A carbon fiber tow has its own strength and modulus, but it is not yet a structural material ready to be installed on a rocket or satellite. Once the fibers and resin become a laminate, and that laminate is made into a component with holes and joints, its performance changes with ply orientation, resin system, cross-sectional geometry and manufacturing quality. Looking only at a fiber product sheet can reveal the material’s advantages while giving a misleading picture of the finished component’s capabilities.
This article examines continuous carbon fiber reinforced polymer composites (CFRP). It first distinguishes their mechanical properties, then explains how fibers become composites and how layups distribute performance. Finally, published fiber-grade data and thermal expansion coefficients illustrate how to approach preliminary material selection. [1]
Strength, Modulus and Stiffness Answer Different Questions
Carbon fibers generally provide high tensile strength and modulus along their axes, combined with low density. Tensile strength is the stress reached before failure in a specified test. Stress is the tensile force divided by the load-bearing cross-sectional area and is commonly expressed in MPa. Tensile modulus describes resistance to stretching in the elastic range: it is the ratio of stress to strain, where strain is the extension divided by the original length. Modulus is commonly expressed in GPa. One GPa equals 1,000 MPa. High strength does not necessarily mean high modulus, and high modulus does not necessarily mean a greater ultimate tensile load. [1][2]
To compare the potential of different materials under the same mass constraint, engineers also consider specific strength and specific modulus: strength and modulus divided by density, respectively. These are useful for preliminary comparisons, but they do not calculate the weight savings of an entire component. The finished structure also contains resin, adhesive layers, core materials, fittings and manufacturing allowances, and the loads may not act along the fibers. [4][5]
Component design also concerns stiffness, or resistance to deformation under a given load. The axial stiffness of a straight member is proportional to EA/L, where E is the modulus in the direction being considered, A is the cross-sectional area and L is the length. Even with the same material, increasing the member’s cross-sectional area or reducing its length reduces extension. In bending, cross-sectional geometry becomes particularly important. Separating the load-bearing face sheets and maintaining that separation with a lightweight core can improve bending stiffness per unit mass, but core shear and face-sheet–core debonding then become additional failure modes to check. [3][7]
For the same homogeneous plate under identical span and boundary conditions, doubling its thickness approximately doubles its mass but increases its bending stiffness eightfold. A composite laminate also requires consideration of each ply’s orientation and distance from the mid-plane. Under compression, a thin plate may buckle before the material reaches its compressive strength. The axial tensile properties of a fiber, the load-bearing capacity of a laminate and the stability of a component are three different levels of assessment. [3][7]
Transverse tension, in-plane shear, through-thickness loading and interlaminar separation test different aspects of the resin, fiber–resin interface and bonding between plies. Holes cut local fibers and alter stress distributions. Compressive performance can also be affected by fiber waviness and local instability. Tensile strength and modulus are therefore only starting points for understanding CFRP; they cannot replace data on compression, shear, interlaminar behavior, open-hole performance and joints. [4][5]
Compression is not simply a tensile test with the force reversed. Under compression along the fiber direction, the fibers rely on the resin for lateral support. Local waviness, resin shear and delamination can affect failure. Through the thickness, there are no continuous fibers spanning the plies in the same way as within the laminate plane, so the load path depends more heavily on the resin and interfaces. These differences explain why a single laminate requires several types of testing, and why a fiber tensile-strength value cannot be converted into a complete set of design allowables. [4]
Two Manufacturing Stages: Making the Fiber, Then the Composite
Carbon fibers commonly used in structural applications begin as polyacrylonitrile (PAN) precursor fibers. The precursor undergoes stabilization, followed by carbonization in a controlled atmosphere. Some high-modulus products undergo further graphitization. These treatments change the internal carbon structure, influencing the final strength and modulus. Surface treatment and sizing prepare the fibers for subsequent processing and bonding with resin. The product delivered at this stage is a fiber tow. The “12K” following a grade designation means that a tow nominally contains about 12,000 individual filaments; it is not a strength rating. [1]
The second stage can be illustrated using thermoset prepreg. Unidirectional fibers or woven fabric are pre-impregnated with a resin such as epoxy or cyanate ester, laid onto a mold at specified angles, consolidated under vacuum and cured at the temperature, pressure and duration required by the material system. After demolding, the component still requires trimming, hole machining, fitting installation and inspection. Thermoplastic prepregs also exist, but their forming process involves melting and cooling rather than thermoset resin curing. [1][4][6]
