FLUID HANDLING AND FLUID POWER

This section was edited by Executive Editor Harry Hutchinson.

PIPE FOR ULTRA DEEPWATER

THE PURSUIT OF OIL AND GAS RESERVES OFFSHORE IS ENTERING EVER DEEPER WATERS. The Brazilian oil company Petrobras, for instance, has just installed a pipeline in 1,600 meters of water in the Canapu field in the Campos Basin. The Roncador field off Brazil has many installed pipes beyond 1,800 meters.

Recently discovered subsalt fields off Brazil lie at depths to 2,400 meters. There are wells already drilled to that depth around the world, although only a few have entered production. 

Ultra deepwater scenarios, at depths beyond 1,500 meters, require very thick-walled steel pipelines or pipe-in-pipe systems, which are expensive and difficult to install because of their excessive weight. A development called sandwich pipe is intended mitigate some of those disadvantages. It is a new concept composed of two concentric steel pipes separated by and bonded to a polymeric annulus. The design provides a combination of high structural strength with thermal insulation.

The concept has been patented by Segen F. Estefen of the Federal University of Rio de Janeiro and two of his colleagues in the Ocean Engineering Department, Theodoro A. Netto and Ilson P. Pasqualino. They are working with two other researchers, Xavier Castello, one of Estefen’s doctoral students, and Su Jian, a lecturer in the university’s  Nuclear Engineering Department, and with two commercial partners, TenarisConfab and Dow Brasil, to optimize the concept design for commercial production.

Sandwich pipe prototype, above, was made by TenarisConfab
for experimental tests. Collapsed section, above, was
subjected to testing in a hyperbaric chamber.

So far, experiments on prototypes and numerical analyses have shown promising results. Studies indicate that the technology can reduce steel costs and lower submerged weight, which will facilitate ultra-deepwater installation. Thermal and structural performance are close to current pipe-in-pipe systems, which have independently designed steel pipes and thermal insulation in the annular space, and whose outer pipe is designed to withstand external pressure by itself.

The stiffness, insulation, and adhesion properties of polymer contribute to the reliability of sandwich pipe. Collapse and buckle propagation under external pressure using a hyperbaric chamber as well as bending and adhesion tests are some of the experiments executed in small and prototype scales to examine the ability of the sandwich pipe to withstand ultra-deepwater conditions. The progress of this research has been presented in recent years in the Pipeline and Riser Technology Symposia at the International Conference on Ocean, Offshore, and Arctic Engineering, which are organized by ASME’s Ocean, Offshore, and Arctic Engineering Division.

In this collaborative research project TenarisConfab S/A fabricated prototypes that have been tested at the university’s Subsea Technology Lab. The results will be shared and Tenaris will apply for the rights to fabricate and commercialize the design. Castello’s research is supported by Brazil’s National Petroleum Agency, through its Human Resources Program.

Numerical and experimental studies for the ultimate strength under combined external pressure and bending indicate that sandwich pipe is reliable for application in water depths below 3,000 meters, where pressures reach 300 atmospheres. Among the advantages over single-wall pipe, the new design has a substantially higher bending capacity for equivalent external pressure, with similar steel weight and less submerged weight.

A prototype specimen from Tenaris was subjected to extreme pressure and collapsed under 35.7 MPa (simulating a depth of about 3,550 meters). The inner pressure used in the test was atmospheric, a condition that might occur if oil flow was stopped during operation. "

Numerical-experimental correlations are used to calibrate finite element models that can be used for parametric studies.

A finite-element model of collapsed and partially propagated sandwich pipe.

Three-D finite element models incorporate cohesive elements that connect the annular layer to the steel layers to simulate adhesion properties. Shear tests are used to evaluate the adhesive bonding capacity and to correlate the results with those from numerical simulations. The numerical results showed strong influence of the degree of adhesion between polypropylene and steel on the sandwich pipe’s ultimate strength. Inner and outer pipes of this study are 6 5/8 inches and 8 5/8 inches in diameter and made of quarter-inch thick mild steels.

An ultimate strength, obtained with maximum shear stress of 14 MPa, corresponds to the theoretical perfect adhesion condition in this situation. The results show that even reducing the maximum shear strength to about 10 percent (1.5 MPa, obtained for the worst condition of the shear tests), the structure will still withstand pressure at the design depth of 3,000 meters.

Polyurethane elastomers are being developed by Dow Brasil S/A to meet the thermal and mechanical requirements. The research project used 500 and 750 kg/m3 densities for numerical simulations. As the density increases the mechanical strength improves, but insulation quality worsens. These two densities appear to be good intermediate options in a wide density range.

Sandwich pipes with these materials were numerically collapsed and compared to a PIP system with low-density polyurethane foam as annular material. They were designed for a hypothetical oil and gas field, requiring a 6-inch inner diameter and maximum heat-transfer coefficient of 1.5 W/m2 °C in a 2,500-meter water depth.

