Trends in hydraulic tube production during times of shortages, Part II

May 16, 2023

Wynn Kearns, Indiana Tube Corp.

Editor’s note: This article is the second in a two-part series about the market for and production of small-diameter fluid-carrying lines for high-pressure applications. Part I discussed the state of domestic supply of conventional products for these applications, which is scant. Part II discusses two unconventional products for this market.

Two types of welded hydraulic tubing specified by the Society of Automotive Engineers—SAE-J525 and SAE-J356A—share a common origin, and so do their written specifications. Flat steel strip is slit to width and formed into a tube by roll forming. After the strip edges are polished by fin roll tooling, the tube is heated by high-frequency electric resistance welding and forged between pressure rolls to make the weld seam. After welding, the OD flash is removed by a fixed tool, typically made of tungsten carbide. The ID flash is removed by a fixed tool or controlled to a designed maximum height.

This description of the welding process is generic, and many small process differences exist in actual manufacturing (see Figure 1). Nonetheless, the two have many mechanical characteristics in common.

Tubing failures, and failure modes in general, can be classified under tensile loads or compressive loads. In most materials, failure under tension occurs at a lower stress value than in compression. That is, most materials are much stronger in compression than tension. Concrete is one example. It’s quite strong in compression, but unless it’s cast with an internal network of reinforcing bar (rebar), it pulls apart quite easily. For this reason, steel is tested under tensile load to determine its ultimate tensile strength (UTS). All three hydraulic tubing specifications have a similar requirement: a UTS of 310 MPa (45,000 PSI).

Because pressure tubing requires an ability to withstand hydraulic pressure, a separate calculation and a destructive test, a burst test, may be required. A calculation can determine the theoretical ultimate burst pressure, which takes into account the wall thickness, the material’s UTS, and the OD. Since the J525 tube and J356A tube can have the same dimension, the only variable is the UTS. Providing a common tensile value, 50,000 PSI, the predicted burst pressure on 0.500- by 0.049-in. tube is the same for both products: 10,908 PSI.

Although the calculations predict identical results, one difference in practical application relates to the actual wall thickness. On J356A, the ID weld flash is controlled to a maximum dimension, based on tube diameter, as outlined in the specification. For J525, a product with the flash removed, the flash scarfing process often will undercut the ID intentionally about 0.002 in., resulting in localized wall thinning at the weld zone. Although the wall thickness is filled out by later cold working, residual stress and grain orientation may differ from the parent material, and the wall thickness may be slightly thinner than a comparable tube specified as J356A.

This may actually create a lower burst pressure scenario for J525 as compared to J356A.

Depending on the tube’s end use, the ID flash needs to be removed or flattened (or smoothed) to eliminate the potential leak path, chiefly for a single-wall flare end form. While it’s commonly accepted that J525 has a smooth ID and therefore has no potential for a leak path, this is a misconception. A J525 tube can develop ID striations from improper cold working, resulting in a leak path at the connection.

Flash removal starts by shearing (or scarfing) the weld bead from the ID wall. The scarfing tool, which is fixed to a mandrel that is supported by rollers, sits inside the tube just past the welding station. While the scarfing tool is removing the weld bead, the rollers inadvertently roll over bits of weld spatter, forcing them into the surface of the tube’s ID (see Figure 2). This is a problem for lightly processed tubing such as skived or honed tubing.

Removing the flash from the tube’s interior isn’t easy. The scarfing process turns the flash into a long, tangled length of razor-sharp steel. Although removing it is a requirement, removal is usually a manual and imperfect process. Lengths of tube containing strands of scarf occasionally leave the tubemaker’s premises and get shipped to the customer.

FIGURE 1. SAE-J525 material is produced in batches, an intensely capital- and labor-intensive undertaking. A similar tube product, made to SAE-J356A, is processed entirely on a tube mill outfitted with inline annealing, so it’s a much more efficient process.

For smaller tubing, such as fluid lines in diameters of less than 20 mm, ID flash removal usually isn’t all that important because these diameters have no additional ID finishing steps. The only caveat is that the end user just needs to consider whether the agreed flash control height will create a problem.

Best practices for ID flash control start with precise strip conditioning, slitting, and welding practices. In fact, raw material characteristics for J356A must be more stringent than J525 because J356A is more restrictive in grain size, oxide inclusions, and other steelmaking parameters because of the cold sizing process involved.

Finally, ID scarfing typically requires a coolant. Most systems use the same mill coolant as for the roll tooling, but this can be problematic. Despite filtering and skimming, mill coolant generally contains no small amount of metal fines, tramp greases and oils, and other contaminants. As a result, J525 tube requires a wash cycle in a hot alkaline bath or other equivalent cleaning step.

Condensers, automotive systems, and other such systems need clean tubing, and sufficient cleaning can be accomplished on the mill. J356A comes off the mill with a clean ID, controlled moisture content, and minimal residue. Finally it is a common practice to charge each tube with an inert gas to prevent corrosion and seal the end before shipment.

