Creep damage and recommended actions

[kclim]

Have a query with regards to high energy piping, on thermal power stations and creep damage.

There is much data available with regards to how to classify creep damage - we tend to adopt the VGB TW-507 criteria for the relevant materials, which is essentially a 1 (no creep voids) to 5 (macrocracking) ranking, with subrankings for certain stages.

However, we have not found much in the way of actions to take given a certain damage classification. In particular, when the creep damaged areas need to be revisited. There is an old EPRI document (TR-100549 - Life assessment product catalog) which suggests reinspection intervals depending on previous service history, in addition to damage classification. But that aside, I haven't seen to much else available in that vein. Furthermore, the EPRI methodologies tend not to cover vanadium low alloy steels which much of our HEP consists of.

Can anyone shed some further light on this issue?

Thanks in advance.

 

[davefitz]

For low alloy vanadium / niobium/ wolfram-modified alloys fabricated prior to 2007 in the US, there is concern that some forged or cast or hot-worked piping elements were not properly heat treated following shop fabrication. There is also concern that HEP butt welds might not have been adequately monitored during PWHT to prevent over-tempering the material directly under the heating coils. So, in addition to evaluating the operating temperature history of the piping, one may adopt the conservative posture of obtaining a boat sample of any HEP section that exhibits accelerated creep damage, under the suspicion that it is not exactly the alloy that is described in the plant's P4 forms.

"Whom the gods would destroy, they first make mad "

 

[metengr]

Typically, for HEP programs at steam plants across the US, the interval for condition assessment is about every 7 years. This is after a walkdown is performed, initial inspections are completed and review of operating history and materials of construction have been completed, similar to a risk assessment.

Normally, if local creep damage is found in welds by surface NDT or combination of surface NDT and replication, the creep damage is characterized and removed by further surface grinding. If the creep damage clears (regardless of weld repairs), the weld is placed back into service and will be targeted for inspection after 7 years. If the creep damage is re-occurring in 7 years and extends beyond the initial location, a shorter time interval may be necessary in addition to repeated comprehensive volumetric UT, like phased array.

 

[kclim]

Metengr,

Thanks for the info. How does the 7 year period fit in with the typical outage program in the US?

Davefitz,

From the what I have read, the CMV materials are a lot more problematic than standard CrMo, particularly with regards to Type IV and reheat cracking. We have MDRs with heat charts, although anyone can produce a good looking heat chart. Is there anything specifically you look for when reviewing old documents?

 

[davefitz]

kclim,
I cannot comment re: CMV materials, as I have limited experience with them.

The only CMV elements I have addressed are steam turbine inlet valves , and the only major issue I had seen was related to the dissimilar metal weld between the 1.25 Cr .5 Mo V turbine stop valve and the ( 50% thinner) P91 HP main steam pipe. This weld must include a forged F91 transition piece with a wall thickness gradient slope less than 18 degrees to avoid excessive thermal stress at the weld interface. GE documents require this transition piece be provided , but other STG vendors have not provided that requirement to the EPC's. Without the transition piece , complete failure of this weld has occurred within 12,000 hrs.

"Whom the gods would destroy, they first make mad "

 

[metengr]

kclim;
In the US, fossil plant outage schedules are boiler or turbine driven. The boiler outages are typically spaced every 24 to in some cases 36 months (4-6 week duration). Turbine outages (minimum 8 to in some cases 10 weeks) are as follows; HP/IP 6-8 years and LP 10 years.

Non-turbine piping condition assessments are typically performed during boiler or turbine overhauls. The turbine components, like control valves are done every 4 years.

 

[weldstan]

One of the other major failure areas is the main steam and hot reheat bypass valves and associated attemperators/desuperheaters. Thermal (creep) fatigue failures have occurred in as little as 5000 hrs when near daily cycling of the plant was done.

 

[DBreyer]

If you want to stick to the VGB standards than you could use VGB-R 509 (periodic inspection of pipes in conventional power plants).
Here you will find recommendations for the parts to inspect, the intervals and the recommendations for the different life stages (classification according to the ranking of VGB-TW 507) of the equipment ( e.g. reduce interval, reduce temperature, replace,...).


Daniel Breyer
Inspection Engineer

http://lnkd.in/dFcQA-w