The sandwich pipes use less steel and are lighter than PIP. The polyurethane D500 option had 65 percent of corresponding PIP steel weight, half of its submerged weight, and had the same external diameter of 10 ¾ inches. The D750 is as thermally efficient as pipe-in-pipe. With a 12 ¾-inch outer pipe diameter, sandwich pipe had 73 percent of the steel weight of PIP. All sandwich pipes were designed with an unfavorable 1.5 percent initial ovality for both inner and outer pipes, which yields low collapse resistance and increased stresses on steels and adhesion interfaces.

Results so far lead the team to conclude that, regardless of the optimization process to be applied to the product, the sandwich pipe concept has clear benefits regarding steel costs and installation weight, and is a good option for the ultra-deepwater fields.

REFERENCES

Estefen, S.F., Netto, T.A., and Pasqualino, I.P. (2005), “Strength Analyses of Sandwich Pipes for Ultra Deepwaters.” Journal of Applied Mechanics. Vol. 72, pp. 599-608.

Jian, S., and Estefen, S.F. (2005). “Thermal-Hydraulic Analysis of Heavy Oil Transportation in Heated Sandwich Pipelines.” 24th International Conference on Offshore Mechanics and Arctic Engineering, Proceedings of OMAE 2005. Halkidiki, Greece.

Castello, X., and Estefen, S.F. (2007), “Limit Strength and Reeling Effects of Sandwich Pipes With Bonded Layers.” International Journal of Mechanical Sciences. Vol. 49, Issue 5, pp. 577-588.

Castello, X., Estefen, S.F., Leon, H.R. and Fritz, M.C. (2007), “Collapse of Sandwich Pipes With Different Annular Materials.” Rio Pipeline Conference and Exposition 2007. Rio de Janeiro, Brazil.

Castello, X., and Estefen, S.F. (2009), “Design Aspects and Benefits of Sandwich Pipes for Ultra Deepwaters.” 28th International Conference on Offshore Mechanics and Arctic Engineering. Proceedings of OMAE 2009. Honolulu, Hawaii.

Segen F. Estefen is full professor of ocean structures and subsea technology in the Department of Ocean Engineering and head of the Subsea Technology Lab at COPPE, the Institute for Post-Graduate Studies and Research in Engineering at the Federal University of Rio de Janeiro. He is currently the coordinator of the Symposium on Pipeline and Riser Technology for OMAE conferences. Xavier Castello is a D.Sc. student under the supervision of Estefen.

SEMI-ACTIVE COMPRESSOR VALVE
By Alan S. Brown

A new electromagnetic valve system promises to sharply reduce the cost of maintenance on large compressors, according to the Southwest Research Institute.

According to the manager of the rotating machinery section in SwRI’s Mechanical and Materials Engineering Division, Klaus Brun, who was the principal developer of the technology, “Four production version valves were installed in a high-speed compressor in early April 2008 and have been performing with no valve failure to date, far exceeding the performance of conventional valves in place at the same location.”

Valves are the weak points on the big piston-driven reciprocating compressors used to move natural gas through pipelines. SwRI estimates that the average valve system lasts only about six months and costs $30,000 in parts and labor to replace.

The short lifetimes are not hard to understand. New pipeline compressors operate at 500 to 1,000 rpm and 4,000 to 8,000 hp. They snap their valves open and closed, and eventually crack the valve plates on which they are mounted.

Most compressors damp valve motion to slow failure. Passive valves use springs to slow their momentum, but they are quickly compromised by dirt, debris, sticking, improper lubrication, and eventually by high-cycle fatigue. Active valves that use external pressure transducers or shaft encoders for control tend to be more complicated and expensive.

Brun’s design uses an electromagnet to produce a soft landing on the valve seat when closing and on the valve guard when opening. Instead of a spring, it uses a magnetic coil that does double duty. First, its motion produces a signal in an electric inductive motion sensor that senses the valve’s position. The sensor activates the coil, which applies an opposing force to slow down the valve prior to impact.

The system is called semi-active because it is activated only by gas pressure and only controlled just before impact. The design eliminates springs (the weak points in passive designs) and costs about one-third as much as other active designs, according to Brun. As a result, it can last up to 10 times longer than passive valves.

The technology also gives compressors more surge capacity, letting valves open and close faster during peak use without sacrificing valve life. Unlike other active systems, the compressor will continue to operate (albeit noisily) if the electrical power goes off.

SwRI developed the technology under a program funded by the Gas Machinery Research Council, the U.S. Department of Energy’s National Energy Testing Laboratory, and BP plc. Cook-Manley Co. plans to manufacture the valve for industrial oil and gas applications.        

NEW FOR OLD
By Harry Hutchinson

A 30-year-old petrochemicals plant in France was motivated to go wireless because of its age. The plant at Carling Saint Avold, owned by Total Petrochemicals, makes a range of products, including ethylene, propylene, methane, styrene, polyethylene, and polystyrene. The products are derived by steam cracking, which mixes petroleum with gas, oil, or naphtha and heats them with steam at 800ºC.

The company was particularly interested in monitoring the boiler of its aging steam cracker. It wanted to add several temperature measurement points. By comparing the internal and external temperature of the boiler walls and identifying heat loss, the company can calculate the material’s resistance and infer its thickness. It will help the company judge when to replace the boiler.

