04.12.1 / CDS Training Manual

Predict T700/PR520
effective lamina properties.

A bottom-up micromechanics study using constituent values and three fiber-volume regions published for a NASA triaxial-braid model.

Complete manuals · download a whole collection

Each collection includes an editable Word manual and its matching searchable PDF, with linked contents, explanatory text, diagrams and references. Theory equations use native Word mathematics. Individual-page PDF buttons remain available.

Complete User Guide148 pages · 10 chapters · Updated 2026-09-18
↓ PDF↓ Word
Getting Started Handbook139 pages · 8 chapters · Updated 2026-09-18
↓ PDF↓ Word
Training & Exercise Manual386 pages · 115 chapters · all 104 exercises · Updated 2026-09-18
↓ PDF↓ Word
Complete Theory Manual255 pages · 56 chapters · Updated 2026-09-18
↓ PDF↓ Word
Models & Workflow Manual85 pages · 9 chapters · Updated 2026-09-18
↓ PDF↓ Word

These are dated reading editions of the public learning content. Interactive studies and account-controlled classroom notes and databases stay online. For current changes, follow the chapter links back to the website.

Current Workbench, reference editions and downloads

Use the online User Guide for current controls, the released model directory for selectable models and compatibility, and the Training Manual for connected exercises. Wider theory references do not mean every formulation is enabled in Workbench.

Dated GUI captures and compiled Web r18 PDF manuals retain their stated scope; they are not a substitute for current hosted Workbench instructions. Find the overview PDF, guides and exercise databases in Your CDS library.

Two-row simulation workflow · Open full-size map ↗
Linked-input preview, not solved results. Gray blocks are not configured.
Predict T700/PR520 effective lamina properties. two-row simulation workflow with connected input blocks, fatigue and optimization
∑ Selected models & submodels

These are the exercise’s linked choices, not solved results. Open a first-layer model to see its submodels and scope.

Kamal–Sourour autocatalytic · Kamal–Sourour Epoxy Cure

Reusable material model. It contributes only when assigned to a material used by this simulation; separate model records can supply independent properties.

Calibration & applicability

Illustrative model defaults — replace with characterized resin kinetics

∑ Theory & assumptions
CHILE (degree of cure) · CHILE cure-dependent modulus

Reusable material model. It contributes only when assigned to a material used by this simulation; separate model records can supply independent properties. Cure-dependent modulus does not automatically scale strength allowables.

Calibration & applicability

CHILE(α), section 2.4 of Materials 2019, 12, 259. Example parameters, not measured EP180 data. Fixed Poisson ratio; no viscoelastic relaxation or Tg softening. Incremental elastic stress integration at saved process intervals: check time-step convergence. Strength allowables are independent measured inputs, not scaled with modulus.

∑ Theory & assumptions
Temperature-dependent tabular · Temperature-dependent Thermal

Reusable material model. It contributes only when assigned to a material used by this simulation; separate model records can supply independent properties. Micro uses the 23 °C reference; Process evaluates the same table at local temperature, without extrapolation.

Calibration & applicability

Check source data, applicable environment and validity limits in the model record; saved defaults are not experimental validation.

∑ Theory & assumptions
1D Fickian diffusion · Layered Fickian Diffusion

Reusable material model. It contributes only when assigned to a material used by this simulation; separate model records can supply independent properties.

Calibration & applicability

Check source data, applicable environment and validity limits in the model record; saved defaults are not experimental validation.

∑ Theory & assumptions
Halpin–Tsai · T700 / EP180 UD

One compatible homogenization model per Micro recipe. Separate recipes compare models; their predictions are not blended.

∑ Theory & assumptions
CLT · Linear static with failure indices · Plate — Nx Static Validation · Unidirectional lamina

Shared laminate stiffness drives this membrane/CLT path. Failure criteria are independent comparisons, not blended models. Fatigue is a separate assessment.

Failure criterion · Maximum stress

Primary criterion. Additional envelope comparisons are chosen in Response → Failure; they run independently.

∑ Theory & assumptions
Data travelling between blocks

MicroLaminates
Predicted ply stiffness, strength, density and expansion properties.

