Subsea technology inspired by desert cacti

How a 2002 wind tunnel study on the Saguaro cacti led to the development of uniquely shaped subsea buoyancy modules.

In 2014 AMOG engineers started looking deeper into the results of Talley & Mungal's 2002 wind tunnel study on the saguaro cacti and the factors that may have contributed to their evolutionary selection (1). Despite heights of over 12 meters (40 feet) and a narrow root base, these cacti were found to be able to withstand hurricanes. The hypothesis was that the geometry of the cacti created a fundamentally low drag shape; perhaps an evolutionary selection factor whereby the lowest drag shapes, when compared to an equivalently sized circular shape, were able to survive these high wind speeds.

What does this have to do with subsea structures? 

When it comes to currents and flow, wind and water are not dissimilar.

With more than two decades of experience in offshore engineering and a specialisation in Vortex-Induced-Vibration (#VIV), the engineers at AMOG had long been looking at ways to reduce the drag of smooth cylindrical subsea infrastructure, such as the buoyancy modules needed on risers in deep waters up to 3600m. The significant VIV response experienced by riser units had the potential to cause fatigue issues and an overall amplification of the drag. 

There are 2 key reasons why this matters: 

1. Economics - high currents (and the drag effects on top and bottom angles) have the potential to lead to costly suspension of operations. Sometimes, this can require the complete disconnection and retrieval of the riser string.
2. Safety. High currents and riser string retrieval on offshore structures is complex and presents an increased the safety risk for those charged with undertaking the work.

Of course, the VIV challenge is not new to offshore engineers, but a previous study of the existing solutions by Taggart and Tognarelli (2) outlined the need for a new or different technology to fill the gap where helical strakes or fairings were limited in their effectiveness. 

So, AMOG set about studying, researching, and testing the hypothesis: that the unique geometry of the cacti could be used to reduce drag and suppress VIV in offshore currents. 

AMOG developed a two-phase model testing campaign (a sub-critical Reynolds Number flow, and a large scale post-critical Reynolds Number flow) to investigate the potential benefits of hydrodynamic performance to address the industry gap. It was an iterative process, culminating in only the most successful design being commissioned for scale model production and flow tank testing.

Research Phase 1 - Sub-critical Reynolds Model Testing

This planned research program included 70 physical specimens produced via 3D printing and more than 150 tests in a flume tank at Monash University. Specimens were assessed for drag and VIV characteristics at low Reynolds flows (sub-critical in the range between 3,000 to 16,000). The tests were conducted in fixed, freely vibrating, and forced vibrating conditions in the FLAIR (Fluids Laboratory for Aeronautical and Industrial Research) water channel at Monash University. 

The technical details of Research Phase 1:  The nominal cylinder was 40 mm diameter with a wetted length of 0.6 m and the cylinder was exposed to uniform flow in a water channel where velocities could reach a maximum of 0.8 m/s. In- line and cross-flow forces were measured via load cells. The cylinder was constrained to vibrate in the cross-flow direction only with motions being measured via an LVDT. 

More than 150 tests were conducted in a planned intermittent process over a twelve-month period, allowing for careful interpretation of the results along the way to feed back into optimising the next set of cross-sections to build. The cylinders were arranged in the preferred default arrangement with the grooves and ridges alternating down the long axis (generally four alternating sections in these tests), such that ridges and troughs were lined up at the ends of each alternate section, thus disrupting the phasing of vortex shedding. 

Key parameters identified for systematic variation included: 

• Trapeziodal/triangular shaped ridges (“T-Series”) 
• Rounded ridges (“R-Series”) 
• Heights of ridges 
• Width of ridges 
• Shapes of ridges 
• Width of grooves 
• Equal ridges vs alternating unequal ridges 

The results were that both the “T-Series” and the “R-Series” provided drag reduction by their modification of the flow field. 

But it wasn't enough to just find the shape with the greatest drag reduction or VIV suppression. For commercial viability, the solution also needed to consider the following objectives and constraints:

• A maximum allowable outer diameter such that the buoyancy could be installed through the drill floor and be stacked
• The uplift should be maximised, hence the cross-sectional area should be maximised
• It should have reduced drag and VIV suppression properties
• It was strongly desirable to achieve reduced drag and VIV suppression without active moving parts
• The modules could not be large and cumbersome and had to be handled & stacked easily, and
• It had to be readily moulded into subsea buoyancy

Most of the shapes tested seemed to disturb the flow by virtue of their geometry and the hypothesis was that they were interfering with the vortices' formation by stretching, contracting, and/or splitting them; leading to less severe vortices and ultimately less drag. But, overlaying the list of requirements above, revealed the optimal number of longitudinal grooves was found to be about 8 or 9, with 16 grooves proving worst in terms of drag performance.

