Assessment of eguipment lifetime using the metal magnetic memory method

Dr., professor Anatoly Dubov

According to the "Standard guideline for metal inspection and prolongation of TPS boilers, turbines and pipelines basic elements life", it is suggested to consider crack resistance as the main parameter characterizing the survivorship of power equipment units operating in conditions of cyclic loads. However, this characteristic is determined on samples, and transferring of laboratory testing results to real operational conditions does not provide an objective assessment of metal condition.

The article considers the abilities of the method of metal magnetic memory (MMM) to detect crack initiation zones directly on equipment and to trace the development of metal fatigue failure process in these zones. Based on 100% power equipment inspection using the MMM method it is suggested to detect all potentially dangerous zones with developing defects and to timely remove them during the repairs. Thus, an opportunity is offered to assess the real equipment life.

Paper [1] presented in detail the problems of aging equipment’s residual life assessment.

Among the basic scientific-technical problems of power equipment life assessment the following should be singled out:

• the lack of scientifically grounded concept of engineering diagnostics and life estimation;

• insufficient effectiveness of traditional non-destructive testing (NDT) methods and means at early diagnostics of fatigue damages and investigation of metal’s structural-mechanical properties;

• low effectiveness of the current techniques for calibrating strength calculation due to the lack of actual structural-mechanical properties of metal by all equipment elements and units;

• the lack in the broad practice of effective NDT methods and means allowing performing 100% equipment inspection in order to assess the stress-strained state and individual life of each unit and the entire aggregate.

The scientific-technical problems of equipment reliability assuring and prolongation of its life are aggravated by the lack of the required financial finds. According to estimation presented in the Concept of Technical re-equipment of RSC "UES of Russia" Electric Power Plants till 2015, the man-hours for providing of expert-predicted life, at its prolongation beyond the body life, may be about 50% from the cost of complete replacement of similar equipment’s power units. It is indicated that this high level of costs should be aimed at a 100% diagnostic examination of equipment, execution of calibration strength calculations and the analysis of technical-economic documentation on the experience in equipment operation. It is obvious that these costs were reasonable based on the experience in power equipment metal inspection gained by RSC "UES of Russia" counting on a 100% examination by traditional destructive and non-destructive testing methods.

Thus, nowadays at large Electric Power Plants, where equipment has exhausted its body life, a deadlock condition has formed. There is no money for equipment replacement and even for its 100% examination, and not a single organization will obviously take a responsibility to prolong the life without such an examination! At such conditions Power Plant management has to provide safe and reliable operation of the equipment.

It should be noted that execution of a 100% examination of power equipment using traditional NDT methods (UT, MPI, etc.) is associated not only with the high level of costs, but also it has low effectiveness due to its unsuitability for detection of fatigue damages at an early stage of their development.

According to standard guideline [2], it is suggested to consider crack resistance as the main parameter characterizing the survivorship of power equipment units operating in conditions of cyclic loads. It should be kept in mind that this is a conventional material characteristic determined by the ratio of the current (actual at a specific time and in specific conditions) crack growth rate to the critical rate for a specific material. However, this characteristic is determined on samples, and transferring of laboratory testing results to real operational conditions does not provide an objective assessment of equipment operability and efficiency.

Can assessment of cracks growth be made and their development zones be detected in real conditions directly on equipment?

It is known that the main goal of a 100% examination is to detect potentially dangerous stress concentration zones (SCZ), in which development of corrosion, fatigue and creep damages occur. Exactly for solution of this problem it is suggested to use the method of metal magnetic memory (MMM) the main designation of which is SCZs detection based on express control of the entire equipment surface. No preparatory works are required for this.

Most of power equipment units and elements operate in conditions of cyclic loads, and after their long-term operation fatigue and/or creep damages should be expected, which occur, as a rule, unexpectedly in local SCZs.

SCZs are not only the known beforehand areas, where structure peculiarities create various conditions for distribution of stresses due to external operating load, but also are randomly located areas, where large strains (shear strains, as a rule) occurred due to metal inhomogeneity combined with off-design additional working loads.

