P91 in Service Experience
P91 in Service Experience
PAGE 1
Grade 91 Background
> Used widely in the UK from 1989 onwards for retrofit superheater outlet headers delivering steam at ~568C > Later (early 1990s) used for HRSG headers and HP pipework on CCGTs with steam temperatures typically 540C > More recent application has been on CCGTs operating, or planned to operate, at 568C > Cracking in the UK to date has mainly been Type IV and mainly on retrofit headers (but at least one example had been found on a CCGT operating at 568C) > Thermal fatigue cracking has been found on endcaps (HRSG) > Examples of aberrant material have been found on both retrofit and CCGTs
PAGE 2
Grade 91 Failures
PAGE 3
Grade 91 Issues
> Why do we have a grade 91 problem? > For normal (martensitic) grade 91: The grade 91 Type IV shortfall in creep strength is substantial and has generally not been fully exposed by test programmes An inherent low Type IV strength is further exacerbated by low nitrogen : aluminium ratio > Grade 91 steel is vulnerable to incorrect heat treatment with the consequence that aberrant non martensitic material can, and has, entered service: Components may be supplied in the incorrectly heat treated condition Components may be supplied in the correctly heat treated condition and subsequently mistreated during PWHT after welding
PAGE 4
Grade 91 Issues
> Why is the grade 91 Type IV zone weak? The grain size is very small (<5m) The Martensite lath structure is suppressed (because of the small grains) The precipitates will be largely unaffected by the short heating cycle (but may coarsen slightly) > Aluminium nitride, if present, will not dissolve therefore any vanadium nitride lost to it prior to welding will remain lost
PAGE 5
Stress (MPa)
Time (Hrs)
RWE Power International PAGE 6
Stress (MPa)
Time (Hrs)
PAGE 7
Aluminium (wt%)
0.03
0.02
0.01
Nitrogen (wt% )
RWE Power International PAGE 8
Aluminium Nitride
TEM & FEGSEM images John Bates, Loughborough
0.05 High AlN Low AlN No AlN 0.04 N:Al =1.5 N:Al = 3.0 0.03
Aluminium (wt%)
0.02
0.01
0 0.02
0.03
0.04
0.05
0.06
0.07
0.08
Nitrogen (wt% )
PAGE 9
Aluminium Measurement
0.05 0.045 0.04 0.035 Al (wt%) 0.03 0.025 0.02 0.015 0.01 0.005 0 0 0.01 0.02 0.03 0.04 Al (wt% ) measured by OES
RWE Power International PAGE 10
y = 1.408x y=x
y = 0.9467x
wt% Al (Soluble) wt% Al (OES) wt% Al (MDM)
Grade 91 Issues
> As end users of grade 91 components, generators need to know: Will cracking/failure occur within the operating lifetime? If so, when will it start to become a problem? Once started, how large a problem will it be? > If grade 91 cracking/failure does occur, what are the optimum remedial methods: If cracking is ground out, how quickly will it reappear? If cracking is weld repaired, how quickly will re-cracking occur?
PAGE 11
Stress (MPa)
1000000
Life (Hrs)
RWE Power International PAGE 12
Stress (MPa)
2CrMo (N&T) Mean 120 100 80 60 40 20 10 100 1000 10000 100000 1000000 Bar 257 Type IV KA817 Type IV
Time (Hrs)
RWE Power International PAGE 13
PAGE 14
PAGE 15
Inspection Recommendations
General estimate for the onset of Type IV
130 120 110
590C 580C 570C 560C 540C
Stress (MPa)
Life (Hrs)
RWE Power International PAGE 16
Inspection Recommendations
Particular problem areas requiring special attention
> Terminal welds and branch welds > Non 90 branches (offset branches, welded Y pieces) > Endcaps (fully or partially welded) > Seam welds (whether renormalised or not) > Thermocouple pockets (specific arrangements) > Dissimilar metal joints
PAGE 17
PAGE 18
PAGE 19
PAGE 20
PAGE 21
PAGE 22
PAGE 23
PAGE 24
Aberrant Material
Simulated to match the microstructure observed in service
Normal martensitic structure (VHN 203) Non martensitic structure (VHN 151)
PAGE 25
Stress (MPa)
100000 1000000
Life (Hrs)
RWE Power International PAGE 26
Conclusions (1)
> The grade 91 story continues to develop and still has some way to run with more widespread cracking likely to occur > In the UK, GENSIP (Generators Safety & Integrity Programme) is sponsoring work to enable operators of aged coal-fired plant to operate grade 91 components safely > This is a global issue and collaboration with others (eg VGB, Vrmeforsk, EPRI) will be a key factor in meeting the structural integrity risks associated with the use of this material
PAGE 27
Conclusions (2)
PAGE 28