In recent years, we have witnessed a growing concern of manufacturers and retailers regarding the quality of the jewelry products. Closer attention is being paid to such issues as general aesthetic appearance, surface finish, color consistency, strength and overall durability of jewelry.


Responding to the increasing quality requirements, we have made an attempt to develop some specific methods for testing the finished jewelry products including various chains, bangle bracelets, castings and components. This work was presented at the Santa Fe Symposium in 1995 (1). Presently, we use these tests in the ongoing QA procedures as well as to evaluate the experimental prototypes. At the 1996 Symposium, the importance of QA, and mechanical testing in particular, of the finished jewelry products was emphasized by Santala (2).

A reliable QA procedure requires a sufficient data base. Since the neck chains and bracelets together constitute a significant portion of the gold jewelry sold worldwide, we have concentrated our effort on extending the data base for mechanical properties of different chains. In this paper we present a range of data on mechanical strength of the chains made with the wire: solid rope and herringbone, and with the assembled links: stampato and omega. We also show how some chains fail and make a correlation of the strength with the chain size and design features.

Ganoksin is sponsored by

Chains Made with Wire

The metallurgical aspects of the manufacture of such chains were recently reviewed by Aldo M. Reti and Philip A. Fossaluzza (3) and D.P. Agarwal (4). Under the tensile load, these chains show a significant stretch, usually more than 10%, before they break. Originally, we thought that the load at 10% stretch would be a good measure of the overall chain strength (1). It appears, however, that the break load is the parameter which is used more commonly. Also, the advantage of using the break load is that it does not require sophisticated equipment to measure. Below we present data for two types of chains – the solid rope and herringbone.

Solid Rope

Typical tensile stress-strain curves for the solid rope neck chains of different gauges are plotted in Figure 1 (the corresponding diameter size of the cross section is given in parenthesis). They show that the strength increases fairly rapidly with the chain size. The 14 gauge (1.8mm) chain, for example, breaks at the load around 15 lb (6.8 kg), whereas the break load of 25 gauge (3.4mm) chain exceeds 60 lb (27.2 kg). In other words, a two-fold increase in diameter results in approximately four-fold increase in strength. It appears, therefore, that the wire size as well as the total cross section area of the chain both determine the strength.

Measuring the strength of the chains made with the different alloys (some of which are intrinsically stronger than others), we have determined the strength range of the solid rope chains. Figure 2 shows such a range of break loads for different chain sizes. The rupture mode of the solid rope chains is similar to that of a single wire. Naturally the solid rope shows much higher strength than the hollow rope chains. For example, 18 gauge hollow chain breaks at about 5 lb (2.27 kg) (1), whereas the same size solid chain breaks at 30 – 35 lb (13.6 – 15.9 kg) showing a 6 – 7 fold increase.


Probably the most popular neck chain and bracelet style, herringbone chains are produced by numerous manufacturers who use their own proprietary alloys and techniques. The chain size is determined by its thickness and width. The width is more or less standard for all the manufacturers and is given in mm – 2.5mm, 3mm, 3.5mm, etc. However, the thickness may vary significantly depending on the design and the initial wire size. We have grouped all the tested chains by the width (mm) and the weight per unit length (g/in).

Ganoksin is sponsored by

The table below shows the alloy (R – regular, H – hardenable), thickness (N – normal, L – light), the weight per unit length (g/in) and the break loads for the tested chains.

Table 1 shows that the chains made with the hardenable alloy are much stronger. Figure 3 shows the difference in break loads for the chains of normal thickness made with regular and hardenable alloys. The Table also shows that the reduction in weight by about 50% results in greater (more than 50%) reduction in strength.

highest in the edge areas. There is no direct relationship between the width or the wire size and the strength. Overall thickness is probably the most important factor.

