· Home · Table of Contents · Methods & Instrumentation

Automated Ultrasonic Rotoscan Weld Inspection - 'AGL - Roma to Brisbane Stage & Gas Looping Pipeline Constructed by Clough/Lucas Neile Findlay (Managing Director) Steeltest Australia Jan Van Der Ent (Project Manager) RTD, Holland. Contact

Introduction

In June 2000, the AGL - Roma to Brisbane Stage 4 Gas Pipeline Looping (54kmX 16"/400mm) was the first Australian Pipeline project to be fully inspected via Automated Ultrasonic Weld Inspection, utilizing the RTD Rotoscan system, in lieu of X-Radiography. Steeltest Australia in conjunction with RTD of Holland were awarded the NDT Contract by Clough/Lucas to perform Auto UT on all 4 loops constructed.

Loop 1	Wallumbilla -KP00 to KP12	16" /400mm Æ X 6.6mm WT
Loop 2	Bowenville - KP00 to KP08	16" /400mm Æ X 6.6mm WT
Loop 3	Toowoomba - KP00 to KP16	16" /400mm Æ X 7.9 & 9.5mm WT
Loop 4	Ipswich - KP00 to KP11	16" /400mm Æ X 8.8mm WT

Welding & Construction of the line was carried out by " East Coast Welding" with Steeltest Australia & RTD performing all aspects of NDT.

This paper describes the use of the RTD Rotoscan Automated Ultrasonic Inspection System used to inspect the pipeline Girth Welds during the construction of the above mentioned pipeline.

CONTENT

1. Description of the Rotoscan AUT System
2. Automatic Welding & AUT Process's for Setup Configurations & Trials
3. NDT Project Organisation for both AUT & Radiography
4. NDT Comparisons - AUT V's X-Ray
5. Overall Project Summary

1.0 Description of the Rotoscan AUT System

Introduction
RTD, being the pioneers in the field of Automated Ultrasonic Inspection, provides a field proven mechanised ultrasonic inspection system, the Rotoscan. The system was developed for the inspection of girth welds during construction of long-distance pipelines, both On/Offshore. High inspection speed and instantaneous recording of results characterise Rotoscan. Unlike radiography, it provides immediate weld quality feedback, which can be detrimental to the pipeline construction.

The inclusion of a mapping feature enables the system to visualise the presence of the geometrical welding features such as the position of the weld cap and root penetration, which minimises the possibility of the system generating false calls. Furthermore this feature enables the system to comply with most existing Ultrasonic Codes and acceptance criteria's, because of its capability to detect and quantify volumetric defects.

Moreover, the integrated simultaneous TOFD feature within the Rotoscan system provides additional information to assist in the evaluation of indications. In addition this technique offers the ability to determine and assess the through-thickness dimension of indications. The present Rotoscan is capable of achieving a low False Call Rate (FCR) coupled with flaw sizing capabilities, a user-friendly presentation and storage of results.

The above technical improvements in flaw sizing and recording have allowed the practical application of rejection/acceptance criteria of weld defects that are based upon fracture mechanics (Engineering Critical Assessment) as well as workmanship standards. The development and actual use of such modern acceptance criteria supported the introduction of Rotoscan. Worldwide commercial application has proved that, contrary to the expectations of many people, ultrasonic inspection does not lead to higher weld repair rates than radiography does. Since its commercial introduction in early 1989, over 8,000km of pipeline (some 640,000; welds) have been inspected by RTD. The Rotoscan system has been qualified in numerous countries by a variety of customers and regulators for different welding processes, pipe diameters and wall thicknesses.

The system's many beneficial features include:

• No radiation hazards or harmful chemical waste - environmentally friendly
• Instant online presentation of results.
• Immediate weld quality feed back to reduce repair rates
• Allows use of ECA (Engineering Critical Assessment) and post project engineering
• Substantial reduction in inspection time
• Offshore high-speed inspection - 350 welds per day
• No delays or stoppages to adjacent pipeline workers
• No equipment present inside the pipe (no crawler cut-outs)
• Capable of inspecting welds at elevated temperatures (up to 392°F/200⁰C)
• Acceptance criteria can be adjusted to suit client requirements
• Can accommodate most weld bevels or welding processes SMAW or GMAW

Inspection Principle
Full weld inspection coverage is achieved by placing an ultrasonic probe array on both sides of the weld, each probe within the array examines a depth zone within the weld. This eliminates the need to move the probes to and from the weld, as in the conventional practice in time consuming manual ultrasonic inspection. The ultrasonic information is transferred to a computer for data presentation and analysis.

