·Table of Contents
·Methods and Instrumentation
Special linear phased array probes used for ultrasonic examination of complex turbine componentsJérôme Poguet - Imasonic - France
Petru Ciorau - Ontario Power Generation Inc. - Canada
Gérard Fleury - Imasonic - France
A large variety of linear phased array probes (LPAP) [size, frequency, bandwidth, multi-heads, pitch size] were designed and manufactured for examination of complex-shape turbine components, such as blade roots, rotor steeples, narrow-gap welds, disk blade rim attachment. The probes were designed and characterised for the following features: geometric dimensions, centre frequency, pulse duration, beam features (focal depth, beam spread in both directions, signal-to-noise ratio). Multi-head LPAP were manufactured to optimise detection and sizing for limited accesses contact areas. Examples of probe characterisation, detection and sizing of artificial and natural defects are given for the following applications: blade roots, rotor steeple, and disk rims. Examples of probe beam features are also presented. Some aspects of Imasonic and OPG QA program for phased array probe characterisation will be illustrated
About 20 phased array probes have been specifically developed during the last 3 years.
Phased array technology
Phased array technology has been selected for the beam steering and focusing flexibility and also for the possibility to angle the beam without wedge that would require too much space for some inspections.
Piezocomposite technology was preferred for the phased array technology and also to optimise the electroacoustical performances like sensitivity, signal to noise ratio, bandwidth and pulse length.
High frequency miniature probes
All probes are high frequency probes (from 6 MHz to 17 MHz) for high-resolution inspection.
Miniature probes were necessary to inspect the complex geometry and limited access components.
Further to the electroacoustical performances, OPG and Imasonic put emphasis on the reliability of the probes for their use in the industrial field
The modelling of the probe is done by O.P.G., with PASS software.
The target parameters of this modelling are inspection depth, lateral and axial resolution, according to various inspections angles.
This modelling work allows the definition of the phased array probes, and particularly
Imasonic checks this design with in house software QuickSonic that allows a very quick check of the basic parameters of a linear phased array probe
If necessary, Imasonic may also confirm the design with CIVA software, developed by CEA (France). This very complete software allows accurate beam simulation and interaction with defined defects.
|Fig 1: Schematic representation of a piezo-composite plate with a 1-3 structure - W.A. SMITH|
The piezocomposite material
A specially designed piezocomposite material 1-3 structure (see Figure 1) is implemented in each type of probe. Details could be found in [ 5 ] .
The composite components and geometry are defined according to the temporal and frequency response specifications, while keeping high sensitivity and signal to noise ratio level.
The piezocomposite material is also designed to lower the cross coupling between neighbour elements, which is necessary to properly steer the beam with electronic delay laws. Typical cross coupling is lower than -40dB.
The matching layer
Taking into account the using conditions of the probe (manual, automated, direct contact, contact with a wedge), the matching layer is designed to optimise the energy transfer, to shorten the pulse length and to be wear resistant.
The backing material
The backing material is designed to shorten the pulse length and attenuate the back echo. Specially designed backing materials allow interesting compromise with high damping and high attenuation in reduced dimension
The cable performances are also a key parameter for the overall performances of the probe. Its attenuation must be as low as possible, mainly for high frequency probes. Its electrical impedance is matched to probe and electronic characteristics.
The bending capability is optimised to access small areas, while keeping high mechanical resistance and constant electrical properties.
The electro-acoustical influence of all these components is simulated with software based on KLM model. The target parameters are temporal and spectral response, sensitivity and electrical impedance (see Figure 2).
|Fig 2: Electro-acoustical design with KLM Model|
OPG issued mechanical specifications for each probe based on part geometry and manipulator holding device.
These specifications include overall dimensions, particular geometry requirements, and cable output position.
Imasonic validate this design or propose modifications to OPG to guarantee a high mechanical resistance and watertightness under 50cm of water.
Several probes are used manually or automatically in direct contact, which is very aggressive for the front face. The wear of the front face may have unexpected consequences on the probes like water penetration in the probe or modification of electroacoustical properties.
In the meantime, the implementation of a protective layer on the front face may alter the pulse length and sensitivity due to the additional interface.
For this reason, Imasonic implemented a new hard face material l /4 material that combines an appropriate acoustical impedance for high energy transfer. OPG tested wear resistance which was 10 times higher than conventional front face.
Miniaturisation & Geometry
The restricted space available for inspection and limited access for contact area required miniaturised probes with particular front face geometry.
Imasonic manufactured small probes down to 8 x 8 x 17 mm for 6 x 6mm active area with 20 elements
Inspection speed required parallel inspection of neighbour blade roots or rotor steeple, or parallel inspection of different area of the same part.