Layup is not the only forming method. Filament winding places continuous fibers along prescribed paths on rotationally symmetric shapes. Automated tape laying and automated fiber placement can deposit material along planned paths across large surfaces or complex contours. Process selection depends on geometry, fiber orientation, thickness variations and the ability to control defects; it cannot be determined by the fiber grade alone. [4]
Prepreg has specified fiber areal weight and resin content, but the finished component still depends on placement, consolidation and curing. If the fiber volume fraction is too low, fewer load-bearing fibers occupy a given cross-sectional area. Simply increasing it, however, can impede wetting and resin flow, leaving dry regions and voids. Wrinkles at corners, abrupt thickness reductions at ply terminations and fibers cut during hole machining all change local stresses. [4][6]
The resin transfers loads between fibers and also influences transverse, shear, interlaminar and post-damage performance. The same fiber combined with different resins, or processed using different cure cycles, cannot automatically share one set of laminate design allowables. Delivery documentation should, at a minimum, identify the resin system, prepreg condition, layup, cure cycle, defect acceptance limits and performance test conditions. Nondestructive testing can detect defects such as delamination and debonding within specified detection limits. Manufacturing records provide traceability for layup and curing. These serve different purposes. [4][6]
The Same Material, Different Layup Directions
Take the plate’s lengthwise direction as 0° and its transverse direction as 90°. With the material and total thickness unchanged, adding 0° plies generally improves lengthwise tensile stiffness and load-bearing potential. ±45° plies help transfer in-plane shear and off-axis loads, while 90° plies support transverse loading. The total fiber allocation is limited: assigning more to one direction leaves less for the others. [3][4]
Both the proportions and stacking sequence of the plies matter. A symmetric layup mirrors the plies about the mid-plane and can eliminate coupling between in-plane loading and bending in an ideal laminate. A balanced layup pairs positive and negative ply angles, helping reduce certain forms of in-plane extension–shear coupling. Symmetry and balance address different problems and cannot substitute for one another. [3]
Under ideal conditions, with identical ply materials and thicknesses, [0/+45/−45/90]s has quasi-isotropic in-plane extensional stiffness. The subscript s indicates that the sequence is mirrored about the mid-plane. Here, “quasi-isotropic” refers only to the specified in-plane stiffness; it does not guarantee equal bending stiffness, interlaminar properties or strength in all directions. [3]
Consider two layups made from the same eight plies: one places most plies at 0°, while the other distributes them among 0°, ±45° and 90°. The first devotes more fibers to a single principal direction; the second also accommodates shear and transverse loading. If deflection governs the design, the support span and each ply’s distance from the mid-plane must also be considered. If the component contains holes, inserts or joints, local load transfer after fibers have been interrupted must be checked. Designing by direction involves more than simply aligning fibers with the main load. [3][4][5]
Figure 1. Trade-offs in ply orientation with the same material and total thickness. Illustration based on NASA laminate references; it does not represent the actual layup of a specific structure.
Seven Grades: Comparing Strength and Modulus
To keep the values comparable, Figure 3 uses seven grades from a single publicly available Toray product sheet. Each pair below gives typical tensile strength in MPa followed by tensile modulus in GPa: T300, 3,530/230; T700S, 4,900/230; T800S, 5,880/294; T1100G, 7,000/324; M40J, 4,400/377; M55J, 4,020/540; and M60J, 3,820/588. [2]
These values come from resin-impregnated strand testing and allow comparison of typical axial tensile performance at the fiber level. They are neither individual-filament data nor design allowables for laminate compression, laminate shear or flight hardware. Points farther to the right on the chart have higher axial modulus; points higher on the chart have higher typical tensile strength. M60J resists stretching more than T1100G, yet its listed typical tensile strength is lower. There is therefore no single “highest” grade suitable for every mission. [2]
For example, T300 and T700S both have a listed modulus of 230 GPa, while their typical strengths are 3,530 and 4,900 MPa, respectively. If preliminary selection aims to control elastic extension under otherwise identical conditions, their moduli alone show no difference. If the concern is axial tensile failure at the fiber level, the strength values distinguish them. Laminate design then requires compression, shear, interlaminar and open-hole properties for the selected fiber/resin combination under specified layup and manufacturing conditions. [2][4]
A hypothetical calculation illustrates modulus alone. Two fiber tows are each 1 meter long, have the same total effective fiber area of 1 square millimeter and carry an axial tensile force of 100 newtons. Using ΔL = FL/EA, T700S at 230 GPa extends by approximately 0.435 millimeters, while M60J at 588 GPa extends by approximately 0.170 millimeters. This calculation shows what greater resistance to stretching means. It does not account for the effects of resin, layup, hole edges or joints on a finished component. [2][4]
Figure 2. Typical tensile strength and modulus for seven Toray grades, measured using resin-impregnated strand testing. These are not laminate or structural-component design allowables.
