Construction of Steel Penstocks using HT100 at Kannagawa Hydropower Plant

Construction of Steel Penstocks using HT100 at Kannagawa Hydropower Plant Ken-ichiro AOKI 1 and Masayuki MINAMI 1 1 Tokyo Electric Power Company, Japan Introduction Tokyo Electric Power Company s Kannagawa Hydropower Plant is located in Gunma prefecture, Japan. For the first challenge in Japan, high tensile strength steel (HT100) with a tensile strength of 950 N/mm 2 was adopted for use in the penstocks of Kannagawa Hydropower Plant to reduce the construction cost. To apply HT100 steel for the penstocks, a performance verification test was conducted in advance to make sure that HT100 steel plates and welded joints have required performance, while the requirements for welding procedures were determined. In the actual construction, quality control was carried out adequately on the basis of the determined requirements for welding procedures and other requirements. This paper presents the design and construction of the penstocks of the Kannagawa Hydropower Plant. In the paper, we primarily focus on quality control items specified in consideration of the characteristics of high tensile strength steel and their results. surge tank, into the penstock for units 1 & 2 with a total length of 1,397 m and the penstock for units 3 & 4 with a total length of 1,365 m. Tokyo Electric Power Company decided to introduce high tensile strength steel (HT100) with a tensile strength of 950 N/mm 2 for the penstocks of its Kannagawa Hydropower Plant to reduce the construction cost. Table 1 shows the materials used and their weight. The total weight for units 1 & 2 penstock was about 5,154 t, of which the quantity of the HT100 steel was about 2,331 t. The total weight for units 3 & 4 penstock will be about 5,150 t, of which the quantity of the HT100 steel will be about 1,420 t. The construction of the penstock for units 1 & 2 was started in April 2001, with the installation on the site completed in July 2004. Currently, the inclined shaft and part of upper and lower horizontal sections of the penstock for units 3 & 4 are under construction with the progress rate being about 40% as of May 2009; the installation of the penstock is planned to be completed by April 2010. 1. Overview of Facilities Tokyo Electric Power Company s Kannagawa Hydropower Plant is a pure pumped-storage power plant with an effective head of 653 m generating a maximum output of 2,820 MW with six motor-generators. Its unit 1, with a maximum output of 470 MW, was put into commercial operation in December 2005. The steel penstocks of the Kannagawa Hydropower Plant are embedded in rock mass, extending over a length of about 1,400 m with a difference in elevation of 745 m and having a maximum design head of 1,076 m, of which the static head is 817 m, and inside diameters ranging from 2.3 m to 8.2 m. The steel penstocks consist of upper horizontal section, inclined shaft section, and lower horizontal section. The inclined shaft section of the entire penstock structure is an about 960m long steep gradient pressure conduit, with an inclination of 48 degrees and an inside diameter of 4.6 m. Figure 1 shows a schematic diagram of the steel penstocks of the Kannagawa Hydropower Plant. The steel penstock is bifurcated, at the Branch I located at the base of the headrace Figure 1: Schematic diagram of the steel penstocks of the Kannagawa Hydropower Plant 1-1

TABLE 1: Materials and weights of steel Steel Penstock for units 1 & 2 Penstock for units 3 & 4 4) Thick ness (mm) Weight (t) Ratio of weight Thick ness (mm) Weight (t) Ratio of weight HT100 29-72 2331 45% 27-72 1420 28% SHY685 28-38, 28-42, 62 1) 312 62 1) 907 NS-F 6% 200 2) 8 200 2) 8 18% SM570 22-45 1335 26% 22-44 1541 30% SM490B 20-52, 22-52, 70 3) 1168 23% 70 3) 1148 22% SM400B ---- ---- 0% 22-26 126 2% Total weight ---- 5154 100% 5150 100% 1) 62-mm thick steel is for joining inlet valves. 