A complication with the plan was that the steam cracker wasn’t the only thing aging at the plant. So was the wiring. According to the company, there was evidence of corrosion, infiltration, and armature degradation.

It was to avoid the cost of replacing or adding wiring to connect the additional monitoring equipment that Total installed a wireless system made by Emerson Process Management.

Total Petrochemicals’ Carling Saint Avold plant. An Emerson Smart Wireless system avoided the cost of rewiring for new monitoring equipment.

Total put eight Rosemount 648 wireless temperature transmitters directly on the exterior of a boiler drum and a Smart Wireless gateway on the roof of a building about 300 meters away. Two more temperature gauges were installed in strategic areas between the boiler and the gateway.

The gateway relays the data from the transmitters to a distributed control system via the Modbus serial communications protocol. Operators can keep an eye on the boiler’s condition from the control room.

The company estimated that it saved the cost of installing about a kilometer of new wiring.

An additional wireless device, a Rosemount 3051S pressure transmitter, is providing redundancy for an older wired device that gives a critical measurement in the plant. The transmitter is also an additional path back to the gateway.

According to Emerson, each Smart Wireless device can act as a router for neighboring devices. Smart Wireless uses self-organizing mesh technology to provide reliable communications that avoid obstacles like temporary scaffolding, new equipment, or a parked construction trailer that can interfere with wireless signals. When they meet an obstruction, transmissions are rerouted along the network until they find a clear path to the Smart Wireless gateway.

FLOWMETER GROWTH
By Peter Easton

There is strong growth in the worldwide Coriolis flowmeter market, according to a new research study. The study, The World Market for Coriolis Flowmeters, 3rd Edition from Flow Research Inc. of Wakefield, Mass., reports that the worldwide flowmeter market totaled $662 million in 2007, and will grow at a projected compound annual rate of 7.5 percent through 2012. By 2012, the worldwide Coriolis market is projected to reach $950 million. 

The highest growth rates will be in China and other Asian countries, the Middle East and Africa, and Latin America. The rapid growth in these regions is due to the need for high-accuracy flow measurement in new process and power plants.

Coriolis meters, the highest-accuracy flowmeters on the market according to Flow Research, are most widely used in the chemical industry. However, their use is growing very rapidly in the oil and gas industry. Even though gas was previously a difficult application for Coriolis meters, now their use in gas flow measurement is widely accepted.

In addition, Coriolis flowmeters are increasingly used in the custody transfer of petroleum liquids, where they are replacing positive displacement meters. The growing number of approvals by industry organizations, including the American Gas Association and the American Petroleum Institute, is encouraging the use of Coriolis flowmeters for custody transfer applications.

Another significant trend is the growth of Coriolis flowmeters in large line sizes—above 6 inches. More suppliers have entered this market. Coriolis flowmeters are also beginning to be used for steam applications.  

TREATING HYDRAULICS LIKE SERVOS

ROGER DOWNING, A SENIOR PROJECT ENGINEER AT THE BRADBURY CO. INC. OF MOUNDRIDGE, KAN., RAN INTO A PROBLEM BENDING METAL RAILS. His firm uses a servomotor and ball screw to apply the linear force to make the bend. “We were running right up to the edge of our controllability, and we would occasionally get a fault and the servo would run into the hard stops and tear everything up,” he recalled.

Ordinarily, this would leave Downing with one of two choices: scale up the motor and screw, or switch to a hydraulic system. The latter was the more cost-effective choice, but then Downing and his customers would have to learn to program the hydraulic controller in addition to the electromechanical controllers.

Instead, Downing opted for a third choice: Bosch Rexroth Corp.’s new HNC100-3x digital numerical controller. The unit not only controls electric drives, but generally treats electrohydraulic drives as if they were servos. By decoupling fluid physics from the system’s functionality, the HNC100-3x lets engineers program their hydraulic drives as if they were servos, according to Bosch Rexroth’s new technologies development manager, Michael Liedhegener.

Downing originally learned about the controller while researching a different application. It proved a natural choice when he took on the rail bender, though it required some adjustment. Bradbury uses Bosch Rexroth servos, but had to upgrade to a newer version of the company’s software. He also found the hydraulics harder to parameterize (not surprising, since hydraulics, unlike servos, are nonlinear). Still, it proved easier than working with a separate hydraulic controller, and Bosch Rexroth was there to help him develop the control strategy.

The HNC100-3x controls up to four axes at a time, and offers true axis synchronization. Like other controllers, it has grown smaller and more powerful, allowing faster scan rates and up to 11 freely configurable digital I/Os. The company offers its WIN-PED software to simplify configuration, startup, and diagnosis.

Each HNC100 comes with local CAN-bus for exchanging data and synchronizing individual servo drives. It also communicates with higher-level control systems through Profibus-DP, Interbus-S, and CANopen fieldbuses. In order to control both electrical and hydraulic drives, it must use a Sercos interface. Sercos is an open standard introduced by Bosch Rexroth and used in its electrical drive control products.