Fig 1: shows a cross section of a typical mechanised weld divided into a number of weld passes.

Fig 2: shows a plan view of the probe arrangement with focused or natural focussed beams.

Fig 3: shows a cross section and plan view of a typical SMAW weld.

Fig 4: shows a cross section for transverse detection.

AUT Pipe Scanners
The Rotoscan can be equipped with different scanner types to cover a wide range of pipe diameters starting from 2" up to 56" diameter.

All scanners are designed to be handled by one operator. For the standard scanner, a second operator is needed to handle the guiding band. An electric drive motor provides the scanning motion. An encoder measures positions around the circumference. The micro scanner (2-6") as displayed is of a horseshoe shaped, clamp-on type. Micro-scanner for 2 to 6"

The large diameter scanners (12" and up) are mounted on a track, which is preferable the same as used for automated welding. The larger diameter scanners are guided along on a pipe band, which needs to be mounted before the inspection commences. The scan frame can be quickly changed. Changeovers can be completed under 5 min.

Flex-head scanner on 18" pipe

Vehicle & Umbilical


For Onshore Pipelines, the Rotoscan is fitted to a 4WD vehicle where all inspections are carried out from. The umbilical cable is required for connection between the scanner and the electronics, which are housed inside a 4WD vehicle or an instrumentation cabin (Offshore). Operating lengths are between 50 and 150 feet.

Ultrasonic Instrument
Computerized Multi-Channel Ultrasonic Instrument, up to 64 sequences (channels), with programmable setting for:

• Transmitter and receiver channels 1- 32
• Programmable Gain 0-80 dB (1 dB increments)
• Gate start and length. (min. increment 0.1mm)
• Delay (mm)
• Threshold level 5% - 100% Full Screen Height (FSH)
• Recordable output selection between; Amplitude and/or Transit Distance, Mapping, TOFD and Coupling.
• Selection between First / Highest amplitude detection within the gated section.

Computer
Computerised data storage with dedicated software, offering color-enhanced user-friendly and coherent presentation enabling quick and unambiguous interpretation of inspection results.

Printer
Printer to produce instant field inspection records as required.

Calibration


A calibration plate, made of an original piece of the pipeline material to be inspected, is prepared with artificial defects such as flat bottom holes and/or notches, which represent actual flaws. Artificial defects are present in each weld pass. These plates are independently checked (DNV) for accuracy prior to use.

Coupling System
Coupling medium (water or glycol) supply system for the acoustic contact between ultrasonic probes and pipe surface. This couplant is also used to cool down the outer pipe surface to a temperature just below 212°F, prior to scanning.

Probes


As part of 'Project Preparation' Ultrasonic probes are designed and manufactured by RTD to the specific weld configuration and material (angle, frequency and focus) to be inspected for each individual pipeline project. Probes can also be designed to withstand high temperature pipe surfaces, in particular for Offshore Barge construction.

Data Presentation
The Rotoscan system incorporates computerised data presentation and storage. The onboard computer enables ultrasonic signals to be digitised, which is essential for coherent C-Scan mapping and the use of Time of Flight Diffraction (TOFD) technique within the system.

In addition, the software allows for automatic judgement of indications and defect tally list generation. The system also provides weld cross sections showing the position/depth of indications within the weld.

As an example, a typical weld defect indication as produced by Rotoscan is shown alongside a photograph of the sectioned weld showing the defect. The example is derived from the RTD Rotoscan Image Reference Collection used for training purposes.

Circumferential position 2434 mm.

Macro - Carbon Steel Circ Butt Weld (SAW) - 25.7mm WT.

