· Home· Table of Contents
· Fundamental & Applied Research
Detection and Sizing of Cracking from the Inner Surface using ID Creeping Waves
Peter Hayward, Certification Board for Inspection PersonnelCBIP,New Zealand.
Contact
| · Home · Table of Contents · Fundamental & Applied Research | Detection and Sizing of Cracking from the Inner Surface using ID Creeping WavesPeter Hayward, Certification Board for Inspection PersonnelCBIP,New Zealand. Contact |
Once an inner-diameter (ID) connected crack has been detected using a standard or specification scanning requirements then the height of the crack needs to be evaluated. The initial process generally involves scanning the item to the requirements of a standard such as AS 2207. The sizing methods in AS 2207 use the dB drop methods e.g. 6dB and 20dB or the last significant echo method (LSE).
Another method that can be used for finding the height of the ID connected flaw is to apply the creeping wave technique using a single crystal probe.
These methods of advanced flaw sizing technique use refracted longitudinal and shear waves to determine the height of the crack.
The results obtained using the ID creeping wave technique can be evaluated using Tip Diffraction, Bi-Modal, or High Angle longitudinal waves.
This paper will discuss the application of the above techniques and the results.
Keywords: Ultrasonic, welds, cracks, creeping waves, sizing
The reliability of ultrasonic inspection for the detection, identification and sizing of flaws is of considerable importance. To ensure the integrity of the structure or pressure equipment we need to consider the influence the flaw type has on the design life where cracking is found. This is particularly significant when previously unknown cracking is found during inspection surveys. This paper discusses the traditional ultrasonic sizing methods and the more recent advanced methods using the creeping wave.
Ultrasonic inspection is often used to determine if newly fabricated pressure equipment meets the fabrication requirements. This also applies for in-service ultrasonic inspection when cracking and other serious defects are suspected and need to be detected. While the location and characterisation of a reflector is the first obstacle, the second is the sizing of the reflector once is has been found. This is a major factor when the reflector is a crack and the results of the inspection are to be used to determine the equipment's fitness for purpose. While the length of a surface breaking defect such as a crack can often be determined by using one of the complementary NDT methods such as liquid penetrant or magnetic particle inspection the determination of the crack height needs a reliable inspection method. When the crack is on the inner surface of a pipe, tube or vessel and access to the inside surface is not possible both the length and height need to be determined using a reliable NDT method. The evaluation of crack height is one of the more controversial topics in ultrasonic flaw detection. The traditional sizing methods in conventional manual ultrasonic inspection are the 6dB and 20dB methods along with the maximum amplitude and DGS.
The 6dB sizing method is practical and straightforward and is normally used when the reflector is wider than the width of the impinging ultrasonic beam while the 20dB sizing method is used when the impinging ultrasonic beam is wider than the reflector. The choice of which sizing method to use often depends on operator preference. The 6dB sizing method is shown in FIG 1 below.
Fig 1:
|
The sizing methods using advanced techniques such as inside diameter (ID ) or outside diameter (OD) creeping waves, bi-modal and refracted compression waves may be considered to be more accurate than the traditional sizing methods. Using any of the sizing methods requires reflectors of a known length and height, equipment of known performance and personnel suitably trained and qualified.
Fig 2:
|
The I.D. creeping wave probe is essentially a sub surface wave that runs along the inside surface of the material being scanned. The creeping wave is produced, refer FIG 2 above, by using a probe that generates a 30° shear wave, a 70° refracted compression (longitudinal) wave and a second 70° refracted compression wave.
The probes used in this paper to produce creep waves were a WSY70-2 and a WSY70-4 single element transducers. The probes of this design were developed in the 70's to inspect stainless steel welds and to detect under cladding cracks in the nuclear, petrochemical and pulp and paper industries. Additionally a 45° shear wave probe and Bi-modal probes were also used. The above probes can be used to find and give reliable initial information on the depth (height) of a crack as follows:
Fig 3:
|
Note: The above, FIG 3, shows the typical types of CRT presentation when all three of the signal responses discussed above are shown on the CRT. The standing wave between 5 and 15 is not shown.
Prior to testing the effectiveness of this method of sizing the ultrasonic flaw detector was set up as discussed below.
The initial calibration was done using a block of similar material and thickness with notches running from the ID of 20%, 40%, 60% and 80%. The ID creeping wave and 30 - 70 - 70 signals were set to approximately 40 and 50 on the CRT with the amplitude of the latter being adjusted to give a suitable evaluation setting. This adjustment was necessary to control the base line noise level to around 5%.
The calibration method used permits the operator to classify the reflector as shallow, mid wall or higher. Following this initial calibration the reflector heights of the test block were checked.
The test blocks, carbon steel 25mm thick, had induced cracks of varying heights, all of which were surface braking on the ID (back wall). The blocks were scanned using the 30 - 70 - 70 probe and then evaluated for height using a standard shear wave probe 45° and a twin crystal Bi-Modal probe. The results were recorded as shown below and are reported in percentages of material thickness.
| sample | 45°Tip Diffraction | Bi-Modal | Actual |
| A | 40% | 50% | 55% |
| B | 50% | 54% | 61% |
| C | 32% | 49% | 43% |
| D | 35% | 40% | 39% |
| E | 35% | 48% | 52% |
| F | 40% | 44% | 48% |
| G | 7% | ND | 9% |
| H | 40% | 61% | 65% |
| I | 45% | 52% | 52% |
| J | 15% | ND | 15% |
| Table 1: | |||
The above results were evaluated using the conventional pulse echo 'A scan' presentation and also using radio frequency (RF) presentation. The amplitude of the creeping wave signal relies upon the refracted angle of the probe used. The greater the refracted angle the less gain is required, refracted compression waves above 55°will produce an ID creeping wave.
From the above results the creeping wave can be used to size vertical planar reflectors emanating from the ID with an accuracy of better than 15%. If the material being scanned is divided into zones, the following criteria may be used:
The initial sizing is done using the half-skip method that will require weld reinforcements to be removed. This technique will be of limited use, as a sizing method, if welds are left as welded. Operators need to be trained in the various methods and be able to recognise the correct signal amongst the many low level (noise) signals. The weld roots can produce spurious responses that may also need to be resolved.
This method of sizing is easy to apply and can be used to quickly determine the approximate height of the reflector and may be more reliable than the traditional 6dB and 20dB methods because they are based upon the arrival times of the echo. This can be very useful when the results of the traditional sizing methods are being questioned.
Contact Energy for loan of blocks and probes.
Materials and Testing for machining blocks.
New Zealand Welding Centre for the welded plates and blocks.
He author also wishes to thank CBIP, HERA and the NDTA for their support.
| © AINDT , created by NDT.net | |Home| |Top| |