Lip interference is often augmented through use of a garter spring. A garter spring is a helically coiled spring formed into a ring (see Figure 50). If present, the garter spring rests in a radiused groove molded into the head section of the lip. Seals without springs are common in applications in which the fluid being sealed has relatively high viscosity (such as grease). Because thick fluids don’t flow very readily (and thus require a fairly large leak path to be problematic), a sealing lip without a spring will generally suffice. If the fluid is thin, however, it can flow more quickly though a much tinier space. A spring-loaded lip may be needed to make sure less viscous fluids such as water and oil don’t escape.

The axial distance between the centerline of the garter spring and the contact point is known as the R value. A positive R value means the spring centerline is located toward the air side of the seal relative to the contact point, and this is desirable. A negative R value means the spring centerline is located toward the fluid side, which will result in immediate leakage if the R value is still negative after the seal has been installed on the shaft.

A spring fulfills two main functions. First, it contributes to the total radial sealing force, or load, between the lip and the shaft. (Load is also based on the sealing lip’s inherent “beam force,” as well as the “hoop force” generated when the lip is stretched outward slightly during installation onto the shaft.) Second, the spring also helps make sure the desired amount of load is maintained even when the lip material itself might swell and soften due to, for example, chemical exposure at high temperatures. An example of radial load loss is shown in Table 19. (Note: 1 N/mm = 5.71 lb/in)

A lip that has swollen away from the shaft is less able to maintain consistent contact with the shaft without the aid of a spring. Inconsistent contact makes the development of a leak path likely. The spring artificially stiffens the lip, and this helps hold the lip in place. Seals without springs will leak sooner than seals with springs. For example, consider an NBR seal for a 76.2 mm (3.000 inch) shaft operating at 2165 RPM in SAE 30 engine oil at a sump temperature of 107° C (225° F). Without a spring the average life of the seal is 480 hours; the life more than doubles to almost 1000 hours when a spring with an eight-ounce tension (2.224 N) is added to the seal.

The standard spring material is hard-drawn carbon steel wire, an economically priced, general-purpose material. This material is typically designated using a four-digit number, such as 1070, that corresponds to a system developed by the Society of Automotive Engineers (SAE). This system, which mirrors the system of the American Iron and Steel Institute (AISI), assigns a four- or five-digit number to each type of steel. This number is based on the differing levels of carbon and other elements present in the steel. The first digit denotes the primary alloying element (such as a “1” for plain carbon). The second digit indicates the presence of other elements. The last two digits specify the amount of carbon in the steel (in hundredths of a percent).

For example, a designation of 1070 indicates plain carbon steel (1) with no alloying elements (0) and a 0.70% carbon content (70). This may not seem like much carbon, but a little bit goes a long way in adding toughness to a material. High carbon steels (such as are used for hammers and chisels) might contain up to 0.95% carbon, but that’s still less than 1%! The standards for wire used in the production of shaft seal springs call for a carbon content in the 0.50% to 0.95% range (SAE 1050 to 1095, AISI C1050 to C1095).

It is recommended that the garter springs be heat treated, particularly if the spring will be exposed to temperatures of 100° C (212° F) or higher in service. Wire for use in seals that will face exposure to the elements are typically treated with a rust preventative. In highly corrosive environments, stainless steel wire (SAE 30302 to 30304, AISI 302 to 304) may be needed. In some cases, this stainless steel may be treated using a nitric acid solution (a process known as passivation) to further reduce the chemical reactivity of the metal. Springs incorporated into shaft seals that will be used around food or water are required by the Food and Drug Administration (FDA) to be made of stainless steel to prevent development of—and contamination by—rust. Stainless steel wire is more expensive than carbon steel wire, thus adding to the overall cost of the shaft seal.

Whether made of carbon steel or stainless steel, garter springs are produced in three stages. First, reels of metal wire are coiled and cut to length to produce straight springs. The tolerances for the wire diameter are shown in Table 20.

One end of each spring is tapered during the coiling operation to form a nib. The ends of each straight spring are joined together by back-winding each end of the spring, then inserting the nib end into the open end and screwing them together, creating what is called a nib joint. (Back-winding is required to eliminate a twisting tension generated when the ends are joined together (see Figure 50). If the ends have not been back-wound properly, the spring may twist into a figure eight shape after assembly.) Circular springs with a specific assembled inner diameter (AID) result. The tolerances for the AID are shown in Table 21.

The manufacturing process creates an initial tension (N) in the spring by back-winding the coils so that force is required to pull them apart. Spring rate (N/m) is defined as the force required to elongate (deflect) a spring a given amount. Initial tension and spring rate combine to determine the total tension (or spring load) generated by a given spring. The result can be shown as a spring load versus deflection chart (see Table 22). It is recommended that the initial tension be 50 to 80% of the total spring load at the working spring deflection. The tolerance for total spring load is the greater of ± 0.14 N (± 0.5 oz) or ± 20% of the nominal load at the desired working deflection.

A shaft seal lip without a spring is said to be unsprung; thus, the amount of interference between an unsprung lip and a shaft is known as unsprung interference. With a spring in place and the seal installed, the interference between the lip and shaft is known as sprung interference.

Because the springs exert inward tension on the lips, the inner diameter of the springs determines the final diameter of the sealing lip. When the seal is installed, the spring stretches and contributes to the amount of load applied to the shaft. As shown in Table 23, the tolerances for lip I.D. are dependent on shaft diameter.

ANATOMY OF A SHAFT SEAL MAIN PAGE