Article "How Much Shake Can Bolted Joints Take?", Machine Design, 5 October 1972

In this document

This is a technical article published in the American magazine Machine Design on 5 October 1972, signed by Robert J. Finkelston, a research engineer at Standard Pressed Steel Co. pp. 24-27. The arrival stamp "DOCUMENTATION Ke1.30.6" and the date "17 APR. 1973" attest that this excerpt was forwarded to Paul Magès by the Citroën documentation department, with the handwritten annotation "Magazine excerpt concerning assemblies" p. 24.

The problem of vibration-induced loosening

The article explains that transverse vibration can cause a bolted assembly to lose up to 80% of its initial preload due to micro-slip at the contact surfaces, creating a net loosening torque p. 24. Three potential remedies are cited: increasing joint friction, eliminating all relative play, or using fasteners that are intrinsically resistant to loosening — the latter being considered the most industrially viable p. 24.

The test method

A "transverse vibration machine" is described pp. 25-26: a test bolt clamps a mobile base against a fixed base via a threaded sleeve, a forced transverse movement is applied, and a load cell records the loss of preload as a function of the number of cycles. The cross-sectional diagram and the photograph of the machine detail the components (drive spindle, eccentric, transverse force cell) p. 25.

Experimental results

Several factors are ranked by decreasing order of importance regarding vibration resistance: amplitude of dynamic movement, effectiveness of the locking system, thread pitch, initial preload, and hardness of the bearing surfaces p. 24. Some notable numerical data:

Figures 1 to 4 illustrate these relationships in the form of amplitude/cycles and preload/cycles curves pp. 24-26.

Context

Machine Design is a reference American technical magazine for mechanical engineers; this type of technical popularization article circulated widely in automotive design offices at the time.

At Citroën, the reliability of bolted assemblies is a direct issue for the DS and its derivatives, whose hydropneumatic suspension — invented by Magès — includes numerous fasteners subjected to constant dynamic loads (road vibrations, wheel shocks).

Archivist's perspective

This piece is not a production by Paul Magès but a documentary watch item sent to him, likely because the issue of vibration-induced bolt loosening directly concerned the design of the running gear and suspension elements on which he worked.

The delay between publication (October 1972) and internal distribution (April 1973, p. 24) illustrates the operation of the Citroën documentation department, which provided engineers with referenced foreign technical intelligence (reference Ke1.30.6).

It seems that this archive is part of a broader set of Magès' concerns for safety and mechanical resistance, as evidenced by other items in the collection dealing with kinetic energy and forces sustained during impacts, such as the note "Method for breaking or not breaking eggs" or the calculation notes on safety distances, written during the same period (March-April 1973).