This paper has been presented at the 6th EPRI Steam Turbine/Generator Workshop and Vendor exposition - St Louis, Missouri, USA - August 1999
Two field trials on Darlington unit #3 and #4 were performed in January and May 1999, for in-situ ultrasonic examination of L-0 and L-1 blade roots and rotor steeples. The ultrasonic system uses automated and manual phased array technology, capable of high-speed rate and reliable detection and sizing. The system capability was demonstrated on 1:1 scale mock-ups and reference blocks, using EDM notches. Targets as small as 2 x 0.5 mm on steeple hooks and 3 x 1 mm on blade roots could be detected and reliable sized. A custom built UT simulation software: Imagine 3D interfaces with SIMSCAN to produce 1:1 2-D views and generates the spreadsheets with target and probe coordinates and ultrasonic path and angles (refracted and skew) to hit that target. The Ray-tracing simulated results were validated by experimental measurements. Examination of L-0 blade and rotor steeple grooves was performed with 2 phased array Focus systems under networking. Data analysis was done in real time. Manual phased array was performed on L-1 blade roots (hook 1). Special linear array probes were designed by OPGI and manufactured by Imasonic. They performed very well during the 10-days examination period. Accuracy in sizing is ¡¾ 0.3 mm for height, ¡¾ 1 mm for length. Location of indication in 2-D specimen layout is within ¡¾ 2 mm space envelope for blade root, and within ¡¾ 4 mm for rotor steeple. New developments under way will lead to a better sizing and plotting, to an increase speed and to a reduced file size. Representative results from the 2 field trials/examinations will be presented. The system(s) will be full-speed commissioned by May 2000.
Darlington NGS and
Specialized Examination and Maintenance Department (SIMD) funded a
4-year project to build and commission an automated ultrasonic system
capable to in-situ examine ABB- type blade roots and rotor steeple
grooves of low-pressure turbine rotors (see Figure 1).
RD Tech- Québec-Canada developed Focus phased array machine and associated software-Tomoview, based on SIMD requests. Imasonic-Besançon/France manufactured the linear phased array probes (LPAP), most of them with unique features (prototyping). The feasibility study results were published in EPRI Steam Turbine Workshop - July 1997-phase 1 (1), and EPRI Phased Array Seminar in Portland – Sept. 1998 – phase 2 (2). Since last fall, SIMD developed the new LPAP with hard face and reduced contact area, built wing and steeple manipulators, implanted more realistic targets and developed examination procedures for different configurations. Two field trials were performed in January and May 1999. Figure 1: L-0 blade (a), steeple (b), L-1 blade(c). |
The January field trial objectives were:
The May field trial / examination objectives were:
The paper will detail some aspects of proof-of-principle examinations of low-pressure turbine components.
Reference Blocks and Ultrasonic Equipment
Capability demonstration of phased array system in automated and manual modes was performed on the following reference blocks (see Table 1):
Reference block ID | Nr. of EDM notches | Reference
target size [ mm ] |
Remarks |
L-0 Blade Roots |
|||
RD Tech #3 |
36 |
3 x 1 | T2 / Jan. 1999 |
RD Tech #1 |
37 |
3 x 1 / 4 x 1.5 | T1/T2 Jan. 1999 |
Silver Blade |
34 |
5 x 1 | T3 / May 1999 |
SP-1 |
38 |
4 x 1.5; 3 x 1; 5 x 1 | T1 / T2 Jan 1999
T3 May 1999 |
SP-2 |
No targets |
- |
Comparison |
L-0 Rotor Steeple Grooves |
|||
ABB Mock-up |
35 |
3 x 1; 4x1.5;
9 x 0.45; 9 x 1 |
Jan. 1999
May 1999 |
Steeple Block |
39 |
3 x 1; 4x1.5;
9 x 0.45; 9 x 1 |
Jan. 1999
May 1999 |
L-1 Blade Root – Hook 1 |
|||
L-1 Block |
26 |
3 x1; 2 x 0.5; 4 x 2 | Manual UT/ May 1999 |
AC 878 |
5 |
3 x 1 | Manual/Automated UT
May 1999 |
Table 1: Reference blocks for capability demonstration of different techniques.
