Spherical pressure vessels are manufactured in accordance with the American Society for Mechanical Engineering Boiler and Pressure Vessel Code (ASME BPVC) and are typically managed under the American Petroleum Institute API-510 Code: In-service Inspection of Pressure Vessels and and ASTM E 1930-02, Standard Test Method for Examination of Liquid-Filled Atmospheric and Low-Pressure Metal Storage Tanks Using Acoustic Emission. In-service testing of spherical pressure vessels using acoustic emission is an economical and technically sound approach to operating the vessel using risk based inspection (RBI) strategies. An example spherical pressure vessel that was tested with acoustic emission is shown below. The sphere was made in accordance with American Society for Mechanical Engineering Boiler and Pressure Vessel Code Section 8 Division 1. The spherical pressure vessel shell is 23 mm thick and all welds were tested with magnetic particle and radiographic inspection during fabrication and before hydrotesting. Once placed into service, acoustic emission testing of the sphere may be used to screen the weld for active fatigue, corrosion, and erosion based flaws.
Spheres that are tested with acoustic emission are commonly made from SA516-GR70 steel and are typically purchased with a corrosion allowance of at least 1.6 mm. In addition to the shell plates which can be tested with acoustic emission, there are numerous attachments that are welded to the sphere, shell that are also tested with acoustic emission. This includes welded reinforcement pads for the support legs, flanges inlets and outlets at the top of the sphere, a drain valve at the bottom of the sphere, and cooling water piping.
Acoustic emission testing may be used to detect fatigue cracks that form in and around nozzle and penetration welds due to numerous pressurization cycles. Acoustic emission testing may be used to also detect corrosion product, corrosion activity, and corrosion related cracks which may include stress corrosion cracking (SCC), corrosion fatigue cracking (CFC), and intergranular corrosion cracking (IGC) and transgranual corrosion cracking (TGCC) depending on the material of construction used. Generally speaking acoustic emission is not used to detect erosion related defect in spherical pressure vessels.
Prior to acoustic emission testing of spheres, a review of previous spherical pressure vessel documentation is required. This includes;
- NDT reports – including acoustic emission inspection reports
- Maintenance reports
- Operating historical records
- Risk Based Inspection (RBI) analysis
- Repairs, alterations, service or rating change
- Fitness For Service (FFS) previous analysis
- Construction drawings and calculations: MAWP, minimum thickness, etc.
- Data sheet, data report, data book
Setting up a spherical pressure vessel acoustic emission test is a multi-team effort between the owner, inspection company and safety personnel. The most time consuming aspect of the acoustic emission testing is installing the acoustic emission sensors on the tank shell. The acoustic emission sensors used are typically in the 150-300 kHz range. The AE sensors must be distrubited over the sphere evenly to ensure that acoustic emission sound wave from fatigue cracks and corrosion related activity and cracks can travel to the sensor and be received with sufficient signal-to-noise ratio. And example acoustic emission sensor layout used during an acoustic emission test of a sphere is shown in the figure below. In this scenario, 24 sensors were used to monitor the sphere for acoustic emission 6 rows of sensors. Acoustic emission sensors may be installed at significant cost using scaffolding. Typically it takes up to 1 week to install the scaffolding and this approach may only be worthwhile if other inspection methods are leveraged into the process. It is more cost effective to install the acoustic emission sensors using a rope access technician. Acoustic emission sensors be installed and removed in a single day.
The acoustic emission instrumentation consists of sensors, analog filtering, and recording equipment. The instrumentation is capable of recording acoustic emission data above the amplitude threshold and had sufficient channels to locate AE sources in real time. Hit detection is required for each channel. Amplitude distribution is recommended for flaw characterization. The acoustic emission instrumentation acquired and recorded count, hits, and amplitude information on a per channel basis.
The acoustic emission software used displays the sphere in 3-D along with the position of the acoustic emission sensors. The red squares superimposed on the sphere are acoustic emission events. An event is located on the sphere shell when 3 or more sensors receive and acoustic emission hit. Relative time delays are stored, and triangulation is used to approximate the source of the acoustic emission on the sphere. The acoustic emission events located on the 3-D plot below were from a small crack located on the reinforcement pad used to brace the support leg to the shell.
The acoustic emission pressurization sequence will follow ASME BPVC Article 12 Pressure increments and shall generally be to 50%, 65%, 85%, and 100% of maximum test pressure with an option to pressurize to 110% the maximum test pressure. Hold periods for each increment shall be 10 min and for the final hold period shall be at least 30 minutes. During hold time, acoustic emission hit and events are recorded along with the respective acoustic emission features including acoustic emission amplitude, duration, energy and count. Additionally the rate at which acoustic emission is emitted is also recorded.
Normally, the pressure test will cause local yielding in regions of high secondary stress. Such local yielding is accompanied by acoustic emission which does not necessarily indicate discontinuities. Because of this, only large amplitude hits and hold period data are considered during the first loading of vessels without post-weld heat treatment (stress relief). If the first loading data indicates a possible discontinuity or is inconclusive, the vessel shall be re-pressurized from 50% to 100% of the test pressure with intermediate load holds at 50%, 65%, and 85%. Hold periods for the second pressurization, if necessary, shall be the same as for the original pressurization. The vessel shall be pressurized with fluid.