Flaw Sizing Methodology
The RTD Rotoscan meets the requirements of API 18/19th Edition both Workmanship and Appendix A requirements, by using the following applications :

• The API requirements for Good Workmanship standards are described in the standard procedure, this is a straight-forward criteria based on amplitude and length.
• As to the requirements of the Appendix A concerning interaction: Imperfection length and through thickness size is used to define interaction. This is achieved through the Automatic Interpretation Module (AIM).
• Through thickness sizing of embedded imperfections is primarily done on the basis of signal amplitude, using a conversion from the calibration reflectors present in the calibration block (flat bottomed holes) to strip-like reflectors (lack of fusion type). This conversion is probe type dependent.
• In addition, the number of zones in which the indication occurs is taken into account in the sense that signal signature and TOFD are used to ascertain that signals are from the same defect:
• If yes, signals are from the same defect, size is based upon the the largest amplitude supported by TOFD. If at least one of the amplitudes exceeds 100% it is concluded that the defect occurs in more than one zone, and the relationship with calibration reflectors is more or less lost. In that case a combination of TOFD and pulse-echo (at -12dB) is used to measure height.
• If no, signals are not from the same defect, the defects are sized and reported separately. The interaction rules occur in that case.
• In the case of surface breaking defects, sizing is more complicated because amplitude as a function of notch / defect height is not linear and to a degree ambiguous. The notch depth will therefore have a height equal to the the critical height calculated on the basis of the API / ECA requirements, and the role of TOFD is more pronounced for confirmation.

Amplitude Sizing for Appendix A Requirements
The requirements as laid out in the API 1104 Appendix A, require that both the length and depth (defect height) be established, which is not possible with Radiography methods. The technique used is based on the amplitude method, were a relationship is present between reflector size, soundpath and gain (amplification).
Flaws in GMAW welds methods are predictable in location with a planar character.
The defect reflectors simulated in the calibration standard are typical incomplete fusion.
However instead of disc shaped reflectors, slots or strip like reflectors are used.


Evaluation of Stacked and Inter-Zonal Flaws
Flaw signatures (from pulse-echo channels) provide information to the extent of stacked flaws or inter-zonal flaws. When the echo dynamics of adjacent transducers is the same, then two probes record the same flaw. In this case the highest amplitude is selected to calculate the equivalent flaw size.

• Flaws with different echo dynamics require individual sizing.
• Ultrasonic signals from flaws exceeding 100% FSH, require evaluation at -12dB level.

Below, a typical screen display with the Rotoscan Automatic Interpretation Module (AIM) is shown.

The defects are marked either red or green based on the acceptance criteria and the 'API Appendix A' interaction rules. The interaction rules are performed as a separate evaluation through the AIM module. All pertinent information from both indications 1 and 2 is collected. The AIM module will calculate if interaction exists. These results will be transferred (when affirmed by the operator) into the indication list section. (see next page) The AIM module supports the 5 cases as defined by the Appendix A.

Indications List Section
The defect list section is displayed in one of the pull down menus. This information is immediately collected from the interpretation (AIM) of the examined weld. The following defect categories are recognised;
• Defect start, (from zero point) end and length
• Depth from OD. surface
• Height of the defect
• Weld pass
• Registration in dB
• Diagnosis, accept/not accept.
• Interpretation (Defect ID)
• Channel ID
• Acceptance Code

Header Information


The header information section is displayed in the top right corner of the screen. The information section contains the RTD logo, the selected filename, present date and user comments as required

X-Section


The X-Section of the weld is displayed in the middle of the header. When a cursor is moved over the Rotoscan data the corresponding position within the welds is displayed as a black bar. A green bar is displayed if the cursor passes an acceptable defect indication. A red bar is displayed if the cursor passes a non-acceptable defect indication

Coupling Monitors


The coupling monitor channels display a red color where the amplitude drops below the threshold level or a green color otherwise. The total number of coupling monitoring channels matches the number of inspection channels.

Distance Measurement


On the right hand side of the image, the circumferential position in mm, with increment resolution of 10 mm is visible. (The circumferential increment is 2 mm).

