adherends), moisture diffusion can only occur through the edges of the bond. In many cases, the bond length and width di￾mensions may be such that moisture equilibrium of the adhesive cannot be achieved within a reasonable time period. Estimates of the required diffusion time can be calculated if the diffusivity of the adhesive has been previously determined from neat adhesive specimens (see Section 6.6.8 on moisture diffusiv￾ity). Even if it is estimated that moisture equilibrium can be achieved within the timeframe of the test pro￾gram, tracking of moisture uptake is another problem. Since the mass of non-absorbent metal adherends may be several orders of magnitude greater than the mass of the bonding adhesive, accuracy in deter￾mining equilibrium from periodic weighings is poor at best. Travelers consisting of aluminum foil adher￾ends bonded together with the same adhesive as the test article, and in the same bondline thickness and same bond length and width dimensions as the test article, have been used in an attempt to reduce the mass of the adherends relative to the adhesive while still limiting absorption to the bond edges. Theoreti￾cally these travelers, when placed in the conditioning environment along with the test articles, offer in￾creased accuracy in determining when moisture equilibrium has been reached. However, the foil and adhesive masses must be known accurately, and foil corrosion introduces another potential interference. Thus, this practice has not been widely adopted. A possible work-around for the corrosion issue is the use of stainless steel or other corrosion-resistant foil, although this has not been documented. Since conditioning to equilibrium is often either impractical or inaccurate, fixed time conditioning is the only real option in many cases. Although the entire bondline does not, in general, reach a constant mois￾ture content, the region near the edges of the bond will be at, or close to, the equilibrium moisture level. This is the same region where shear and peel stresses are typically highest in a bonded joint under load, and from which the failure of the test article will initiate. Therefore it can be argued that, although the en￾tire bondline is not at the desired moisture level, the areas where bond failures will initiate are at the de￾sired level. 1000 hour exposures at 85-95% relative humidity and elevated temperature (up to 185°F (85°C) for 350°F (177°C) curing epoxies) have been used by some labs as an accelerated fixed time con￾dition for relatively short overlaps. However, this approach and rationale should not be used as a general excuse for short exposure times. Since structural tests of bonded joints evaluate the joint as a system, and not just the bonding adhesive in isolation, other effects of conditioning, such as metal adherend sur￾face preparation degradation, may also contribute to bond failure. Such effects should be taken into con￾sideration when selecting a fixed time environmental condition

MIL-HDBK-17-1F Volume 1,Chapter 7 Structural Element Characterization 7.3.4 Damage characterization specimens For testing of post-damage specimens(such as compression after impact).a different result may be obtained depending on whether conditioning was performed prior to or subsequent to the damage event. This may be due to several effects: 1.A moisture conditioned panel may have a different compliance and/or matrix hardness compared to the same panel prior to conditioning.This difference in compliance and/or hardness may result in different types and/or levels of damage for the same test parameters and energy.For example, the delamination area may be less for the conditioned panel due to increased compliance, whereas the front surface dent depth might be higher due to matrix softness. 2.A panel which is conditioned after the damage event might absorb moisture in a non-Fickian manner.That is,in addition to Fickian absorption at the molecular level,liquid water (or other fluid)may start to accumulate in matrix cracks and delaminations.This phenomenon could inter- fere with weight gain measurements,as these measurements may not accurately represent mois- ture absorbed by the matrix polymer.Consequently,this will affect the accuracy of moisture equi- librium and moisture content determinations.Non-damaged travelers are recommended in this case. While there may be valid reasons within a design development or qualification program for condition- ing either before or after impact,it is important to keep these effects in mind and to document the order in which impacting,conditioning,and testing were performed. 7.3.5 Sandwich Structure Conditioning of sandwich structures requires consideration of several issues,depending upon the specific materials of construction and the failure mode under test.Table 7.3.5 shows 12 common combi- nations of materials and failure modes. TABLE 7.3.5 Sandwich materials and failure modes. Non-perforated Metallic Composite Face Sheets Face Sheets Metallic Core Organic Core Metallic Core Organic Core Face Sheet Failure (Tension Compression) 2 3 4 Core Failure (Tens.Comp./Shear) 5 6 7 8 Adhesive Bond Failure (Tension /Shear) 9 10 11 12 *Note:Table entries refer to numbered notes which follow If the core is metallic(aluminum honeycomb,for example)(as in Combinations 1,3,5,7,9,and 11 in Table 7.3.5),then only the environmental condition of the face sheets and bonding adhesive applies.If the core contains organic constituents(such as in polyamide/phenolic,glass/phenolic,or foam cores,as in Combinations 2,4.6,8,10,and 12),then the condition of the core material may be of interest,unless core failure is not an expected mode.The following lists each of the 12 combinations in Table 7.3.5,and suggests specific considerations and approaches relative to environmental conditioning. 7-4

