'set' in a threaded joint ? 2

[Gerry45]

Im just starting to investigate the technology of bolted joints and have come across the term 'set' or 'embedment'. The literature I have suggests that this phenomena, which is caused by a crushing/collapse of the contact surface asperities etc, tends to reduce preload built into the joint. It must therefore be allowed for in the initial calcs.

What puzzles me a little is that we are concerned with metals here. I didn't think that metals showed any significant tendancy to 'relax' if kept under load, as would say a polymer material. I would imagine metals to act more like the situation one would get if putting a nut and bolt through a couple of rough-sawn planks of wood. The surfaces would crush DURING the tightening process, and tightening torque would not begin to rise until this crushing process had been completed. In such a case, loss of preload due to 'set' at the time of torquing, and would therefore not need to be allowed for in the initial calcs. I do understand however that in doing the same with lumps of plastic there would be a loss of preload with time due to the visco-elastic behavior of these materials.

Clearly I must be missing something. Perhaps someone would like to relieve me of my ignorance !?

 

[TVP]

Aluminum and magnesium alloys exhibit significant losses of preload AFTER initial tightening. Steel joints do not suffer nearly as much from this phenomenon. If elevated temperatures are likely, or if vibration is present, then loss of preload must be taken into account.

Set/Embedment is what you are describing-- plastic deformation of the bearing surface(s) during initial tightening.

 

[CoryPad]

In Handbook of Bolts and Bolted Joints, Bickford refers to short-term relaxation (what you call set), which contrasts with stress relaxation (creep, viscoelasticity, etc.). Sources of short-term relaxation include: surface asperities, thread plasticity due to poor dimensional control and insufficient thread engagement, oversized holes, and lack of perpendicularity between fastener bearing surfaces and longitudinal axis.

VDI 2230 includes guidelines for calculating initial preload loss due to relaxation.

If you are looking for the best fastener information, try these:

Handbook of Bolts and Bolted Joints, available at:

http://www.dekker.com

VDI 2230, Systematic Calculation of High Duty Bolted Joints, available at:

http://www.beuth.de

(note: English translation of newest edition not available until January 2003, but well worth it)

ASTM Special Technical Publications 1236 and 1391, Structural Integrity of Fasteners, available at:

http://www.astm.org

 

[nashjp]

Gerry,

You are correct, most of the setting happens during tightening. Most of the setting has happened before you remove the wrench. After a few minutes it is very hard to measure any further setting.

However, external loading on the joint is the main cause of joint set or embedment. The joint must experience its maximum external load before it will stop setting. (If the max load is not beyond the capacity of the joint.) If the clamp load is insufficient to hold the joint together, then progressive setting can occur. This then leads to less clamp load and eventual loosening of the bolt after repeated cylces.

 

[Gerry45]

Nashjp,

Thanks, the mechanics of this are beginning to make some sense to me now. I hadn't previously considered the effect of the external load in creating that final bit of 'set' which causes a reduction in the initial preload.