Amplitude and Transit Distance Data Presentation


The marked defects can be identified by a label. Each label contains the type of defect, it's position and length. An abbreviation of the code can be selected as well. All defect labels, can be individually turned on or off. (Up-Stream side only, all three labels turned on)

The pulse-echo channels show both the Amplitude (Amp) and the Range or Transit Distance (TD) of the signal in the gate. Each channel represents an inspection zone, with the total number of channels displayed, related to the number of zones and wall thickness of the weld. The layout is symmetrically built-up whereby the channels on the left-hand side are used for probe functions at the up-stream side of the weld centerline. The right hand side displays the channels at the down-stream side of the weld. Each channel represents a depth zone within the wall thickness of the weld counting from the weld Cap to Root. The Transit Distance Amplitude channels show the registration of analogue amplitude together with the transit distance of each individual probe in use. The amplitude line is drawn in a black line 0 to 100% and the transit distance presented in a solid grey bar. The width of the grey bar is related to the measured transit distance, with the transit distance 'TD' being then used to determine the flaw's position in the weld. The rejection threshold is also represented on each channel by a series of dots. The usual evaluation threshold is 20% and the computer automatically indicates any signal exceeding this threshold.

Mapping Image Data Presentation
The gate-and-threshold based pulse-echo concept is expanded by a mapping-based presentation. Instead of recording only the maximum amplitude and range of the signal detected within a gate, the A-scan within the gate is digitised and recorded and displayed as a single line. The amplitude of the signal at each point along its length is color coded, and these lines are stacked together to produce an image. This gives a graphical 'map' of the root and cap areas, and it is easier to monitor changes.

The zones usually selected for mapping are:

• those looking at the body of the weld to look for internal flaws such as porosity
• the root, which aids interpretation of the root signals by identifying the presence and location of the root penetration.
• the cap, which aids interpretation of the cap signals by identifying the presence and location of the cap reinforcement.

All mapping channels are displayed in the middle of the Rotoscan screen. Proper alignment of the guiding band to the weld centerline can easily be monitored as well.

Mapping Advantages
The mapping facility offers the possibility of pattern recognition, which appeared to be the key feature for reliable inspection of welds with narrow roots (manual stick, narrow gap) as well as recognition and quantification of volumetric defects such as porosity. This has been made possible by the fact that the improved technology enables unambiguous signal interpretation such as root penetration and cap position (Geometry Indications)

Mapping Display


Fig 3: Principle of mapping in the root

TOFD Advantages
TOFD (Time of Flight Diffraction) in combination with standard pulse-echo techniques as used in Rotoscan, can offer unique advantages over the application of pulse-echo techniques alone, especially since modern high-speed signal processing enables the application of this technique at the normal Rotoscan inspection speed (typically 2 - 4 inch/sec). Principle showed below.

The TOFD presentation can be switched to the centre on the monitor for more detailed evaluation and can be further enhanced with a separate pull-down menu featuring;

• Linearise TOFD, this allows direct defect depth measurements from the image
• Filter TOFD, allows the image to be smoothed to improve quality.
• Straighten TOFD, will straighten the lateral wave.
• Remove TOFD lateral wave, allows lateral wave to be removed, which is very effective for looking for flaws near to the upper surface.

Toolbar
To achieve a quick interpretation and navigation of the program a number of functions can be accessed through hot-keys in the tool bar at the top of the screen presentation.

Data Storage and Quality Management
Directly after the scan the inspection data is stored onto hard drives, then at the end of the pipe string or end of the day, copied onto Optical Disc. A hard copy of the inspection result can be produced on a paper format, in black and white. For archiving purposes at the end of the project, the data is then stored on optical disc (CD-ROM) from which the screen images can de derived in colour, if required.

2.0 Automatic Welding & AUT Process's for Setup Configurations & Trials

In preparation for the AGL Roma to Brisbane Gas Looping Pipeline Project, a Project Bar Chart with an overview of all required activities was established, in order to meet the planned project start date in March 2000.

The objective was to carry out a pre-qualification inspection on Automated welded joints in co-operation with Vermaat Automated Welding from Holland, before mobilisation to Australia.