MIL-HDBK-17-1F Volume 1, Chapter 7 Structural Element Characterization 7-4 7.3.4 Damage characterization specimens For testing of post-damage specimens (such as compression after impact), a different result may be obtained depending on whether conditioning was performed prior to or subsequent to the damage event. This may be due to several effects: 1. A moisture conditioned panel may have a different compliance and/or matrix hardness compared to the same panel prior to conditioning. This difference in compliance and/or hardness may result in different types and/or levels of damage for the same test parameters and energy. For example, the delamination area may be less for the conditioned panel due to increased compliance, whereas the front surface dent depth might be higher due to matrix softness. 2. A panel which is conditioned after the damage event might absorb moisture in a non-Fickian manner. That is, in addition to Fickian absorption at the molecular level, liquid water (or other fluid) may start to accumulate in matrix cracks and delaminations. This phenomenon could inter￾fere with weight gain measurements, as these measurements may not accurately represent mois￾ture absorbed by the matrix polymer. Consequently, this will affect the accuracy of moisture equi￾librium and moisture content determinations. Non-damaged travelers are recommended in this case. While there may be valid reasons within a design development or qualification program for condition￾ing either before or after impact, it is important to keep these effects in mind and to document the order in which impacting, conditioning, and testing were performed. 7.3.5 Sandwich Structure Conditioning of sandwich structures requires consideration of several issues, depending upon the specific materials of construction and the failure mode under test. Table 7.3.5 shows 12 common combi￾nations of materials and failure modes. TABLE 7.3.5 Sandwich materials and failure modes.* Non-perforated Metallic Face Sheets Composite Face Sheets Metallic Core Organic Core Metallic Core Organic Core Face Sheet Failure (Tension / Compression) 1 2 3 4 Core Failure (Tens. / Comp. / Shear) 5 6 7 8 Adhesive Bond Failure (Tension / Shear) 9 10 11 12 *Note: Table entries refer to numbered notes which follow If the core is metallic (aluminum honeycomb, for example) (as in Combinations 1, 3, 5, 7, 9, and 11 in Table 7.3.5), then only the environmental condition of the face sheets and bonding adhesive applies. If the core contains organic constituents (such as in polyamide/phenolic, glass/phenolic, or foam cores, as in Combinations 2, 4, 6, 8, 10, and 12), then the condition of the core material may be of interest, unless core failure is not an expected mode. The following lists each of the 12 combinations in Table 7.3.5, and suggests specific considerations and approaches relative to environmental conditioning

MIL-HDBK-17-1F Volume 1,Chapter 7 Structural Element Characterization 1.Here everything (except the adhesive)is metallic,and failure is expected in the face sheets. There is no need to condition such test articles since the face sheet strength is not usually af- fected by moisture exposure (except for corrosion effects,which are not within the scope of MIL- HDBK-17).Even if an unanticipated failure occurs in the adhesive,conditioning would have had a minimal effect on the outcome,since the adhesive is shielded by the skins (except at the edges)from the conditioning medium. 2.As in Combination 1,the metallic face sheets shield the adhesive and core from the conditioning medium.Therefore,even though the core is organic,there is no need to condition such articles, assuming that edge absorption can be ignored. 3.In this combination the face sheets are composite and are expected to fail.Therefore the mois- ture condition of the skins is of interest and conditioning to moisture equilibrium is desirable.For this configuration,it is difficult to track the test article itself(or even sandwich travelers)during conditioning because of possible liquid accumulation within the metallic cells(assuming the core is a cellular material).In such cases (where there is the assumption of one-sided exposure of the face sheets),it is convenient to prepare solid laminate travelers made of the same material and stacking sequence as the face sheets but twice the thickness.These travelers are placed in the conditioning environment along with the test article.Two sided exposure of the double thick trav- elers is equivalent to one-sided exposure of the skins on the test article.When the traveler has reached equilibrium,so have the face sheets on the test article. 4.When failure is expected in the face sheets and the core and face sheets are organic,the mois- ture content of the core is not of particular interest.Therefore,the technique of using solid lami- nate travelers twice the face sheet thickness (as discussed in 3 above)can be used.This has the added benefit of precluding accumulation of condensation in the cells of the core.Since liquid accumulation in the organic core cells is less likely than with metallic core,moisture tracking of the test article or sandwich travelers can usually be employed as an alternate method. 5.In this case core failure is anticipated.Since the core is metallic,testing of conditioned articles is not needed. 6. See Combination 2. 7.In this combination the face sheets are composite (allowing moisture to reach the interior of the sandwich),but the core(which is expected to fail)is metallic.Assuming an insignificant moisture effect on the metallic core properties,no conditioning should be needed for this configuration. 