Ultrasonic equipment consists of the following items:
Probe ID |
Fn / Nr. Elements/ Size |
Type of wave |
Remarks |
2 / 2B |
10 / 20 / 6x6 mm |
L, T |
L-0 Blade T3 –
May 1999
L-1 Blade T1 – May 1999 L-0 Steeple T2 – May 1999 |
3 / 3B |
10 / 32 / 7 x 10 mm |
L, T |
L-0 Blade T1 –
Jan. 1999
L-0 Steeple T2 – Jan. 1999 |
3C – hard face |
11 / 32 / 7 x 10 mm |
L |
L-1 Blade T1 - May 1999 |
5 |
5 / 32 / 10 x 32 mm |
L |
L-0 Steeple T2 - Jan. 1999 |
7 |
7 / 16 / 16 x 14 mm |
L |
L-0 Blade T1 - Jan. 1999 |
8 – hard face |
11 / 32 / 4 x 10 mm |
L |
L-1 Blade T1 - May 1999 |
10 – hard face |
12 / 16 + 16 / 5 x 5 mm |
L |
L-0 Steeple T2 - May 1999 |
Table 2: LPAP used in two field trials / in-situ examinations.
SIMSCAN / IMAGINE 3D Simulation Results.
O.P.G.I. in cooperation with UTEX Scientific and NDE Focal Point Technologies developed a simulation software package Imagine 3D and SIMSCAN. The scope of this development was to validate the focal laws, get the ultrasonic path, probe trajectory and refracted/skew angles for different examination scenarios. The physical dimensions of LPAP were incorporated into model. A 1:1 print display of 2-D with simulated UT rays was also developed. The simulation data were validated by experimental results on SP-1 blade, on reference blocks.
Figure 2 and 3 represents the simulation displays of L-0 blade root (hook 1) T3, and rotor steeple (hook 5). The results could be display into a spreadsheet and representative charts are used for procedure development.
Figure 2: SIMSCAN results for T3- CVX, L-0 blade. | |
Figure 3: Simulation results for T2, CVX L-0 Rotor steeple, hook 5. |
An example of correlation between refracted angle and probe movement is presented in Figure 4 and 5.
Figure 4: SIMSCAN results for refracted angles L-waves – T1 –L-0 blade-inlet-platform. | |
Figure 5: SIMSCAN results for refracted angles T-waves on convex side, L-0 blade-hook 1 |
SIMSCAN / Imagine 3D were used to optimize the probe trajectory for manual phased array /mono-crystal of L-1 blade root inspection. The simulation results were incorporated into examination procedures. SIMSCAN / Imagine 3 D provide examination layouts for all 3 objects (L-0 blade, L-0 steeple, L-1 blade). An example of dependence of refracted angles on probe position is presented in Figure 4a and b. More details about ray-tracing simulation could be found in (3).
Ultrasonic Results / Field TrialThe field trial from January 1999 was focused on the following examinations:
L-0 blade root
- a combination of side and top raster scanning with 2 LPAP, which could cover 125 mm from inlet and outlet
- the reference reflectors were: 3 x 1 mm for top technique ( platform and 4 hooks)
- the reference reflectors were 6 x 1.5 mm for side technique ( platform and 2 hooks)
- nr. of scanning files: 16, with a scanning time of 5-7 minutes / file.