2) 200-mm thick steel is for stiffening the branch II. 3) 70-mm thick steel is for reinforcing the branch I. 4) The penstock for units 3 & 4 is now under construction, so the values are estimated. TABLE 2: Allowable stress and yield point of steels Steel Thickness Allowabl Yield Standard (mm) e stress point (N/mm 2 ) (N/mm 2 ) HT100 <75 400 885 Technical Guidline for HT100 SHY685NS- F <50 50-100 200 SM570 16-40 40-75 SM490B 16-40 40-100 SM400B 16-40 40-100 330 320 315 240 235 175 160 130 115 685 665 650 450 430 315 295 235 215 JIS G 3128 JIS G 3106 JIS G 3106 JIS G 3106 2. Design of the Penstocks 2.1 Design for Internal hydraulic pressure In the design of penstocks embedded in rock mass, internal hydraulic pressure is partially borne by rock mass surrounding penstocks with the aim of reducing costs through a reduction in the thickness of steel pipes. In this design, the ratio of internal pressure borne by rock mass is designed according to rock mass properties, such as the elastic modulus and plastic deformation coefficient. The plate thickness of the penstock is determined so that the stress acting on the penstock as a result of the consideration of the internal pressure borne by rock mass may not exceed the allowable stress of the penstock. In addition, the design is established on the safe side so that the stress acting on the penstock may not exceed the yield point of the steel material even if rock mass is assumed to bear no internal hydraulic pressure. Table 2 shows the allowable stress and yield point for the steel material used. Joint efficiency is determined for the places of welding and the sampling ratio in the nondestructive inspection, and can be made 100% by welding in a workshop and inspection over the entire welded length. Tokyo Electric Power Company compared costs of nondestructive inspection over the entire welded length with reducing costs through a reduction in the thickness of steel pipes assuming a 100% joint efficiency. For the penstocks used in the Kannagawa Hydropower Plant, a reduction in the plate thickness was adopted by assuming a 100% joint efficiency after the inspection of entire longitudinal joint lines. 2.2 Design for External hydraulic pressure To ensure safety for external hydraulic pressure, penstocks embedded in rock mass are designed to have a safety factor of 1.5 or more against the critical buckling pressure calculated using the theoretical formula. For inclined shafts of which an adequate drain slope can be expected, Tokyo Electric Power Company adopts a design in which drain facilities are installed to reduce external hydraulic pressure. Figure 2 shows the design of the drain facilities of the Kannagawa Hydropower Plant. The drains consist of both a direct drain for draining water from the void between the penstock and the infilling concrete and an indirect drain for draining seepage water, which comes from the rocks, from the void between the concrete and the rocks. Even when one of the drains becomes ineffective, the other reduces the external hydraulic pressure to prescribed level. The groundwater level in the natural ground surrounding the penstocks is assumed to be on the grand level according to investigation. The drain facilities are designed to reduce the groundwater potential near the penstock in the inclined shaft section. For the penstock for units 1 & 2, the design external hydraulic pressure was set at 40% of the hydraulic pressure corresponding to the natural ground cover on the basis of the two-dimensional seepage flow analysis. For the penstock for units 3 & 4, on the other hand, with the aim of reducing plate thickness, the design external hydraulic pressure was set at 30% of the hydraulic pressure corresponding to the natural ground cover because a ratio of the measurement value for the design external hydraulic pressure of the penstock for units 1 & 2 was very small. 1-2

Figure 2: Design of the drain facilities 2.3 Results of the Design In designing the material and plate thickness of penstocks, the plate thickness was decided as follows; First, the largest plate thickness, for each unit section (with a length of 3 m), was calculated by each candidate material among ; (1) plate thickness dictated by the internal hydraulic pressure, (2) plate thickness dictated by the external hydraulic pressure, and (3) the smallest plate thickness dictated by the restrictions in construction. Then, the most cost-effective material was decided with the costs for materials, manufacturing and installation taken into account. In addition, while welding workability was being given consideration to avoid an abrupt change in plate thickness and an extreme change in the material between adjoining unit sections, materials and design plate thickness were adjusted adequately before determining the final design. Figure 3 shows the results of the design of the penstock (for the inclined shaft) of the Kannagawa Hydropower Plant. For the penstock for units 3 & 4, the plate thickness of HT100 steel is smaller than that of the penstock for units 1 & 2 as a result of a review of design conditions such as a change in the design external hydraulic pressure. Penstock for units1&2 Penstock for units 3&4 1SM400 2SM490 3SM570 4SHY685 5HT100 Figure 3: The design of the penstock (for the inclined shaft) 1-3

