Tolerancing helical springs

In my 30 years experience as a spring designer, tolerancing of helical springs is probably the trickiest part of the job and the most common issue with customers. One reason: The metal wire from which springs are formed can have widely varying mechanical properties between batches, especially tensile strength. Operators making the springs have no control over raw-material properties so they must compensate for the variations during manufacturing.

Most spring-coiler machines, notably compression-spring formers, use standard tooling to vary factors such as coil count, pitch, body diameter, and wire feed. Typically, spring rate k (lb/in.) or load at height are the design goals. First consider spring rate.

Wire size, body diameter, and active coil count, all influence spring rate. Wire diameter is held to industry standard tolerances, such as ASTM A228 for music wire. Wire diameter is often set in the design phase because it's not practical or cost effective to change it on the shop floor after setup. That leaves body diameter and coil count.

But body diameter, especially for compression springs, generally have tight OD or ID specs. Coil-spring rate goes as the cube of diameter, which makes it highly sensitive to small changes. But spring diameter may need tweaking several thousandths of an inch to reach a target spring rate in some cases, putting it out of tolerance. Diameter of automobile valve springs cannot be changed significantly to gain rate because springs must fit collars, valve guides, and pockets, for example.

A focus on statistical process control, Six-Sigma, and C_pk integrity further complicate changing rates with diameter. Springs that hold spec on mean rate may be rejected because of a low C_pk on body diameter, the same parameter used to tweak rate. Ironically, some blueprints ask to meet both specs simultaneously!

A change of active coil count is a second, simpler option for adjusting rate. Wire-feed mechanisms control coil count and are highly accurate as a rule, especially those in NC machines. Many NC wire formers have diameter controls that compensate for diameter "drift" to keep a process in line.

Spring makers will appreciate that designers not lock-in coil count but instead make it a reference dimension. In fact, designers requiring springs with a precise rate should not tolerance coil count at all. This makes it possible to hold diameter and adjust coil count to obtain a precise rate. The approach also speeds production and helps cut delivery times.

The next issue is load at height. Spring rate alone controls how much load is produced at a given height. Springs made with wire on the low side of a diameter tolerance will also have lower spring rate and subsequently deflect more for a given load.

Most spring makers change free length to boost load height and leave rate unchanged. Knowing this, designers should always allow a generous tolerance on free length or, better yet, make free length a reference dimension. Free-standing length is rarely important on preloaded springs. In fact, slightly different free lengths on compression springs from order to order may mean a spring maker is correctly adjusting loads. From a spring maker's perspective, tightly toleranced free lengths make fast, accurate, load corrections more difficult, which is reflected in higher costs and longer delivery times for customers.