D8: Laminate - Service Flashcards
Why are composites bad at absorbing impact damage?
- Because they are brittle & exhibit very little plastic deformation
- Means they have low energy absorption compared to materials with a plastic zone (area under stress strain curve)
What does BVID stand for? What does it mean as a design technique? Why is it necessary?
- Barely visible impact damage
- Assume that there is internal damage that we cannot see (1inch diameter hole corresponding to delaminated zone)
- Because impact damage shows little sign on surface but propagates through the material, delaminating and shattering fibres in a cone. Can’t be detected with visual inspection, so have to ensure components are strong enough even if BVID is present.
What 8 things can we do to design for impact?
- Use toughened epoxies or TPs for matrix
- Minimise grouping of plies
- Use kevlar & glass hybrids with CF
- Use fabric plies as outer surface (more compliant)
- Use 45 deg outer plies to protect 0 deg plies
- Avoid minimum gauge
- Design for replaceability & reparability
- Model impact dmg as a hole of equiv. thickness for initial analysis
What are reduced allowables? Why are they necessary?
- A limit on general strain, typically 0.4% at ultimate load
- To ensure that BVID doesn’t grow throughout the structure’s lifetime
What are the limits for visible and non-visible damage, respectively?
- Component must be repaired before strength degrades to less than limit load
- Damage must not degrade strength to less than limit load
How is impact testing carried out? What is the most limiting test for composites?
- Indenter dropped onto a clamped specimen, followed by residual strength tests to assess the effect on the component’s structural integrity.
- Residual hot wet compressive strength after impact test
How does composite fatigue performance compare with metals?
Better, provided that the laminate is fibre dominated in direction of loading & loading is in-plane.
What are the 3 phases of composite fatigue? How long do they last relative to each other?
- Wear in, stable growth, wear out
- Short, long, short
What does the wear-in fatigue phase consist of? When does it stop?
- Failure of weak fibres
- Failure of weak matrix regions (e.g. areas of porosity)
- Failure of weak interfaces
- Material essentially stress-relieving
- When the material reaches a stable “characteristic damage state”
What does the stable-growth fatigue phase consist of?
- Damage accumulation slows down
- Matrix cracks and interface disbonds slowly grow and couple into larger cracks
- Occurs particularly at high stress concentrations
What does the wear-out fatigue phase consist of?
- Areas of cracks amalgamate to form delaminated zones
- Zones grow as 2D cracks between adjacent plies
- Propagation accelerates until area is weak enough to fail under the cyclic load
What’s different about the location of fatigue damage accumulation in composites vs metals?
In composites, damage accumulates throughout the material, in metals damage tends to be a singular predominant crack at a high stress concentration
What fatigue loading conditions are required to degrade composite stiffness?
Very high stresses at very high cycle numbers
What shape are composite S-N curves in tension?
Characteristically very fact, even as cyclic stress approaches material static ultimate stress.
How does the fatigue performance under compressive loading compare to tensile? Why?
- Less impressive
- Fatigue by delamination growth can be significant
- Temperature and moisture can be significant because of weakened matrix support of fibres
What type of cyclic loading is the most damaging in composites? Why?
Cycling between tension and compression, because of interactions between different failure modes
How do we account for statistical scatter when designing for fatigue?
- Extreme value statistics (Weibull)
- Use a stress factor (instead of a life factor, because of significant scatter at high stress levels & flat SN curve)
What are 5 ways to design against fatigue damage? Why are they necessary?
- Use conservative designs to lower stress & strain levels
- Use damage-tolerant design
- Demonstrate no damage growth of impact specimens
- Verify fatigue life by testing to several lifetimes, and in different conditions
- Avoid out-of-plane loads
What are the 4 levels of fatigue test performed and how many are performed?
- Major test, complete airframe (usually one)
- Sub component assembly test (3 per component)
- Structural features, joints & notches etc (many, to verify scatter)
- Coupons (many, to verify scatter)
Why are fatigue tests for composites longer and more expensive than those for metals?
- Increased sample size required to account for increased scatter
- Load frequencies must be kept low (5-10Hz) to prevent the build up of internal heat (bad dissipation)
- Tests performed at higher stress levels for fewer cycles cannot be extrapolated down
How does the notch sensitivity of metals and composites? Why?
