March 2020 Newsletter

Fatigue Failure of an M16 Flanged Bolt Newsletter - PDF Version

  Investigating fastener failure can provide invaluable insight into the correct application and assembly of fasteners. In the image below we can see the failure of an M16 flanged fastener of strength grade 12.9. This fastener failed as a result of fatigue failure.

  Following a detailed investigation of the assembly, the torque specification was too low. The torque specified was 180 Nm that resulted in a preload of 71000 N. This preload was insufficient to prevent the forces applied to the bolt opening the joint causing high alternating stresses to be sustained by the bolt. The problem was resolved by increasing the tightening torque applied to the bolt so that a higher proportion of its yield strength was used.
Fatigue failure of an M16 bolt

  The strength of the fastener was not being utilised, reducing the preload that the fastener was capable of providing. A tightening torque of 300 Nm was subsequently specified that resulted in a preload of 119000 N that was sufficient to sustain the forces applied to the joint.

  In many applications the clamping force provided by tightening fasteners is of critical importance in determining the success, or otherwise, of the structural integrity of an assembly. A great deal of attention is often placed on ensuring that bolted connections are installed in a controlled manner so that a predictable clamping force is achieved. The most popular controlled method of tightening a threaded fastener is by applying a specific tightening torque. Below the yield point of the fastener, the relationship between the applied torque and the clamp force provided by the fastener, is linear. That is, double the torque and you double the clamp force. Once tightened, the clamp force provided by a bolted connection can decrease. The decrease can occur without any rotation of the thread, as in the case of stress relaxation, embedding, creep and similar effects, or, the bolt  or nut may rotate decreasing the clamp force as in the case of self-loosening. Subsequently, concern over the loosening of bolts, in many applications, necessitates quality control measures to check their tightness. Tightness is usually assumed to be the measured torque value but in the majority of applications, it is the clamp force rather than the torque that is the critical factor. The problem is that there is no low cost method of assessing the clamp force provided by a previously tightened fastener. Currently, the tightness of a bolt/nut assembly is usually assessed by a torque based method, the approach is referred  to as torque auditing.

Did you know?

  If a nut is untightened immediately after being tightened, the torque needed to untighten it will be less than that needed to tighten it. This is due to the portion of the torque which actually stretches the bolt always acting in the untightening direction. Hence when a nut is immediately untightened, usually about 20% less torque is needed than was needed to tighten it in the first place.

  As the time passes from when the nut was tightened, the torque needed to untighten it tends to increase. After half a day or so, the release torque can be typically up 10% greater than the tightening torque. This is due to a number of effects including embedding of the contact surfaces and changes in temperature affecting the friction conditions. Such effects typically increase the friction and hence a greater release torque is required. Because of such friction changes, the use of torque auditing methods (measuring the backoff - or crack-off torque or the tightening or crack-on torque) to assess if the bolt being tightened correctly is only accurate if it is completed shortly after the assembly was tightened.

Corroded nut

  The greater the changes in the temperature and environment experienced by a bolted joint, the more rapid are the changes in the friction and subsequent change to the release torque. In sub-sea applications, sustaining elevated or low temperatures, significant changes in the friction conditions can rapidly occur. The properties of any lubricant will change over time, which in some applications, can lead to galling of the surfaces resulting in a dramatic increase in the release torque i.e. by as much as 50% to 100% of the tightening torque.

  Over a prolonged time period, oxidation and corrosion of the thread interfaces and nut face can occur making removal of the nuts problematic without destroying the bolt. In such circumstances nut splitters and similar measures are needed for nut removal.

  To continue reading this article please head to

Subscribe to Bolt Science's Newsletter

* indicates required