Paper [3] considers the physics of metal fatigue damaging and offers the development model of this process opening the possibility of material’s state quantitative assessment at using the MMM method.

It is established that fatigue damaging of metal has three phases:

• the first stage is preparatory, it is characterized by comparatively high rate and lasts for comparatively short period of 1,0-1,5% of the limiting number of cycles;

• the second phase - being the basic one - is an accumulative phase characterized by very slow development of propagation process in one direction - to the depth (from units to tens of microns) and lasting very long - for 90-95% of the limiting number of cycles;

• the third phase - is the final one flowing very rapidly and causing occurrence of microcracks in "random sites" and their propagation at a very high speed to the depth and along the length and their growing into macrocracks.

It should be noted that the first two phases of metal fatigue damaging development in conditions of cyclic loading are well investigated, but the third phase, which became a subject of numerous investigations, remains a secret to a great extent.

Fig.1 shows graphic representation of the three phases of metal fatigue damaging accumulation process in the form of dependence of metal weakened (loosened) layer thickness δ on the number of load cycles N.

Paper [3] gives quantitative representation of this diagram based on design investigation for pipe sample in conditions of symmetric cyclic loading along the radius.

Transience of the third phase and uncertainty of damaging initiation site did not allow till date the detailed investigation of events preceding the failure directly on equipment.

Let us further consider the abilities of the MMM method to trace the development of metal fatigue failure process directly on equipment.

The main diagnostic parameter by the MMM method is the magnetic leakage field gradient Нр (dНр/dx) or this field variation intensity coefficient (Кin)¹, registered at scanning with a special magnetometer sensor along the equipment surface. It was established that this very diagnostic parameter due to magnetometric effect directly reflects the energy state of surface and depth metal layers in SCZ [3]. And the maximum field gradient value, determined on the metal surface with accuracy up to one millimeter, corresponds to the source of crack occurrence. In the area of the most intensive strain and, finally, failure process the domain structure suffers sufficient changes. Sizes of domains, whose directions coincide with glide direction, reach critical sizes. As a result, the maximum-size domain "breaks" and a microcrack appears. Design investigations in paper [3] demonstrated that the domain of iron can have a volume covering up to ten grains. At present Energodiagnostika Co. Ltd, due to its large experience in power equipment examination, possesses quantitative values of Кin characterizing metal limiting state by strength conditions and micro- and macrocracks² development.

¹ K_in=|ΔH_p|/Δx where |ΔH_p| is the modular difference of the Н_р magnetic field intensity between the adjacent measurement points located at a distance Δx. At Δx->0, К_in=dH_р/dx.

² It should be noted that cracks division by micro and macrocracks has till date a conventional nature in the technical literature.

Fig.2 shows inspection results of No.29 stage disk rim of No.3 unit K-300-240 turbine average-pressure rotor (APR) at Konakovo TPS (July, 2001). Нр field and its gradient dНр/dx distribution shown in Fig.2, а characterizes the disk rim metal’s state. A "replica" for metal structure investigation was taken in the Нр field and the maximum gradient value local variation on the disk. Fig.2, b shows the photo illustrating the results of metallographic investigation. Microcracks with 1 or 2 microns opening can be seen.

Fig.2a. Results of disk rim inspection. Distribution diagram of the Н_р in SCZ.

Fig.2b. Results of metallographic investigation.

Fig.3, а shows the results of MMM-inspection along the output edge of K-300-240 turbine LPR stage 38 blade No.12, unit No.5 at Konakovo TPS. A "replica" for metal structure investigation was taken in the zone of abrupt local variation of the Нр field and its gradient. Fig.3, b shows the results of metallographic investigation. A microcrack with 5-8 microns opening was revealed in the zone of the field maximum gradient value dНр/dx.