In order to meet the required objectives, the following actions were planned;

• Set-up of Rotscan Inspection system for 16" /400mm Dia Thin Wall Pipe, after receipt of the pipe material, weld bevel configuration & inspection specification
• Accurate measurement of the pipe material sound velocity and wall thickness's
• Design of Probe Layout and manufacture of probes to target each depth zone of the weld
• Design/Manufacture of Rotoscan calibration blocks in accordance with ASTM-E-1961-1998
• System configuration to suit 16" /400mm Dia Pipe
• Rotoscan Procedure Development
• Pre-Qualification of weld inspection
• Mobilise to Australia for Auto Welding Trials
• Setup Auto Welding Equipment to East Coast Welding Mainline equipment
• Train East Coast Welders with the Automatic Welding Process
• Produce Auto Welding Pup pieces (6.6, 7.9, 8.85 & 9.5mm) for Auto UT, X-Ray and Mechanical testing
• Establish Defect detectability comparisons for AGL & Clough Lucas QA/QC (of which AUT proved far more sensitive in particular to 'Fusion' type defects. Certain defects such as 'Lack of Side Wall Fusion' (LS) were not visible with X-Ray, but clearly visible with AUT & Macros.

Once all of the above were completed the equipment was mobilised to the commencement of the Mainline. Initially due to terrain encountered, the mainline commenced with 'Manual Welding', then once the right of way flattened out, Automatic Welding resumed. For the first 5km of welding, all welds were inspected with both AUT and Crawler X-Radiography, with the AUT proving to be far more accurate and sensitive. Another advantage to the project, AUT was also able to accept more weld defects detected with X-Ray due to the ability and accuracy of AUT being capable of measuring depth and height of the defects, which X-Ray could not assess.

3.0 NDT Project Organisation

AGL - Roma to Brisbane Gas Looping Pipeline Project

NDT Project Organisation

4.0 NDT Comparisons - AUT V's RT

During the initial setup trials various defects were not visible or visible considerably smaller in length with Radiography (X-Ray) compared with AUT results. These areas were also confirmed with Macro's. An example is shown below;

Depth Zone	Defect & Position	Result
Z1-US	LS - 22 - 40mm	Not visible on X-Ray
Z1-DS	LS - 142 - 430mm	Detected by both techniques
Z3-DS	LR - 142 - 380mm & LR - 460 - 648mm	Not visible on X-Ray (Stacked Defect)
Weld QW 4

4.0 NDT Comparisons - AUT V's RT

5.0 Project Summary

• Throughout the Project all Terrains were encountered (traditional rain-forest to the steep inclines/descents of the Toowoomba Range), where the Rotoscan fitted to a 4WD Truck, tackled all sections, of which some sections were not accessible with pipeline crawlers.
• Welders were provided with immediate results, with the 2 man AUT crew remaining approx 4-5 welds behind the front-end capping crew, wherever possible.
• The 16" /400mm Dia pipe (including 100mm Overlap) was inspected in the following time frames;


• Attach the CRC band onto the pipe : 45 sec
• Check position related to weld center : 15 sec
• Install the scanner onto the guide band : 10 sec
• Inspection of 16"/400mm pipe at 70mm/sec : 20 sec
• Evaluation/Sentence UT Scan : 0 sec (if defect free) 15-30 sec (if defect found)
• Remove guide band from pipe : 10 sec

Total cycle time : 100 to 130 sec (Approx 2 Minutes)

• Inspection Production per day - 60 to 120 joints per day including Tie-Ins
• Where any repairs were given, the same Mainline crew completed their own repairs during breaks or at the end of the work day, eliminating the need for a separate repair crew, then the repair inspected immediately after sufficient cooling
• Overall Project Repair rate less than 1%
• Rotoscan inspection results available immediately after scan is completed. Results saved to hard drive automatically, then onto optical disc after the completion of each string or at the end of the work day.
• Welder Qualifications carried out on the mainline, with immediate AUT results, enabling the welder to continue welding immediately.
• Final report for the project delivered to Clough/Lucas on (2) CD Roms with full viewing capabilities of each weld scanned. No need for large quantities of RT film to be stored.
• Various cost savings for the project including;


• Increased pipe string length, allowing reduction in Tie-ins required
• Reduced weld repairs due to immediate feedback of weld quality
• Ability to inspect concrete coated pipe for various sections in lieu of radiography
• Completed critical Tie-In welds as required - ie : Road X-ing (4 Joints) - 9 minutes.

Overall Automated Ultrasonic Weld Inspection has proven to be the way of the future in Australia, bringing us to par with the likes of Europe, Canada, USA & Offshore, where AUT is used a majority of the time.

© AINDT , created by NDT.net

|Home|    |Top|