3. Procedure of the penstock installation Figure 4 shows an overview of the penstock installation of the Kannagawa Hydropower Plant. The penstocks were constructed by cutting steel plates, preparing edges and bending the plates in manufacturer s plant. And then, due to transportation limits, these were constructed by transporting the half-pipe parts to the site, welding the parts into 3 m-long unit pipes and then into 15 m-long unit pipes in the temporary workshop at the site, and installing in the inclined shaft and welding the unit pipes. The installation of steel penstock in the inclined shaft was a critical path of the construction schedule. Since the adoption of HT100 steel allowed one to reduce the penstock weight in the inclined shaft section, the length of an installation unit pipe was increased from 12 m (four 3-m long unit sections), the longest in the past, to 15 m (five 3-m long unit sections), thereby reducing the number of times of installation in the inclined shaft and hence reducing the time necessary for installation by 20%. In the inclined shaft, a construction method that was based on the installation of two unit pipes taken as one work cycle was adopted. Consequently installing steel a unit pipe and filling concrete helped shorten the construction period. The installation speed of HT100 steel sections including concrete filling was 11 days/30 m. The mean installation speed of the entire inclined shaft was 10 days/30 m. Regarding the weld processes adopted for the penstocks at the site, the SAW(submerged arc welding) method was adopted for straight pipes and the SMAW(shielded metal arc welding) method for bend pipes at the temporary workshop, while the method of automatic MAG(metal active gas) welding from one side of the inner surface was adopted for the inclined shaft, with the SMAW method adopted for other on-site welding operations. 4. Quality Control for the Penstocks 4.1. Overview of Quality Control for the Penstocks To apply HT100 steel for the penstocks, Tokyo Electric Power Company primarily focused on the prevention of brittle fracture, while quality control items and criteria for steel plates and welded joints were determined, especially in consideration of toughness and weldability. Figure 5 shows the penstocks installation procedure and important inspection items of quality control for the penstocks of the Kannagawa Hydropower Plant. Tokyo Electric Power Company s approach to quality control is fundamentally based on both the checking of all records of inspection by the fabricator and witness sampling inspection. In addition to this, specifically for important inspection items, the witness inspection of all articles is predominantly used. In the following, the important items and results of quality control are described using the pieces of data obtained during the construction on the penstock for units 1 & 2 for examples. Place Steel mill Workshop of fabricator Temporary workshop Inclined shaft Figure 4: Overview of the penstock installation Steel penstock installation procedure Material making Half pipe manufacturing ( transportation ) Unit pipe of 3m fabricating Assembly Longitudinal welding Unit pipe of 15m fabricating Asssembly Circumferential welding Painting Installation Assembly Circumferential welding Completion Inspection Material Inspection Half pipe Inspection Receiving&Inspection Groove Inspection Visual Inspection Nondestructive Inspection Groove Inspection Visual Inspection Dimension Inspection Nondestructive Inspection Before painting Inspection After painting Inspection Groove Inspection Visual Inspection Nondestructive Inspection Figure 5: Important inspection items of quality control for each procedure 1-4