MaterialsMicro
Constituent stiffness, strength, density and thermal / moisture properties.

ModelsMaterials
Model choices and calibrated parameters.

MechanicalSimulation
SIMULATION selects this case and its analysis model; the case owns its applicable cycle and input references.

LaminatesMechanical
Ply angles and thicknesses, stiffness, mass and ply properties.

GeometryMechanical
Shape and dimensions for the selected structural analysis.

SimulationOptimization
Linked inputs and current-property response for candidate evaluation.

Reference validation · 2026-09-18

Not benchmark validated

No matched quantitative reference comparison completed for the full exercise.

Recorded baseline only—not a certification of the current database, edited inputs, or every output. Analytical agreement is not experimental material or failure validation.

Source validation, comparisons and tested inputs

Execution: Completed (not validation)

The copied T700/EP180 database is a connected starter, not T700/PR520. Complete the constituent setup below before comparing with NASA.

Tested inputs

Input SHA-256: f332a9c41bf7dc1aaa5eb2521be80c1e6feeef085db48ea61d81a3475b42b7b9
Solver source SHA-256: 30c8a614b6cb004a26da8adaf8ab7c534c37c4d534836921939f35a549e1bcc5

  • Constituent-to-lamina screening only. The NASA braid subcells, calibrated LS-DYNA damage and coupon validation are not reproduced.
Separate published-data fixture: Benchmark discrepancy

Source-matched constituent fixture compared with NASA MAC/GMC, 1% exploratory equivalence criterion declared before running. A different homogenization model need not match; mismatch is retained, not fitted away.

Acceptance: absolute error ≤ absolute tolerance + relative tolerance × |reference|.
QuantityCDSReferenceError %Tolerance (relative + absolute)Result
E1 at Vf=0.375 (GPa)88.7588.50.28248591% + 0.00002 GPaPass
E2 at Vf=0.375 (GPa)5.517246.22-11.298391% + 0.00002 GPaDiscrepancy
G12 at Vf=0.375 (GPa)2.232562.6-14.132311% + 0.00002 GPaDiscrepancy
E1 at Vf=0.733 (GPa)169.658169.50.093215341% + 0.00002 GPaPass
E2 at Vf=0.733 (GPa)8.649279.9-12.633641% + 0.00002 GPaDiscrepancy
G12 at Vf=0.733 (GPa)4.704457-32.793571% + 0.00002 GPaDiscrepancy
E1 at Vf=0.8 (GPa)184.8184.70.054141851% + 0.00002 GPaPass
E2 at Vf=0.8 (GPa)9.6774210.9-11.216331% + 0.00002 GPaDiscrepancy
G12 at Vf=0.8 (GPa)5.934076-1.0988331% + 0.00002 GPaDiscrepancy
Fixture inputs
{
  "fiber": {
    "E1": 230,
    "E2": 15,
    "G12": 27,
    "nu12": 0.2
  },
  "matrix": {
    "E": 4,
    "G": 1.44,
    "nu": 0.38
  },
  "volumeFractions": [
    0.375,
    0.733,
    0.8
  ],
  "voidFraction": 0,
  "model": "Rule of mixtures"
}
Separate published-data fixture: Benchmark discrepancy

Source-matched constituent fixture compared with NASA MAC/GMC, 1% exploratory equivalence criterion declared before running. A different homogenization model need not match; mismatch is retained, not fitted away.

Acceptance: absolute error ≤ absolute tolerance + relative tolerance × |reference|.
QuantityCDSReferenceError %Tolerance (relative + absolute)Result
E1 at Vf=0.375 (GPa)88.7588.50.28248591% + 0.00002 GPaPass
E2 at Vf=0.375 (GPa)6.622526.226.4713831% + 0.00002 GPaDiscrepancy
G12 at Vf=0.375 (GPa)2.904062.611.694621% + 0.00002 GPaDiscrepancy
E1 at Vf=0.733 (GPa)169.658169.50.093215341% + 0.00002 GPaPass
E2 at Vf=0.733 (GPa)10.477619.95.8344441% + 0.00002 GPaDiscrepancy
G12 at Vf=0.733 (GPa)7.0001170.0015714291% + 0.00002 GPaPass
E1 at Vf=0.8 (GPa)184.8184.70.054141851% + 0.00002 GPaPass
E2 at Vf=0.8 (GPa)11.4366210.94.9231191% + 0.00002 GPaDiscrepancy
G12 at Vf=0.8 (GPa)8.80865646.810831% + 0.00002 GPaDiscrepancy
Fixture inputs
{
  "fiber": {
    "E1": 230,
    "E2": 15,
    "G12": 27,
    "nu12": 0.2
  },
  "matrix": {
    "E": 4,
    "G": 1.44,
    "nu": 0.38
  },
  "volumeFractions": [
    0.375,
    0.733,
    0.8
  ],
  "voidFraction": 0,
  "model": "Halpin–Tsai model"
}