Thermal Expansion Is Also a Composite Property
The coefficient of linear thermal expansion, α, describes the relative change in a material’s length for a temperature change of 1 K. Its unit is 1/K, often expressed as ppm/K, or parts per million per kelvin. For composites, the measurement direction must also be specified. In a unidirectional ply, the fibers constrain thermal expansion and contraction along their axes, while transverse behavior depends more strongly on the resin and fiber arrangement. The thermal expansion coefficient of the fiber itself cannot substitute for that of a composite ply or multidirectional laminate. [3][8]
Two examples illustrate the scale of these values under specified conditions. Material inputs used in a NASA structural study of unidirectional T300/5208 carbon/epoxy give +0.432 ppm/K along the fibers at 0° and +29.16 ppm/K perpendicular to the fibers at 90°, under room-temperature, dry conditions. The fiber volume fraction was approximately 62%–67%. These are two directions within the same material system, not a comparison between two fiber grades. [10]
Hexcel’s table of typical prepreg laminate properties, at approximately 60% carbon fiber volume fraction, lists +0.3–0.7 ppm/K for high-strength unidirectional carbon/epoxy and +2–3 ppm/K for woven-fabric laminates. These published typical values apply to different reinforcement forms. They cannot be used to assign a universal coefficient to each fiber grade, such as T300 or T700S. [11]
The effective α of a multidirectional laminate varies with ply materials, angles, proportions and thicknesses. Adjusting the layup can reduce expansion in a specified direction, but it also changes stiffness in that direction and load-bearing performance in others. Material comparisons should record the coefficient alongside the material system, layup direction and test conditions. Detailed design should then use the corresponding laminate data. [3][9]
The coefficient may be positive, close to zero or negative in a particular direction and temperature range. It is generally not a constant that remains unchanged from room temperature across all temperatures. A fiber tow product sheet listing only strength and modulus does not provide enough information to infer a laminate’s α. As with the fiber-grade chart, thermal expansion data initially support comparisons between material systems; they must then be related to the test conditions for the actual layup. [8][9]
From Material Selection to a Structural Design
Begin by defining the main objective. If unidirectional tensile failure governs, compare strength values obtained using the same test basis. If elastic deformation governs, consider modulus together with cross-sectional geometry and length. If loads include compression, shear or multiple directions, move to laminate and joint data. For dimensional stability, also examine the effective thermal expansion coefficient in the relevant layup direction. Strength, modulus and α are inputs to different questions; they cannot replace one another. [2][3][4]
Next, treat the fiber, resin, prepreg, layup and manufacturing process as a single material system. Use specimens to verify tensile, compressive, shear and other properties, and test representative layups, open-hole configurations and joints to examine critical weaknesses. Assess the finished component against dimensional, defect, connection and load requirements. The fiber grade provides a starting point for selection; the material system and manufacturing quality determine whether its advantages survive in the finished part. [4][5][6]
The value of carbon fiber composites lies in their ability to place more material in the directions and locations where it is needed. That same flexibility requires designers to answer three questions together: what the fibers can provide, what the layup and resin turn them into, and how testing proves that the finished component meets its requirements.
The same principle applies beyond composite structures: a satellite project needs hardware and verification matched to its mission, rather than the highest specifications in every category. Drawing on China’s expanding satellite production capacity, STARPATH GLOBAL offers international customers competitively priced satellite platforms, payloads and assembly, integration and test (AIT) equipment. To explore options suited to your mission and budget, contact STARPATH GLOBAL to discuss your technical requirements and procurement needs.
References
[1] Toray Composite Materials America, Carbon Fiber Terminology; terminology for PAN carbon fiber, CFRP and prepreg.
[2] Toray Composite Materials America, TORAYCA Carbon Fiber Card, p. 2; typical tensile data for seven grades measured using resin-impregnated strand testing.
[3] NASA-RP-1351, Basic Mechanics of Laminated Composite Plates; laminate stiffness and coupling, not strength assessment.
[4] ECSS-E-HB-32-20, Structural Materials Handbook, Part 3, §31, continuous carbon fiber composites; Part 4, §34, manufacturing quality control.
[5] ECSS-E-HB-32-20, Structural Materials Handbook, Part 2, §22; joint design.
[6] ECSS-Q-ST-70-15C, Non-destructive Testing; nondestructive testing. For process control, see [4].
[7] NASA NESC Academy, Sandwich Structures Failure Modes and Their Prevention; sandwich-structure failure modes.
[8] NASA, Transverse Thermal Expansion of Carbon Fiber/Epoxy Matrix Composites, 1983; differences between longitudinal and transverse thermal expansion.
[9] NASA, Sensitivity of the Coefficients of Thermal Expansion of Selected Graphite Reinforced Composite Laminates to Lamina Thermoelastic Properties, 1992; the relationship between effective coefficients and layup.
[10] NASA-CR-159302, Design and Analysis of a Stiffened Composite Fuselage Panel, Table 2; thermal expansion design inputs for unidirectional T300/5208 at 0° and 90°.
[11] Hexcel, Prepreg Technology, p. 28; typical thermal properties of high-strength carbon/epoxy prepreg laminates.