J525 tubing is normalized after welding, which is followed by a cold-working (drawing) operation. After cold working, the tubing is normalized again to meet all mechanical property requirements.

The normalizing, drawing, and second normalizing steps require transporting the tube to the furnace, the draw bench, and the furnace again. These steps entail other separate substeps, such as pointing (before drawing), pickling, and straightening, depending on the specifics of the operation. These steps are expensive, gobbling up untold resources in time, labor, and cash. A cold-drawn tube is associated with a 20% scrap loss in production.

J356A tubing gets a normalizing step after welding while it is still on the mill. The tube doesn’t touch the ground, progressing from the initial forming step to a finished tube in one uninterrupted series of stages on the mill. An as-welded tube like J356A is associated with a 10% scrap loss in production. All other things being equal, this means that J356A tube can be produced at a lower cost than J525.

While the two products perform in a similar manner, they are not identical from a metallurgical standpoint.

Cold-drawing J525 tubing requires a normalizing pretreatment twice, after welding and after drawing. The normalizing temperatures (1,650° F or 900° C) lead to surface oxides, which usually are removed after annealing by a mineral acid, typically sulfuric acid or hydrochloric acid. Acid pickling has a large environmental impact in terms of air emissions and a metal-rich waste stream.

Further, normalizing temperatures in the reducing atmosphere of a roller hearth furnace can deplete the carbon at the steel’s surface. This process, decarburization, leaves behind a surface layer that has a much lower strength than the original material (see Figure 3). This can be especially critical for thin-walled tubing. If the wall thickness is 0.030 in., even a slight decarburization layer at 0.003 in. thick reduces the effective wall by 10%. This weakened tube can fail in service from load or vibration.

FIGURE 2. The ID scarfing tool (not shown) is supported by rollers that ride along tube’s ID. Good roller design reduces the amount of weld spatter that is rolled into the tube wall. Nelson Tool Corp.

J356 tubing is processed in batches, which requires annealing in roller hearth furnaces, but it doesn’t end there. A variant, J356A, is processed entirely on a mill using inline induction, which is a much faster heating process than for a roller hearth furnace. This reduces the anneal time, which shrinks the opportunity window for decarburization from minutes (or even hours) to seconds. This provides J356A with a uniform anneal, absent of oxides or a decarburization rind.

Tubing made for use in hydraulic lines must be ductile enough for the requisite bending, flaring, and forming. Bending is necessary to get the hydraulic fluid from point A to point B, through a variety of twists and turns along the way, while flaring is the key to providing a method for making an end connection.

In a chicken-and-the-egg scenario, drawn tubing—which therefore has a smooth ID—was developed for single-wall flare connections, or perhaps the reverse occurred. In this connection style, the inside surface of the tube seals against a seat on a male fitting. To make a leakproof metal-to-metal seal, the tube’s surface finish must be as smooth as possible. This fitting originated in the 1920s for use in the U.S. Army’s nascent aerial division, the Air Corps. The fitting later became the standard 37-degree flare in common use today.

Other flares work for many situations, such as the bubble and double-wall flares (see Figure 4).

Since the COVID-19 period began, the supply of drawn tubing with a smooth ID has dropped substantially. The material that is available tends to have much longer lead times than in the past. This change in the supply chain can be resolved by re-engineering the end connection. For example, a request for quote that calls for single-wall flare and specifies J525 is a candidate for a substitute, a double-wall flare. This end connection can use any of the hydraulic tubing types. This opens the door of opportunity to use J356A.

In addition to the flare type connections, the O-ring face seal (see Figure 5) is in common use, especially for high-pressure systems. Not only does this connection type tend to leak less than a single-wall flare because it uses an elastomeric seal, but it’s more versatile—it can be formed on the ends of any of the common hydraulic tubing types. This sets up tube fabricators for a broader supply chain choice and better long-term economics.

The history of industry is full of examples of conventional products that get established so thoroughly that they develop a momentum, which makes it difficult for the market to change direction. A competing product—even one that is markedly less expensive and is shown to fulfill all of the requirements of the original product—can have a difficult time getting a foothold in the market if it’s looked upon with suspicion. This is often case when a purchasing agent or a specifying engineer considers an unconventional substitute for an incumbent product. Few want to take a perceived risk.

In some cases, a change might be not only warranted but necessary. The COVID-19 pandemic has caused an unexpected shift in the availability of certain types and sizes of bulk tubing for steel fluid lines. The product areas affected are tube fabrication applications for automotive, appliance, heavy equipment, and any others that use high-pressure lines, especially for fluid power.

This gap can be filled, perhaps at lower total cost, by considering established but niche types of steel tubing. To select the proper product for the application, a little research is necessary to determine fluid compatibility, operating pressure, mechanical loading, and connection type.

A close look at the specifications shows that J356A can be a true J525 equivalent. It’s available at a lower cost through a proven supply chain in spite of the pandemic. If dealing with the end form issues is less taxing than procuring J525, this could help OEMs resolve the logistical challenges in the COVID-19 era and well into the future.