Compression Spring Equipment, Design and Non-Axial Performance

By Mark Hayes

The equipment used to manufacture and to test springs is mostly unique and exclusive to our industry. As other articles in this edition will explain, you need a machine with high stiffness and great precision in order to make compression springs with close tolerances. You need a similarly endowed test machine to accurately measure the load and rate of compression springs.

Figure 1: Schematic relationship between shear force and lateral rate.

However good your spring coiling and testing equipment is, though, it is inevitable that an axially loaded compression spring will produce forces that are not axial and, usually, not wanted. That’s because you cannot make a perfect spring that is geometrically symmetrical and accurately square, despite the claims of the best manufacturers of spring coiling equipment about how good the latest generation of equipment has become.

The point of this Cautionary Tale is to relate that these non-axial forces are inevitable and will change as the spring is compressed. What is more, they can be measured and, by both manufacture and design, they can be minimized – but not eliminated.

A compression spring, when loaded, will exert shear forces and a torque. Because the axis of the main load is not coincident with the geometrical axis, the shear force will give rise to a tilting moment acting upon the end coil.

From the considerable research work undertaken by IST, it has been shown that the non-axial performance of a spring is directly related to the lateral rate (see Figure 1, below). The higher the lateral rate, the higher the potential non-axial forces. The lateral shear force can be as high as 15 percent of the axial load. However, as Figure 1 shows, the maximum lateral shear force may not occur at the maximum deflection because a spring tends to “square up” under load. We at IST know this because we build machines to measure these non-axial forces. These are conventional spring load testers with six load cells instead of one. The load cells are arranged so as to measure the magnitude of the forces at all positions of the available deflection range of the spring. Having measured these forces, they are then analyzed using a computer to quantify their magnitude and the position at which the forces are acting. Hence, the test machine will: measure the axial load, quantify the position at which this axial force is acting compared with the geometrical axis (the center of the top and bottom end coils, which have to be accurately located for this type of testing), and measure the shear forces and torque output.

Figure 2: Effect of number of total coils on shear force. These springs were accurately manufactured with exactly one dead coil at each end.

Having explained how the non-axial forces are measured, it is important to recognize that these forces are minimized by accurate spring manufacture. Laying-on the end coils at each end in an accurate and symmetrical way is the most important and difficult to visualize. The end squareness and parallelism are also obvious influences. Nevertheless, no matter how efficiently and accurately you coil your springs and grind their end coils, there will still be significant non-axial forces.

So how can these forces be minimized by design? The greater the number of coils, the lower the non-axial forces will be. What is more, a whole number of coils will give significantly lower non-axial forces than designs with x + ½ active coils, as shown in Figure 2, above.

If there are x + ½ active coils, there will be x coils contributing to the axial load on one side of the spring, but x + ½ on the other side; so it is easy to imagine why such springs give relatively high non-axial forces.

Much more could be written about this subject, so I encourage readers to request further discussion, specifying which aspects of this topic they would like more information on in a future Cautionary Tale.