4.2 HT100 Steel Plates and Welding Materials Table 3 and Figure 6 show the quality control criteria and test results for HT100 steel plates. Generally tensile strength tests are conducted using C-direction of the steel plate. However, the preliminary performance verification test revealed that the L-direction strength bearing hydraulic pressure tended to be smaller than the strength in the C-direction in HT100 steel. In the actually executed work, for this reason, the tensile strength test was conducted both in the L- and C- directions. Since the impact test also showed that the toughness near the middle of the plate thickness tended to be inferior to that at the 1/4-thickness position, the toughness was checked at positions of 1/4-thickness and 1/2-thickness in the actual construction. Table 4 shows the mechanical properties of HT100 deposited metal and the control standard for the quantity of hydrogen. The specifications of the deposited metal were characterized by tensile strength for which slightly softer nature than the base metal was required in consideration of the balance with the toughness and weldability and by the requirement for the quantity of hydrogen. TABLE 3: The quality control criteria and test results for HT100 steel plates Plate thickness <= 50mm Plate thickness > 50mm n=41 n=12 Element C P S Ceq PCM C P S Ceq PCM Maximum 0.11 0.008 0.001 0.572 0.273 0.12 0.006 0.001 0.603 0.311 Minimum 0.09 0.001 0.001 0.533 0.254 0.11 0.002 0.000 0.592 0.288 Maximum content 0.14 0.01 0.005 0.59 0.29 0.14 0.01 0.005 0.62 0.33 Ceq=C+Mn/6+Si/24+Ni/40+Cr/5+Mo/4+V/14 PCM=C+Si/30+Mn/20+Cu/20+Ni/60+Cr/20+Mo/15+V/10+5B Figure 6: The quality control criteria and test results for HT100 steel plates 1-5

TABLE 4: Mechanical properties of HT100 deposited metal and the control standard for the quantity of hydrogen. Welding method SMAW SAW MAG (Reference) HT100 steel plate Yield strength 785 785 785 885 (N/mm 2 ) Tensile strength 930-1000 930-1000 930-1050 950-1130 (N/mm 2 ) Elongation (%) 12 12 12 12 Charpy impact absorption energy (J) Ductile fracture percentage (%) Hydrogen content (ml/100g) 47 (at 10 ) 47 (at -10 ) 47 (at -10 ) 47 (at -55 ) 50 50 50 50 6 3 2 ---- 4.3 Welding Procedure Control Table 5 shows the requirements for welding procedures. The requirements for welding procedures for HT100 steel were specified on the basis of the results of the preliminary weld cracking test. The results of the weld cracking test showed the following findings: a. In single pass welding, the preheat temperature for preventing weld cracking was similar or higher in a U-groove weld cracking test than in a y-groove weld cracking test. Thus both tests should be conducted to determine preheat temperature for HT100. b. A multi-layer weld cracking test showed that postheating (150 C for 2 hours) was highly effective for preventing cracks. In SHY685NS-F, postheating could be omitted when a nondestructive test was conducted at over 48 hours after the completion of welding operation. However, postheating was indispensable for preventing cracks for HT100 when preheat temperature was similar to that for SHY685NS-F.; and c. Use of welding materials for SHY685NS-F in single pass welding by SMAW and MAG enabled preheat temperature for preventing cracks to be lowered. Thus, use of welding materials for SHY685NS-F in SMAW and MAG for root pass welding and tack welding was effective for preventing cracks. For SM570 and SM490B, a special specification of weld crack sensitivity PCM against weld cracking of not exceeding 0.2% was also applied, and preheating was omitted except on cold days. Table 6 shows the track records of heat input control for HT100 welded joints. Compared with the control standard of 45 kj/cm, the average heat input in the actual construction is about 34 kj/cm in the SAW and SMAW methods and about 16 kj/cm in the automatic MAG method applied in the inclined shaft. In controlling the welding atmosphere, the water vapor pressure was calculated from the results of measurements of the ambient temperature and humidity, and the relationship between the values obtained and the control standard was used in order to limit the time for which the welding materials were left alone. Figure 7 shows the track records of the control of water vapor pressure at the temporary workshop. Steel HT100 and SHY685NS-F Thickness (mm) Not exceeding 50 TABLE 5: The requirements for welding procedures Minimum preheating temperature ( C) SMAW MAG SAW 100 80 Over 50 125 100 Interpass temperature At least the preheating temperature and not exceeding 230 C Limit heat input (kj/cm) The mean not exceeding 45, and the maximum not exceeding 50 Postheating