- cascade of focal depth to cover the range 25 mm – 125 mm (3-4 groups)
- size of T1 scanning file: 80 – 150 Mbytes
L-0 rotor steeple
- create 2 groups of focal laws for 2 different LPAP to compensate for manipulator rail off-set vs. concave/convex sides radius
- run a line scanning on each side for both hooks, from 120 – 360 mm
- the reference targets for Hook 1: 3 x 1 mm, and for Hook 2 : 4 x 1.5 mm;
- nr. of scanning files = 2, with a scanning time of 10 minutes / file
- two groups of focal laws to cover H1 and H2 depths
- size of T2 scanning file: 20 – 35 Mbytes
The ultrasonic results for probe 2, 3, 5 and 6 on more than 30 targets concluded:
- detection capability: 2 x 0.5 mm in T2 mode
- detection capability: 4 x 1.5 mm in T1 mode – range 25 – 50 mm
- detection capability: 5 x 1 mm in T1 mode – range 50 – 125 mm
- sizing capability: 3 x 1 mm in T2 mode (¡¾ 0.5 mm for height, ¡¾1 mm for length) – range 0 – 30 mm;
- sizing capability: 4 x 1.5 mm in T1 mode, range 45 – 125 mm
The proof-of-principle – phase 2 from May 1999 was focused on "make-sense" inspection strategy, expand manual phased array/mono-crystal examination to L-1 blade and improve the quality of UT data, increase the productivity by using portable laptops for acquisition, improve the manipulator¡¯s performance (see Fig. 6), and use hard-face probes. The following examinations were performed:
L-0 blade root
- raster examination of 100 – 340 mm on platform and hook 1
- reference reflector: 5 x 1 mm
- nr. of scanning files: 2
- scanning time: 5-7 minutes / file
- size of scanning file: 50 - 65 Mbytes
Examples are given in Figure 7 and 8.
L-0 rotor steeple
- line scanning from 50 mm – 440 mm
- reflector size: 9 x 0.45 mm for H1, and 9 x 1 mm for H2, irregular shape
- nr. of scanning files = 2
- scanning time: 3-5 minutes / file
- size of scanning file: 10-15 Mbytes
Examples are given in Figure 9-12.
L-1 blade root
Conclusions
- manual phased array / mono-crystal for 0 – 40 mm, hook 1-concave side, inlet/outlet
- manual phased array for 60 – 120 mm, hook 1-convex side
- reference target: 3 x 1 mm
- use templates for probe trajectory
- use T-waves for better sizing
Examples are given in Figure 13-15.
Detection capability for T3, L-0 blade root is 3 x 0.5 mm. Targets as small as 9 x 0.15 mm and 2 x 0.5 mm could be detected on H1-rotor steeple. Targets of 0.5 mm height could be detected with an accuracy of ¡¾ 0.3 mm. Reflector length is measured with an accuracy of ¡¾1 mm, and their location is within ¡¾2 mm for blade roots and ¡¾4 mm for steeple.
The field trials from January and May 1999 proved to be successful. Significant improvements were achieved, and the quality of examinations increased. More than 700 items were examined in the critical areas. The combination between manual UT and automated phased array produced reliable results, and a high productivity. Significant improvements were identified, and corrective actions are under way. The ultrasonic results proved the phased array examination has a redundant information regarding the detection, sizing, and defect shape. The system(s) will be fully commissioned by May 2000.
The authors wish to express their thanks to O.P.G.I. – S.I.M.D., and DNGS management, for granting the publication of this paper. One of the authors (P. Ciorau), wishes to express his thanks to the following peoples: Alex Manukian and Gery Proulx- AEP-USA, Mike Lundey and Phil Knox – Florida Light & Power-USA, for their support to this project.
A. Lamarre, N. Dubé, P. Ciorau, and B. Bevins: "Feasibility study of ultrasonic Inspection using phased array of turbine blade root" – 5-th EPRI Workshop Steam Turbine/Generators – July 1997.
P.Ciorau, D. MacGillivray, A. Lamarre, and F. Jacques: "Feasibility study of ultrasonic inspection using phased array of turbine blade root and rotor steeple grooves-part 2" – EPRI Phased Array Seminar-Portland, Sept. 1998.
D. Mair, P. Ciorau, D. Owen, and T. Hazelton: "Ultrasonic Simulation – Imagine 3D and SIMSCAN. Tools to solve the inverse problem for complex turbine components"- to be presented to 26-th QENDE Annual Review of Progress July 25-30, Montreal, 1999.
Petru Ciorau, Doug MacGillivray, Trek Hazelton, Lionel Gilham, Dale Craig – ONTARIO POWER GENERATION Inc. ,Pickering – Canada