8.Both the face sheets and the core are absorptive in this combination,and the moisture level of the core (which is expected to fail)is of primary interest.This can be a difficult configuration to assess relative to moisture conditioning.The mass of the skins is frequently greater than the mass of the core;however,the equilibrium moisture content of some core materials may be greater than that of the composite skins.In addition,absorption through the edge of small sand- wich travelers may represent a significant proportion of the total moisture absorbed(which may not be the case for test articles with higher surface to edge ratios).Whether tracking is done us- ing the test article or travelers which mimic the test article geometry,accurate determination of equilibrium in the core will be compromised if face sheet absorption is dominant.One possible procedure is as follows: Determine the equilibrium moisture content of the core material alone for the environment of interest using methods discussed in Section 6.4.8(with modifications as needed). Prepare a large quantity of sandwich travelers that mimic the geometry of the test article. If the surface to edge ratio of the test article is much larger than the travelers,mask the edges of the travelers with foil tape or other suitable barrier material. Place the test article(s)and the travelers in the conditioning environment. 7-5

MIL-HDBK-17-1F Volume 1, Chapter 7 Structural Element Characterization 7-5 1. Here everything (except the adhesive) is metallic, and failure is expected in the face sheets. There is no need to condition such test articles since the face sheet strength is not usually af￾fected by moisture exposure (except for corrosion effects, which are not within the scope of MIL￾HDBK-17). Even if an unanticipated failure occurs in the adhesive, conditioning would have had a minimal effect on the outcome, since the adhesive is shielded by the skins (except at the edges) from the conditioning medium. 2. As in Combination 1, the metallic face sheets shield the adhesive and core from the conditioning medium. Therefore, even though the core is organic, there is no need to condition such articles, assuming that edge absorption can be ignored. 3. In this combination the face sheets are composite and are expected to fail. Therefore the mois￾ture condition of the skins is of interest and conditioning to moisture equilibrium is desirable. For this configuration, it is difficult to track the test article itself (or even sandwich travelers) during conditioning because of possible liquid accumulation within the metallic cells (assuming the core is a cellular material). In such cases (where there is the assumption of one-sided exposure of the face sheets), it is convenient to prepare solid laminate travelers made of the same material and stacking sequence as the face sheets but twice the thickness. These travelers are placed in the conditioning environment along with the test article. Two sided exposure of the double thick trav￾elers is equivalent to one-sided exposure of the skins on the test article. When the traveler has reached equilibrium, so have the face sheets on the test article. 4. When failure is expected in the face sheets and the core and face sheets are organic, the mois￾ture content of the core is not of particular interest. Therefore, the technique of using solid lami￾nate travelers twice the face sheet thickness (as discussed in 3 above) can be used. This has the added benefit of precluding accumulation of condensation in the cells of the core. Since liquid accumulation in the organic core cells is less likely than with metallic core, moisture tracking of the test article or sandwich travelers can usually be employed as an alternate method. 5. In this case core failure is anticipated. Since the core is metallic, testing of conditioned articles is not needed. 6. See Combination 2. 7. In this combination the face sheets are composite (allowing moisture to reach the interior of the sandwich), but the core (which is expected to fail) is metallic. Assuming an insignificant moisture effect on the metallic core properties, no conditioning should be needed for this configuration. 8. Both the face sheets and the core are absorptive in this combination, and the moisture level of the core (which is expected to fail) is of primary interest. This can be a difficult configuration to assess relative to moisture conditioning. The mass of the skins is frequently greater than the mass of the core; however, the equilibrium moisture content of some core materials may be greater than that of the composite skins. In addition, absorption through the edge of small sand￾wich travelers may represent a significant proportion of the total moisture absorbed (which may not be the case for test articles with higher surface to edge ratios). Whether tracking is done us￾ing the test article or travelers which mimic the test article geometry, accurate determination of equilibrium in the core will be compromised if face sheet absorption is dominant. One possible procedure is as follows: • Determine the equilibrium moisture content of the core material alone for the environment of interest using methods discussed in Section 6.4.8 (with modifications as needed). • Prepare a large quantity of sandwich travelers that mimic the geometry of the test article. • If the surface to edge ratio of the test article is much larger than the travelers, mask the edges of the travelers with foil tape or other suitable barrier material. • Place the test article(s) and the travelers in the conditioning environment