Page 154 - Failure Analysis Case Studies II

Failure Analysis Case Studies If
D.R.H. Jones (Editor)
0 2001 Elsevier Science Ltd. All rights reserved 139

INVESTIGATION OF FAILED ACTUATOR PISTON RODS

T. F. RUTTI* and E. J. WENTZELf
Boart Longyear Research Centre, PO Box 1242, Krugersdorp 1740, Republic of South Africa
(Received 3 February 1998)

Abstmet-Actuator pistons had failed after very short service life by intergranular crackingin a circumferential
weld. Analysis of the welds and subsequent heat treatments showed that the austenitic filler material used to
weld the carbon steel resulted in carbide precipitation during nitriding. Combined with high residual stresses.
these carbides promoted intergranular fracture at normal operational loads.
It is shown how the changes that had been made to the initial design and processing of the actuators were
not entirely compatible with each other and reduced the “metallurgical safety factor”.
Recommendations for improvements were successfully implemented. 0 1998 Elsevier Science Ltd. All rights
reserved.
Keywords: Welding, nitriding, heat treatment, preheat.

1. INTRODUCTION
Boart Longyear Research Centre was subcontracted to investigate the failure of actuator piston
rods. An adapter was welded to the end of each piston rod and fractures in the welds were resulting
in failures of the actuators. The pistons were subjected to two static loads of no more than 50 kg
onto which a 30-35 Hz cyclic load was superimposed. Down-time of the machines, and possible
injury to the operators, were endangering the manufacturer’s position in the market.
This report illustrates how the changes-some of them beyond the manufacturer’s control-that
had been made over the years to the initially sound materials selection, design and processing of the
piston rods, resulted in a gradual erosion of the “metallurgical safety factor”. Eventually, the
combination of materials and processes resulted in an inadequate product that was unable to
withstand operational loading conditions.

2. BACKGROUND

A carburised adapter made of En32 (carbon case-hardening steel) was press fitted into the end of
the piston rod made of En9 (“55” carbon steel). The area on the adapter that was to be welded was
blanked off during carburising with a refractory compound to prevent an increase in carbon content.
The rod and adapter were machined so that after press fitting there was a single-V groove with a
90” included angle that allowed the two components to be circumferentially welded using a single
pass GTAW with an AWS A5.4 E312-16 filler rod (29 Cr, 9 Ni-generally recommended for welding
steels of low weldability). No preheating was applied. The completed piston rod was then either
nitrided or phosphated, depending on customer requirements. It appeared that only the nitrided
piston rods were failing.
After reports that the first few components had failed by fracture of the weld, two “rosette welds”
(essentially plug welds or tack welds through a hole in the rod onto the adapter) were added where
the adapter extended into the rod. These welds were supposed to act as shear pins and prevent the
adapter from being pulled out of the piston rod once the circumferential weld had cracked.

*Author to whom correspondence should be addressed.
t Present address: Combined Design Engineers, PO Box 754, Rondebosch 7701, Republic of South Africa.
Reprinted from Engineering Failure Analysis 5 (2), 9 1-98 (1998)