A unique combination of computer modeling and field testing results in improvements to a basic but critical link in the stimulation equipment chain.

Hydraulic fracture stimulation is one of those oilfield operations where success or failure hinges on a few moments when key pieces of equipment must perform as expected. If a fracturing job is shut down prematurely due to equipment malfunction, the consequences can be costly. All of the elements in the pumping system must be engineered to perform reliably.
One of these simple yet critical elements is the flexible piping used to transfer the fracturing fluid from the fracturing pumps to the wellhead. In turn, the critical links in this piping system are the swivel joints used to connect the pipe sections. These joints must be flexible enough to handle the vibration and pulsation that comes from the positive displacement pumps, while at the same time containing the high pressure and managing the abrasive nature of the fracturing fluid.
Recognizing that the trend in fracture stimulation was toward higher pressures and larger volumes, FMC Fluid Control began a program 10 years ago to collect field data, compare it with proprietary computational fluid dynamics (CFD) analyses and identify the critical design factors that influence swivel joint reliability. The result was the TripleStep (TSi) swivel joint, which has been shown to significantly outlast competitive swivel joints by a factor of 2.5 to 5 in lines carrying abrasive fracturing fluids.
Features
Swivel joints usually are assembled from two 90° elbows and short straight pieces connected by sets of sealed bearing races. These terminate with male and female hammer union ends (Figure 1). The TSi swivel joint differs from competitive swivel joints in three important ways: thickness of the elbow wall, design of the ball races and metallurgy.
During FMC Fluid Control's development program, classical stress calculations, finite element analysis and burst testing indicated the inner arc portion of the elbow sections is subjected to the maximum stresses due to internal pressure. Accordingly, the TSi elbow is constructed with a thicker wall on the inside. These elbows are manufactured by bending a concentrically upset-forged steel tubing section, a process that produces an elbow thicker on the inside of the bend. Competitive joints are made by offsetting the bore before bending, producing an elbow with about the same wall thickness on the bend outside and inside.
Analysis of field performance data also confirmed CFD modeling and flow testing results indicating the most aggressive erosion occurs not on the bend region at the middle of the elbow, but farther along near the exit of the elbow and beneath the bearing races. This led to a distinct feature of the TSi swivel joint, the stepped ball race (Figure 2), which allows greater thickness in the male ball race end without reducing the flow bore of the swivel joint or increasing its weight. This design allows the bearing in the three races to support about equal amounts of hydrostatic end loads. In a nonstepped swivel joint, much more stress is seen in the cross section under the bearing race nearest the bend.
A third feature of the TSi swivel joint is improved resistance to brinelling and erosion resulting from improvement in the metallurgy of the alloy steel used in the joint's manufacture. Brinelling, a form of mechanical damage, is a permanent deformation of the bearing surfaces where the balls contact the races, resulting from excessive contact stress.
Flow testing
FMC tested its design in a full-scale flow loop facility constructed so that TSi and competitive swivel joints could be subjected to the same erosive proppant flows to obtain comparative wear data. A typical test used a slurry of 12-lb/gal 20/40 Northern Sand in water, with flow rates set to achieve 40-ft/sec velocity through the swivel joints. At each 48-hour mark, wall section thickness measurements were taken to gauge erosion at four locations: elbow wall, behind the ball race, at each of three bearing race walls and along the packing face. This was done for both elbow pieces of the swivel joint. After each 48-hour test period, the old sand was replaced to maintain abrasiveness.
These thickness measurements were compared to the manufacturer's minimum allowable wall thickness specifications and the number of hours logged before erosion had reached that minimum was determined. Competitive swivels allow a much smaller erosion wear in the critical area under the male ball races, compared to the 3-in. TSi. The 3-in. TSi took two-and-a-half to five times longer to reach the minimum spec than the competitive swivels under identical flow conditions. At the point where a competitive swivel had reached its limit, the comparably eroded TSi still had a higher burst pressure.
The tests showed conclusively that the maximum erosive wear takes place in the wall under the male bearing race. While elbow wall thickness can be monitored ultrasonically, erosion under the bearing races is difficult to detect without disassembly of the swivel joint. Should the bearing race erode through during fracturing, immediate job shutdown would be necessary.
By modeling the effects of pressure and erosion on swivel joints, focusing on the areas most likely to fracture under stress and testing the results under realistic operating conditions, FMC Fluid Control has developed a joint that helps complete challenging fracturing jobs.

Comments