- Under static loading, metals are notch-insensitive, because the relatively high loads exceed the elastic limit, alleviating concentrations through ductile deformation.
- Composites can’t deform plastically to alleviate the stress concentrations, so they’re notch-sensitive under static loading.
- Under the relatively lower cyclic stresses, metals are notch-sensitive as the stresses aren’t high enough for ductility to alleviate concentrations.
- Accumulation of damage at notches in composites can reduce stress concentrations under lower cyclic loading, making them notch insensitive.
What does the crack propagation phase look like in metals vs composites?
- Singular crack growth vs multiple regions of damage
- Predominant under tension vs compression
- Damage growth perpendicular to direction of loading vs relatively undefined direction
What is the inspection window like in metals vs composites? Why?
- Cracks in metals are very small for most of their lives, but growth accelerates. Therefore, window between a crack becoming big enough to see & becoming catastrophic is small.
- Relatively large window for composites, because of BVID allowance we tolerate damage up to an inch wide which is easy to detect with NDT. Also, strict strain limit of 0.4-0.5% keeps damage growth slow and stable.
How are the properties of metals and composites affected by fatigue damage respectively?
- Metals affected in a small region at the crack tip (plastic zone)
- Composites suffer from general degradation of the entire loaded material.
How does the environment affect metals vs composites?
- Metals relatively unaffected by moisture and temperature (affected by corrosion)
- Composites suffer a significant reduction in compression ability (matrix support), and a long-term degradation of properties
How do the testing spectra of metals and composites compare? Why?
- Metals include high freq low cycles because they are sensitive to them. Neglect higher loads as they can have a beneficial crack blunting effect.
- Composites neglect high freq low loads because they are insensitive to them. Have to include lower frequency high loads because sensitive.
What is the first stage of classical phenomenological fatigue analysis? Why can’t it be applied to composites?
- Load idealisation: load spectrum reduced to sets of constant amplitude cycles. Done based on stable hysteresis loops of plastic deformation in the damage initiation phase.
- Because it doesn’t have a damage initiation phase and doesn’t exhibit plastic deformation.
What is the second stage of classical phenomenological fatigue analysis? Why can’t it be applied to composites?
- Data manipulation: limited S-N data extrapolated, under the assumption that diff. cycles of load at diff. combos of mean and alternating stress have equivalent damage effects.
- The assumption doesn’t hold up for composites, different modes of failure can occur under the different types of loading.
What is the second stage of classical phenomenological fatigue analysis? Why can’t it be applied to composites?
- Data manipulation: limited S-N data extrapolated, under the assumption that diff. cycles of load at diff. combos of mean and alternating stress have equivalent damage effects.
- The assumption doesn’t hold up for composites, different modes of failure can occur under the different types of loading.
What is the third stage of classical phenomenological fatigue analysis? Why can’t it be applied to composites?
- Damage accumulation: assume that the damage accumulation is non-interactive/linear, and is independent of: load mixing, current damage state, and waveform/frequency.
- The exact opposite is true for composites.
How does classical fatigue analysis hold up when applied to composites?
- Pretty good for GFRP despite lack of validity of initial assumptions
- Very limited agreement for CFRP
What are the 3 design allowables, from largest to smallest?
- Material means
- Material allowable (knocked down using statistical methods (A or B) to account for scatter)
- Design allowable (includes hot+wet property reduction, long term degradation, hole & defect tolerance etc)
What are the 4 design limits, from smallest to largest?
- Operating levels
- Design limit (largest load likely to be encountered during lifetime)
- Proof load (load that must be sustained with limited permanent effects, structure can still perform function after)
- Ultimate load/design ultimate (catastrophic failure)
When are A and B basis scatters used?
- Primary and secondary structures respectively (A is more conservative)
What are the design strain limits at ultimate load for CFRP (direct and shear) and GFRP (direct)? What’s different about how they’re defined?
- CFRP: 0.4-0.45% for direct (comp-tens), 0.8% for shear. Driven by static strength.
- GFRP: 0.4-0.5%. Driven by fatigue.