Ganoksin is sponsored by

Chains Made with Assembled Links


These chains are made with the links stamped out of the flat sheet and connected by a wire pin as shown in Figure 6. They are used mainly to make bracelets. Under the tensile load, the links separate when the connecting pin breaks. This typical break mode is shown Figure 7. Therefore the tensile strength of the Figure 4 shows a segment of the stretched herringbone chain just prior to the break. The break occurs in the edge area of the chain where the wire is bent as shown in Figure 5. The stress concentration is not uniform in these chains and is actually the chain is determined by the strength of the connecting pin. Overall, the strength of the stampato chains is fairly high and may vary between 25 – 60 lb (11.3 – 27.2 kg) depending on the alloy and the size of the wire.

The stampato chain may break prematurely at the loads as low as 12 lb (5.4 kg) when the link is weaker than the connecting pin. This situation is shown in Figure 8. Since the links in the stampato bracelets usually have a large exposed surface, the resistance to dent becomes another important issue. With the link thickness being equal, the harder the alloy of the link, the more it resists to the indentation load. In the previous tests of the bangle bracelets described in (1), we used a right-angle probe to introduce the indentation load and measured a yield. The surface pattern of the stampato links is not as uniform as that of the bangles.

Therefore, in order to obtain consistency we needed a smaller probe. A 1/16 inch ball probe from the Rockwell hardness tester appeared to be an ideal solution. Unlike in the previous tests, we have measured the load at a dent of 0.01 inch (0.25 mm) and found that this has more practical meaning than just a yield point since one can visually see such a dent. We have measured an indentation load at 0.01 inch dent for various bracelets made with different alloys. The load range appears to be between 5 – 15 lb (2.3 – 6.8 kg).


In the omega neck chains and bracelets, the flexibility is provided by assembling the links on the reinforcing wire, mesh or strip. The tensile strength, therefore, practically depends on the strength of the reinforcing material. Numerous tests show that the overall tensile strength of these chains is quite high – in the range of 40 – 100 lb (18.1 – 45.4 kg).

Ganoksin is sponsored by

The omega links also have a large exposed surface. To evaluate the resistance to dent we have used a right-angle probe and measured the load at 0.01 inch (0.25 mm) dent. This load depends mainly on the link thickness and the hardness of the alloy, and varies between 5 and 25 lb (2.3 – 11.3 kg) for different chains.

Discussion and Conclusions

  1. The tensile strength of the chain products have a wide range of variation depending on the alloy and size. In general, we may consider the value of 7 – 8 lb (3.2 – 3.6 kg) as a borderline between “strong” and “weak” chains. However, some small size and lightweight chains may break at the loads below 7 lb (3.2 kg) without being defective.
  2. The range of variation of the indentation load at 0.01 inch (0.25 mm) also is wide. In general, we have noticed that when this load is below 5 lb (2.3 kg), the visible dent can be made by a finger nail.
  3. We have obtained a range of data on the mechanical strength that can be incorporated in the QA procedure for the finished chain products.
  4. The evaluation of the strength of chains made with the different materials shows that, for some styles and designs, it is beneficial to use the hardenable alloys to manufacture the chain links and components


This paper is based on the presentation given by the authors at the Santa Fe Symposium on Jewelry Manufacturing Technology, May 1997 and published in the 1997 Symposium Proceedings, edited by D. Schneller.


  1. D.P. Agarwal, Greg Raykhtsaum and Marinko Markic, “Mechanical Testing of Finished Jewelry and Components”, The Santa Fe Symposium on Jewelry Manufacturing Technology, Edited by Dave Schneller, 1995, 367-382.
  2. Timo Santala, “The Weakest Link”, The Santa Fe Symposium on Jewelry Manufacturing Technology, Edited by Dave Schneller, 1996, 165-185.
  3. Aldo M. Reti and Philip A. Fossaluzza, “Precious Metal Chain”, The Santa Fe Symposium on Jewelry Manufacturing Technology, Edited by Dave Schneller, 1991, 287-307.
  4. D.P. Agarwal, “Metallurgical Aspects of Chain Manufacture”, Gold Technology, No.17, October, 1995, 16-18.