Degradation Failure Mechanisms

Time Dependent Failure Mechanisms

Conceptual Model

corr_crk_v0.mp4

Click the tutorial video above for a conceptual overview of estimating failure probabilities for time dependent failure mechanisms.

The conceptual model for any time dependent failure mechanism involves the following steps

Estimate exposure
Estimate mitigation as the chance (probability/frequency) of the exposure being reduced or eliminated at the subject location
Estimate resistance as the effective wall thickness that is being degraded by the time dependent failure mechanism
Produce TTF estimate based on the mitigated exposure (mpy or mm/yr) and the effective wall thickness (mils/inches/mm)
Use TTF estimates to drive integrity assessment scheduling
Convert TTF’s to FoF’s in order to compare to other failure mechanisms

See also relevant chapters in book: Chapter 6 Time Dependent Failure Mechanisms

Corrosion

Assessing Corrosion Potential

special discussion of Corrosion algorithms
possibly the most complex of all algorithms, reflecting the real world complexities associated especially with corr control
in the case of corr, the real-world situation can be reasonably modeled to be effectively binary: either maximum corr is occuring or no corr is occuring
every linear foot has a chance of either 1) no corr or 2) full corr
but we can also model this binary as a continuum of possible corr rates in order to capture the benefits of better mitigation
ie, where chances of corr (no mit) are higher, it is modeled as higher corr rate; where chances of no corr (mit is working) is higher, it is modeled as lower corr rate

For Ext Corr:

coating gaps/mile/5280 x CP gaps/mile/5280 = active corr pts / mile
activ corr pts / mile = prob of no mitigation in each linear foot = (1 – mit eff)
if mit = 1, then corr = 0.000001 mpy (eff zero)
if mit <1, then corr = mpy x (1 – mit)
TTF99 = eff wall / mpy = PXX TTF based on corr rate that occurs in absence of mitigation
TTF = eff wall / (mpy x (1 – mit)) = probabilistic TTF that is occuring in each linear foot

See Mathematical Basis for recommended approach

Cracking

Fatigue vs EAC: purely mechanical mechanism versus ‘environmentally assisted cracking’ mechanisms (SCC, SSC, HIC, etc)

same conceptual model as for corrosion

fatigue is usually examined in terms of crack initiation, activation, and propagation. EAC often adds concepts of crack colonies and coalescence.

often fewer mitigation options for cracking

can change exposure such as pressure cycling, but this is not technically a mitigation
can increase resistance, but again, not a mitigation
measures such as increasing depth of cover in order to lessen the transmission of traffic loads onto a buried pipe IS an example of a mitigation.

Fatigue from internal pressure. Rainflow estimation tool. Fatigue cycle counting.

Resistance

The component’s wall thickness and stress condition is a critical determinant of ability to resist degradation mechanisms. A modeling technique of effective wall thickness is the recommended approach to efficiently complete the FoF estimation. This is in the context of the overall resistance estimation that considers all loads and stresses associated with any failure mechanism.

A degradation mechanism active in a pressurized component will require both leak criteria and rupture criteria be considered in concert, ie, as degradation advances through the material, either the pressure-containing capacity (rupture resistance) or the
fluid containment capacity (leak resistance) will be lost first. Either results in loss of integrity.

Lack of toughness should also be considered in the resistance assessment. Lower toughness makes crack initiation, activation, and propagation more probable and rupture more likely. At higher stress levels, more toughness is required to arrest a running
brittle fracture. Larger diameter or thinner wall pipes require proportionally higher toughness to prevent running brittle fracture.

TTF

With all three essential elements, exposure, mitigation, resistance, the FoF estimate can now be generated. An intermediate value, TTF, is generated as part of this calculation when degradation mechanisms are involved.

Resistance
Effective wall thickness
TTF, TTF99
PoF from TTF

TTF / Remaining Life Estimates

Some detailed, underlying assumptions accompanying the estimation of TTF and PoF are illustrated in the following example.

For external corrosion, we examine the number of coincident coating holidays and CP gaps–ie, the number of locations where both are occurring at once. We must include scenarios where a coating defect allows migration of electrolyte (water usually) beneath the coating for some distance from the defect. These locations of overlapping coating gap and CP gap should be the only locations where active corrosion may occur.

a min TTF should NOT be directly used to determine integrity re-assessment interval since it has a probalistic aspect; ie, each linear foot contributes to overall TTF
- for integrity reassess, can use either of 2 options:
  - TTF99, this gives a very conservative value since it assumes no mitigation anywhere
  - or, calculate an aggregate TTF for the pipline section for which integrity assess is desired
    - convert all TTF’s (from each linear ft) to PoF’s
    - OR gate all PoF/ft ‘s together to get PoF total
    - extract TTF from the overall PoF of the section for which integr assess is being calculated
      - eg, assume 10 TTF’s of 100 years each
        using PoF=1/TTF (alternative PoF-TTF relationships are available) on each TTF, PoF total = (1/TTF1) OR (1/TTF2) OR ….. (!/TTF10) = 10%/year for 10 ft length
        TTF overall = 1/PoF total) = 1/10% = 10 years
      - can see importance of this procedure, since max TTF would have been 100 years, versus the more correct answer of TTF = 10 years
- might appear to be complicated process, but actually only moving back and forth between PoF and TTF. Since, when seeking an estimate for multiple dyn-segs, TTF’s are not logically additive or dependent, the PoF conversions are needed.
- this is actually a very robust way to discriminate mitigation effectiveness and corrosivity aggressiveness
  - captures benefits of more detailed monitoring of CP and coating, versus using only minimum techniques
- note that when defects are known, the ‘per mile’ unitization needs to be used carefully. for instance, a 1 ft defect with a failure frequency of 1.0/yr has a failure frequency of 5280 per mile-year. this is necessary for math to work as intended. this nuance became apparent in a RA where serious defects–weak link in chain analogy–were not forcing high overall FoF’s because defect FoF values were not entered in units of per mile-year.

TTF / Remaining Life Estimates