conditions At least 2 hours at over 150 C SM570 Not No preheating ---- Not exceeding 60 ---- exceeding 25 Over 25 ---- Not exceeding 80 ---- SM490B ---- ---- ---- ---- SM400B ---- ---- ---- ---- a. For SMAW and MAG for HT100, welding material for SHY685NS-F was used for root pass and tack welding. b. For tack welding of HT100 and SHY685NS-F, welding materials of a strength level that was one rank lower than that for the actual welding. c. Postheating was omitted for SHY685NS-F if nondestructive test was conducted 48 hours after welding or later. d. The preheating temperature for tack welding was at least 25 C higher than the minimum preheating temperature stated above. e. In SMAW and SAW, the preheating temperature was at least 25 C higher than the minimum preheating temperature stated above when the partial hydraulic vapor pressure was over 25.5 mmhg. f. For SM570 and SM490B, steels with PCM of not exceeding 0.2% was used and no preheating was conducted provided that the temperature was preheated to 40 C when the air temperature was 5 C or below. g. When welding was suspended, postheating was preformed using the conditions stated above or the temperature was retained until welding was restarted. 1-6

Water vapor pressure(mmhg) TABLE 6: The track records of heat input control for HT100 welded joints Place of welding Welding method Number of joints Pass maximum heat input (kj/cm) Average heat input (kj/cm) Max. Min. Average Max. Min. Average Temporary SAW 365 48.4 30.9 39.1 39.8 27.7 32.8 workshop Temporary SMAW 20 48.4 37.9 45.0 41.1 27.2 36.3 workshop Field SMAW 87 49.1 26.6 39.9 38.9 19.9 32.0 Inclined shaft MAG 25 37.6 24.6 30.1 18.5 13.1 16.4 Control standard Not exceeding 50 Not exceeding 45 40 30 Contorol Standard:25.5mmHg 20 10 0 2001/11/1 2002/1/31 2002/5/2 2002/8/2 2002/11/1 2003/1/31 2003/5/3 2003/8/2 2003/11/1 2004/2/1 2004/5/2 2004/8/1 Date Figure 7: The track records of the control of water vapor pressure at the temporary workshop 4.4 Offset and Angular Distortion of Welded zones From the viewpoint of preventing brittle fracture, checking the shape of welded zones inducing the stress concentration is one of the most important control items. In the actual construction, quality control was conducted on undercuts, overlaps, reinforcement weld heights, offsets, angular distortions and so forth. Among these, the offset and the angular distortion of a longitudinal joint were put to quantitative control. Figure 8 shows the track records of control of offsets and angular distortions of welded zones of HT100. Offset ratio for thickness(%) Angular distortion( ) 6.0 5.0 4.0 3.0 2.0 1.0 0.0 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Control Standard 5% 20 30 40 50 60 70 80 Plate thickness(mm) Control Standard 2.5 3mm 20 30 40 50 60 70 80 Plate thickness(mm) Figure 8: The track records of control of offsets and angular distortions of welded zones of HT100 4.5 Nondestructive Inspection The Japanese standards require that welded zones in the main pressure bearing parts of a penstock be put to nondestructive inspection after welding operation as needs arise, and that the methods of nondestructive inspection be based on the radiographic test or ultrasonic test as a rule. Table 7 shows examples of the criteria (for lineal flaws) for the radiographic test and ultrasonic test. (1) Nondestructive test on the penstock for units 1 & 2 The nondestructive inspection on welded zones of the penstock for units 1 & 2 of the Kannagawa Hydropower Plant was conducted according to the following policies in consideration of the track records of inspection on penstocks: a. The main pressure bearing parts should be put to the radiographic inspection as a rule; and b. Branch pipe shells and reinforcement beam connections to which the application of the radiographic inspection was difficult due to structural problems should be put to the ultrasonic test. TABLE 7: Examples of the criteria (for lineal flaws) for the radiographic test and ultrasonic test Thickness t (mm) Not exceeding 50 Over 50 RT (for linear flaws) Not exceeding t/3 and 16 mm Not exceeding t/4 and 12 mm UT Not exceeding t/2 and 30 mm 1-7