For this purpose, OPG designed and Imasonic manufactured up to 4 head probes connected in a single connector (see Figure 3).
|Fig 3: Probe 15A & 15 B: 1-20E8-8HP 20 elements / Pitch: 0.4 mm / 10 MHz Probe 15C: 41-68E14-8HP 28 elements / Pitch: 0.5 mm / 7 MHz Automatic inspection of L-0 rotor steeple grooves.|
All probes are checked according to internal procedures and periodically calibrated equipment linked to traceable standards. The goal of these procedures is to test during the manufacturing and the final control the performances of the probe that can guarantee the specified performances in the specified using configuration.
The main parameters checked during the final control are
An example of probe performances ( nr. 9) is presented in Table 1 and Figure 4.
Probe 09: 6L32E18-8HR|
32 elts / 0.55 mm / 6MHz Manual PA L-0 blade Manual UT inlet pipe weld Automatic PA L-0 blade
|Fig 4: Probe #9 Sensitivity homogeneity|
OPG implemented a checking procedure for LPAP probes. Further to Imasonic tests, the goal is to test the beam characteristics with representative delay laws.
Automatic analysis procedure linked to excel sheets have been developed
Table 2 presents the main features to be certified. Index evaluation is presented in Figure 5.
|Fig 5: Example for evaluation of actual refracted angle, wedge delay, and exit point, using specimen features of reference block with side-drilled holes.|
|Centre frequency [ MHz ]||f0||10.8||for one set-up|
|Peak frequency [ MHz ]||fp||11.2||for one set-up|
|Pulse duration [ ms ]||D t-20||0.32||for one set-up|
|Relative bandwidth [ % ]||BW||78.5||for one set-up|
|Focal depth [ mm ]||F0||50||for specific angles/focal laws|
|Depth of field [ mm ]||L-6||14 -86||for specific angles|
|Wedge delay [ ms ]||TOFwedge||3.5 - 6.4||for specific angles|
|Refracted angle [ 0 ]||b||35 - 55 , step 5||for a specific focal law|
|Signal-to-noise ratio [ dB ]||S / N||> 30||for specific angles|
|Start Scan - Index [ mm ]||DXb||14||for specific angles|
|Beam divergence [ mm ]||DX-6dB||3||for specific angles|
|Near-surface resolution [ dB ]||A n"h=mm "||>2||for specific angles|
|Far-surface resolution [ mm ]||A f"h"mm||< 80||for specific angles|
|Skew angle [ ° ]||qskew||N/A||for one refracted angle|
|Beam dimension on X [ mm ]||X-3dB||1.8||for specific angles|
|Beam dimension on Y [ mm ]||Y-3dB||8.5||for specific angles|
|Table 2: Phased array probe main features to be certified by OPG.|
Figure 6-8 illustrate evaluation of beam divergence, depth of field and S/N ratio on various probes. The Experimental parameters are very close to modelling.
|Fig 6: Beam divergence for P3C.||Fig 7: Depth of field for probe 9 (L-waves).|
|Fig 8: S/N ratio for probe 15C - L-waves at 15°.|
Table 3 summarised the detection and sizing accuracy for in situ examination of turbine components of ABB low pressure turbine at Darlington NGS for field trial nr. 2 and nr. 3.
|Date||ABB item||Detection [L x h] mm||Accuracy [±mm]||remarks|
|May 1999||L-0 Blade||5 x 1 - P, H1||1.5||0.7||3.5||Min detection 5x0.5mm|
|L-0 Steeple||9 x 0.5 - H1 9 x 1 - H2, H5||1.5||0.5||4||Min. detection: 9 x 0.15mm Special plotting for location|
|L-1 Blade||3 x 1 - H1||1.2||0.7||2||Manual Phased array|
|April 2000||L-0 Blade||5 x 1 - P, H1||1.5||0.7||3.5||Min. detection: 5 x 0.15 mm|
|L-0 Steeple||9 x 0.5 - H1 9 x 1 - H2, H5||1.5||0.5||4||Min detection: 9 x 0.15 mm Special plotting for location|
|L-1 Blade||3 x 1 - H1||1.2||0.7||2||Scan 4-blades same time manual scans in parallel|
|L-1 Steeple||4 x 1.5 - H1||1.8||0.8||3||Automated scan 4 items simultaneously|
All probes withstand the using conditions with reliable performances for manual and automated inspections. An example is illustrated in Figure 9 and 10.
|Fig 9: L-0 steeple inspection with 3-heads probe||Fig 10: : Inspection of L-1 blades with 4-head probe|
Prototype probes passed the field test with reliable performances.
The design and manufacturing were satisfactory and an in-deep characterisation assured a high sensitivity in detection and accuracy in sizing
Some improvements concerning design and manufacturing have been identified and will be implemented for next full industrial inspection in 2001.
The authors wish to thank OPG-NOSS-SESD-SIMD Management Team for granting the publication of this paper.
|© AIPnD , created by NDT.net|||Home| |Top||