This figure represents as well the results of the Нр field and its gradient measurements in the microcrack zone before grinding (а), after primary grinding for taking a replica (c) and after the secondary grinding (d). In this case grinding on blade No.12 was perform to the depth of about 200 microns. The dark stripe presented in Fig.3, b, corresponding to the crack location on the blade surface, obviously represents the loosened metal layer. The dense metal layer with increased hardness is under the loose layer. The results of experimental investigation of similar blades, presented in paper [4], confirm this.

Fig.3a.	Fig.3b.
Fig.3c.	Fig.3d.

With the increase of the number of load cycles the dense metal layer cracks increasing the size of the loose layer. Thus, a crack develops on the blade surface. The results of experiment with blade No.12 grinding in the developing crack region obviously demonstrate correlation of the diagnostic parameter dНр/dx with the damages metal layer dimensions (depth and width). This correlation is also proved by the physical "magnetodislocation" mechanism of the Нр magnetic field formation in the local SCZ [3].

It should be noted that if the damaged metal layer is not removed in due time, it may cause a serious blade or disk failure during the overhaul life. The rate of a microcrack growing into a macrocrack is obviously different for each specific unit. Such investigations are, as a rule, not carried out in real conditional of critical equipment operation. Application of MMM at a 100% examination of aging equipment will allow detecting crack formation zones and assessing their propagation rate for various power equipment units in future. At present there are some techniques for classification of the degree of metal damaging in SCZs by the field gradient value.

It was established in the course of industrial investigations that the diagnostic parameter values К_in in SCZs for the same-type elements (for example, for blades of the same disk) during the same operation time are, as a rule, different, for instance, due to the different amplitude of cyclic load.

It is known that the damaging development rate in phases II and III for the same-type elements is also, as a rule, different. This rate can be traced by the К_in value. This is the basis of the methodology of life assessment by the К_in value presented in paper [5]. Paper [3] shows that decreasing of metal density in SCZs at accumulation of fatigue damaging is accompanied by increasing of magnetic energy density and by the diagnostic parameter К_in increasing respectively.

The gained experience in a 100% examination of K-300 turbines at Konakovo TPS, К-200 at Cherepovetsk TPS and Zainsk TPS, Т-100 at Severodvinsk TPS-2, PТ-60 and Т-100 at Petrozavodsk TPS and others (more than 50 various types of turbines were examined in total) allows making the following conclusion: SCZs are the sources of damages development (in the form of cracks, as a rule), which occupy not more than 3-5% of the total rotors metal surface or volume. The remaining 95% of the turbine rotors metal volume after their long-term operation are in a satisfactory condition! Thus, the problem of turbine rotors life assessment is solved by timely detection of maximum stress concentration zones and their removal by ordinary grinding in the course of repairs. Similar approach at life assessment with 100% examination by the MMM method is used by Energodiagnostika Co. Ltd. on all types of power equipment: turbines, boilers, steam and water pipelines, etc. And the costs for execution of such an examination are sufficiently lower as compared to the cost of diagnostic works indicated in the Concept of RSC "UES of Russia". For example, examination of all three rotors of К-200 and К-300 turbines will make not more than 1 million rubles (as of January, 2005).

References

1. Dubov А.А. Problems of aging equipment residual lifetime assessment // Teploenergetika, No.11, 2003, p.54-58.

2. GD 10-577-03. Standard instruction for metal control and prolongation of basic power station elements life. Moscow: SPO ORGRES, 2003.

3. Vlasov V.Т., Dubov А.А. Physical bases of the metal magnetic memory method. Moscow: ZAO "TISSO", 2004. 424p.

4. Dubov А.А., Matunin V.М., Ryzhkov F.Е., Chechko I.I. Early diagnostics of blades damaging using the method of metal magnetic memory // Tyazheloye mashinistroyenie, No.10, 2001, p.32-34.

5. Dubov А.А., Dubov Al.An., Kolokolnikov S.М. Method of metal magnetic memory and inspection devices. Training Handbook. Moscow: ZAO "TISSO", 2008. 365p.