The optimal candidate had one of the smallest peak responses among LGS variants tested. However there was less existing data in the literature relating sub-critical A/D (Peak Amplitude over Diameter) responses to post-critical A/D responses. It was therefore important to undertake post-critical A/D response testing of this optimal candidate.

In the end, the "winner" of Phase 1 was... the "R8"

Research Phase 2 - Post-critical Reynolds Model Testing

After R8 was identified as the optimal variant, it was commissioned for large scale (1:3.8) construction and testing at the NRC-C 200m long towing tank facility in St John’s, Canada.

The tow tank facility has a well-established rig, which had previously been used by Oceanic Consulting Corporation to conduct large-scale VIV investigations of risers, risers with fairings and other VIV suppression arrangements.

In-line and cross-flow forces were measured via load cells. The cylinder was constrained to vibrate in the cross-flow direction only with motions being measured via a yo-yo potentiometer and two accelerometers.

A testing case matrix was designed to validate R8's behaviour in post-critical flow. It included the following aspects:

• 30 tests of the R8 large scale models in the fixed, free, and free with additional damping conditions.
• Tests conducted up to 4.5 m/s (a Reynolds Number of 1.6 million).
• Ramp testing conducted in accordance with previously validated methods of Resvanis [3] and steady state tests at various reduced velocities were also conducted to further validate the results
• Different modes of testing to capture other properties/characteristics

The results were:

~30% reduction in drag forces on drilling risers with no loss of buoyancy

Reduced dynamic fatigue loading

Demonstrable VIV suppression

Potential annual savings of up to USD$12M per offshore rig

Improved safety for personnel

The R8 model became the basis for AMOG's patented Longitudinally Grooved Suppression (LGS) Technology, which won a number of awards including a High Commendation in the Victorian Engineering Excellence Awards, the AIRG Medal for Australasian Technological Innovation, and a Maritime Australia Innovation Award.

LGS is now manufactured under license by , for dynamic risers (flexible riser, steel catenary riser, SLR, umbilical risers), fixed risers and jumpers/spools, and pipeline spans, and has been successfully deployed on a number of assets around the world (click for video link to more information).

As offshore energy looks to renewable wave and floating wind technology, LGS is evolving to provide solutions for the cables that are needed to transfer power from these devices.

If you would like to know more about LGS, please contact Hayden Marcollo at AMOG, Matrix Composites & Engineering ( Aaron Begley , Peter Pezet ), or Matrix's US representatives at ABCO Subsea.

Thanks goes to the following individuals and teams who contributed to the successful development of #LGS: James Venning and Monash University for their work running the small scale, sub-critical Reynolds regime testing and providing photos of the manufactured LGS modules, Matrix Composites & Engineering for funding the post-critical Reynolds Number testing, Ron Ryan and the team at Oceanic Consulting Corporation for their work in running the high Reynolds testing program in St. John’s, the NRC-C in St. John’s, Canada for the use of their tow tank, Dr Themistocles Resvanis from Massachusetts Institute of Technology for his extensive advice and assistance during the large scale testing, and Professor Kim Vandiver from Massachusetts Institute of Technology (#MIT) for his advice on ramp testing and the additional damping tests.

References:

1. Talley, S., and Mungal, G., 2002, “Flow around cactus-shaped cylinders.” Center for Turbulence Research, Annual Research Brief.
2. Taggart, S., and Tognarelli, M.A., 2008. “OMAE2008-57047: Offshore Drilling Riser VIV Suppression Devices – What's Available to Operators?”. From Proceedings of the 27th International Conference on Offshore Mechanics and Artic Engineering OMAE 2008 Portugal.
3. Resvanis, T.L., 2014, “Vortex-Induced Vibration of Flexible Cylinders in Time-Varying Flows.” PhD Thesis. September, 2014. MIT.

LGS is a Registered Trademark of AMOG Technologies Pty Ltd.