Download source-validation evidence (JSON) · Full 104-exercise study

Objective

Predict how T700/PR520 effective ply stiffness changes across the three local fiber-volume regions used in NASA’s absorbed-matrix braid model.

CDS scope

CDS reproduces a constituent-to-lamina comparison. It does not reproduce the report’s complete braid subcell, LS-DYNA damage calibration, or coupon validation.

Published basis

1 primary source
NASA/TM—2015-218814Experimental and Numerical Analysis of Triaxially Braided Composites Utilizing a Modified Subcell Modeling Approach

Tables 3 and 4 provide T700 and PR520 constituent properties and MAC/GMC effective-ply values for Vf = 37.5%, 73.3% and 80%.

Starting data

Enter in the displayed units
VariableValueHow it is used
T700 density1.80 g/cm³Fiber physical property
T700 E1 / E2230 / 15 GPaFiber axial and transverse modulus
T700 ν12 / G120.20 / 27 GPaFiber coupling and shear
PR520 density1.25 g/cm³Matrix physical property
PR520 E / ν / G4.0 GPa / 0.38 / 1.44 GPaIsotropic matrix elasticity
Fiber volume fractions37.5%, 73.3%, 80.0%Three separate micromechanics evaluations

Review checkpoints

Comparison, not certification
ResultPublished referenceInterpretation
Vf 37.5%: E1 / E2 / G1288.5 / 6.22 / 2.60 GPaB-braider effective ply
Vf 73.3%: E1 / E2 / G12169.5 / 9.90 / 7.00 GPaA/C-braider effective ply
Vf 80.0%: E1 / E2 / G12184.7 / 10.9 / 6.00 GPaA-axial effective ply

All exercises, case studies and simulation examples → · Your CDS library: overview, manuals and database downloads →

01

Create traceable constituent records

Duplicate the nearest carbon-fiber and thermoset records so the library baselines remain unchanged.

  1. Name the records T700 — NASA case study and PR520 — NASA case study.
  2. Enter the density and elastic values from the starting-data table. For the isotropic matrix, use E = 4.0 GPa, ν = 0.38 and G = 1.44 GPa.
  3. Add the NASA report as the source, mark the values as published research inputs, and leave unsupported strength or transport terms visibly incomplete.
03

Sweep the three published fiber fractions

Use Lamina / Micro optimization or three duplicated model records to evaluate Vf = 37.5%, 73.3% and 80.0%.

  1. Hold all other inputs fixed and evaluate each volume fraction.
  2. Plot Vf against E1, E2 and G12; use a line view for the trend and the table for exact values.
  3. Save one optimization snapshot or export one comparison table containing all three cases.
04

Compare the effective lamina response

Compare CDS output with NASA’s MAC/GMC values in the checkpoint table. Expect the longitudinal trend to be strongest and model-dependent differences in transverse and shear response.

  1. Calculate percent difference as 100 × (CDS − reference) / reference for each modulus.
  2. Inspect the selected model’s assumptions before attributing differences to input data.
  3. Repeat with another supported homogenization model to show model-form sensitivity.
05

Save the predicted lamina for downstream work

Create a new lamina material from the selected prediction and retain the model and source links.

  1. Choose the fiber fraction appropriate to the intended region rather than averaging the three NASA subcells without justification.
  2. Save the generated material with Prediction status.
  3. Open the new record and verify that its provenance identifies both constituents and the micromechanics model.