In-service corrosion assessment for high-temperature plant components
Abstract
In-service inspection is an essential part of integrity management for ageing assets. Condition monitoring locations (CMLs) are established across the plant and component wall thickness is obtained through ultrasonic spot measurements. These measurements are used to trend corrosion and erosion rates, assist with establishing fitness for service and complete remaining life calculations.
In addition to the cost of plant shutdown, deploying ultrasonic testing (UT) operators and providing suitable access with ropes and/or scaffolding, the accuracy of traditional manual ultrasonic testing methods requires a conservative tolerance to compensate for operator error and inherent inconsistency problems. At elevated and high temperatures these tolerances are increased as operators lose confidence in the ability to compensate for velocity variances.
One solution to these inaccuracies is to install sensors. These are often tied into monitoring systems that provide live data to control centres. While the increased data is powerful, it can come at an initial high installation cost where required infrastructure such as wireless gateways are not already in place or where a facility is already committed to an inspection programme. Some sites may have additional concerns around data security, battery life and qualification of the measurement without an authorised signatory.
Here, we propose using a novel hybrid approach, combining the benefits of installed sensors to overcome inaccuracies, and continuing to operate under the existing manual operator data collection model, to offer a lower entry cost alternative to autonomous continuous monitoring.
Commercially available 3 MHz high-temperature UT transducers have been installed under the cladding on 12" schedule 40 carbon steel steam pipework at a steel manufacturing plant, where the pipework is prone to corrosion while in-service, at temperatures exceeding 230°C continuously. Installing the transducers fixes the variability in equipment as well as providing exact reproducibility for repeated thickness measurement location on the asset to better compare to previous readings.
A-scans are collected by a UT operator using a Wand, inductively coupled (wireless near-field communication, NFC) to a passive antenna, positioned on the insulation weatherproofing. In seconds, the operator can collect a UT thickness measurement and push the measurement to the cloud. Integrated calibration blocks in the transducer enable temperature compensation to be achieved by algorithm. Wall thickness trending is reported through the cloud to supervisors who can verify the analysis and result.
For comparison, thickness measurement readings have been taken close-by during outages using the conventional manual UT inspection approach.
From the site trials presented, this acquisition process allows much faster data collection, leading to more CMLs per operator per day, and a significant reduction in operator exposure to environments containing hazards such as temperature, height and ignition risk. Combined with remote analysis and reporting, this system is shown to significantly reduce costs as well as guarantee all inspection reports/data have qualified signatory.
In addition to the cost of plant shutdown, deploying ultrasonic testing (UT) operators and providing suitable access with ropes and/or scaffolding, the accuracy of traditional manual ultrasonic testing methods requires a conservative tolerance to compensate for operator error and inherent inconsistency problems. At elevated and high temperatures these tolerances are increased as operators lose confidence in the ability to compensate for velocity variances.
One solution to these inaccuracies is to install sensors. These are often tied into monitoring systems that provide live data to control centres. While the increased data is powerful, it can come at an initial high installation cost where required infrastructure such as wireless gateways are not already in place or where a facility is already committed to an inspection programme. Some sites may have additional concerns around data security, battery life and qualification of the measurement without an authorised signatory.
Here, we propose using a novel hybrid approach, combining the benefits of installed sensors to overcome inaccuracies, and continuing to operate under the existing manual operator data collection model, to offer a lower entry cost alternative to autonomous continuous monitoring.
Commercially available 3 MHz high-temperature UT transducers have been installed under the cladding on 12" schedule 40 carbon steel steam pipework at a steel manufacturing plant, where the pipework is prone to corrosion while in-service, at temperatures exceeding 230°C continuously. Installing the transducers fixes the variability in equipment as well as providing exact reproducibility for repeated thickness measurement location on the asset to better compare to previous readings.
A-scans are collected by a UT operator using a Wand, inductively coupled (wireless near-field communication, NFC) to a passive antenna, positioned on the insulation weatherproofing. In seconds, the operator can collect a UT thickness measurement and push the measurement to the cloud. Integrated calibration blocks in the transducer enable temperature compensation to be achieved by algorithm. Wall thickness trending is reported through the cloud to supervisors who can verify the analysis and result.
For comparison, thickness measurement readings have been taken close-by during outages using the conventional manual UT inspection approach.
From the site trials presented, this acquisition process allows much faster data collection, leading to more CMLs per operator per day, and a significant reduction in operator exposure to environments containing hazards such as temperature, height and ignition risk. Combined with remote analysis and reporting, this system is shown to significantly reduce costs as well as guarantee all inspection reports/data have qualified signatory.