Table 8 shows the rules for the time to conduct nondestructive tests on the penstock for units 1 & 2. In consideration of the occurrence of delayed cracking in high tensile strength steels having performance equal or superior to that of SHY685NS-F, the time to conduct nondestructive test was specified. Table 9 shows the track records of nondestructive tests on the penstock for units 1 & 2. Nondestructive tests over entire lines were conducted on longitudinal joints bearing hydraulic pressure with the aim of rationalizing the design by means of setting the joint efficiency at 100%. For branch pipes with complicated structures that require a high welding skill, nondestructive tests were conducted on entire weld lines. For HT100 steel circumferential joints, entire line inspection was conducted at the beginning of welding operations, and the rate of inspection was reduced to about 10% after the quality of welding had been verified. These inspections allowed one to make sure that all welded joints satisfied the criteria. tester. Figure 9 shows how the ultrasonic test is carried out. With the tester introduced this time, flaws detecting operation is carried out from both sides on one face of a welded zone simultaneously. First, flaws are searched for coarsely, square by square, at intervals of 5 mm, and if reflected echoes are detected, flaw detection is carried out finely at intervals of 1 mm again. All items of obtained data are recorded in the form of electronic data, and the judgment of data on flaws is carried out automatically. Rules for the time to conduct nondestructive inspection for the penstock for units 3 & 4 (Table 8) and the rate of execution of nondestructive inspection on longitudinal joints (conducted over entire lines) are the same as those for the penstock for units 1 & 2. TABLE 8: Rules for the time to conduct nondestructive tests Steel Place of welding Time of testing HT100 All places At least 24 hours after the completion of welding SHY685 Inclined shaft At least 24 hours NS-F after the completion of welding SM570, SM490B SM400B Temporary workshop At least 48 hours after the completion of welding Postheating conditions 150 C for at least 2 hours 150 C for at least 2 hours None All place No rules None TABLE 9: The track records of nondestructive tests HT100 General penstock body Longitudinal joint Circumferential joint Entire length 57% of the welded length 1034.737(m) 1560.779(m) JIS standard materials General penstock body Longitudinal joint Circumferential joint Entire length 12% of the welded length 2090.011(m) 670.173(m) Branch II Entire length 129.906(m) Branch I Entire length 281.031(m) (2) Nondestructive test on the penstock for units 3 & 4 With the aim of increasing the work efficiency and decreasing construction costs, the ultrasonic test based on an automatic ultrasonic tester was adopted for the nondestructive inspection on welded zones in the construction of the penstock for units 3 & 4 of the Kannagawa Hydropower Plant. Before introducing an automatic ultrasonic tester, tests were conducted on artificially flawed test pieces to verify the performance of the Figure 9: Automatic ultrasonic test equipment Conclusion First challenge of introducing HT100 steel with tensile strength of 950 N/mm 2 to penstocks of a hydropower plant in Japan was started at Kannagawa Hydropower Plant. Especially in case of utilizing new materials, quality management is the key issue to complete the project without fatal defects since many unexpected phenomena might happen and counter measures should be taken immediately for completion of the project. In Kannagawa, therefore, quality management system was established by close cooperation between Tokyo Electric Power Company as the owner, engineer and fabricator with support of steel mill. At the any stage of the construction, quality control was been made jointly. Tokyo Electric Power Company s approach to quality control is based on deciding control items and criteria, carrying out the witness inspections, and analyzing their results. The quality control was conducted from the viewpoint of not only judging the results for control criteria but also improving control criteria and our design. As the result of these efforts, the construction of the penstock for units 1 & 2 completed successfully and no irregular stress and strain has been observed by measuring system since starting operation of unit 1 in December 2005. The construction of the penstock for units 3 & 4 is now on going, and the effort for quality management will be also continuing until completing the construction with achieving targeted quality successfully. 1-8