Tordis is a mature oil field that has undergone a number of lifetime assessments. An important part of that was finding out whether the wall-thickness of the bends in the Central Manifold was within accepted limits.


In the summer of 2011, FORCE Technology mobilised a tool for geometry mapping and wall thickness measurements of bends on the Tordis TCM template. However, what was supposed to be a routine inspection quickly turned out to be more challenging than first presumed. Surprisingly, the bend was covered with 40 mm of rubber coating. 

Measuring with ultrasound through 1000µm of painted coating is normally not an issue, but rubber has poor acoustic properties, absorbs noise and will therefore make it impossible to perform the inspection.

Challenges with inspection at 200 m depths

We knew from previous experience that rubber can be removed with water jetting, but the Tordis TCM was only accessible by ROV from above and the distance from the ROV to the TCM was substantial. Additionally, there were several critical components nearby, and the geometry of the bend made it difficult to perform a manual jetting, especially under the bends. Ultrasonic inspection requires a completely clean surface. 

In December 2011, we were assigned the mission of designing a Coating Removal Tool (CTR) as well as designing a tool that could map the wall thickness of the entire length and circumference of the bend. Proserv was nominated as FORCE Technology's subcontractor for the water jetting tool.

Assessment of the situation

Initially, we recognised that the first and foremost challenge would be to find a point of reference for the scanning. Due to the varied thickness of the rubber that covered both the bend and the manifold, it was clear that they would not suffice as reference points for our tool. We also knew, from experience, that bend don't typically have a perfect bend radius and are not a perfect copy of their drawings. Sections taken along the bends often show some ovality.

There is a valve in the immediate vicinity of the bend, and we chose to use this as point of reference for the tools. According to our experience, drawings from this era do not guaranteed to be "As-Built", which was the case with these drawings as well. In a collaboration with Statoil, we therefore had to make a calculated risk and make due with what information we had available.

Design the optimal solution

After careful consideration, we decided to go with a solution where we could fix a "guide post" on the valve in order to allow for precise placement of both the CRT and bend scanner. We presumed that the bolts on the valves were in accordance with the drawings, which they were. However, due to minor discrepancies in the drawings, a few modifications were needed during the operation. 

In order to allow for a greater amount of geometric flexibility for some anticipated uncertainty, we modified the bend scanning tool accordingly. The sensor design was also adjusted to secure an optimal inspection result. 

Guide post placed on valve.
Guide post and deployment tool.

Equipment customisation and testing

In the end, four customised tools were designed and manufactured; a guide post for placement of tools, a deployment tool for mounting the guide post, the bend scanner as well as the CRT supplied by Proserv. 

Testing and FAT (Factory Acceptance Testing) was performed at testing pools and FORCE Technology locations. Prior to FAT and testing, a complete mock-up of the bend, valve and manifold. The FAT procedure was extensive and aimed at proving that all steps from installation to analysis of data would work according to given criteria and design. 

Following the FAT and  minor modifications, all equipment was "fit for operation" and we were ready to mobilise. 

Mock-up of bend, valve and manifold.
Mock-up in test pool, preparing for FAT.
Transducer in contact with cleansed bend surface.
Deployment tool with guide post on its way down to the TCM for installation.

Offshore operation

After the hazard and operability study (HAZOP) performed by FORCE Technology, we were mobilised for operation at the Edda Fauna on December 14th. The guide posts were mounted on the valves via ROV. The CTR was lowered in a basket and placed on the valves by ROV. The rubber was removed effortlessly, leaving no traces of rubber on the bend.

Following the CTR, the bend scanner was deployed and placed on the valves by ROV. The scanning was performed with a resolution of 2,5 x 2,5 mm, which provided us with 23500 individual point-measurements with high quality, placed in a colour mapping for the thickness evaluations.