How about another torque discussion? Assembly processes

dwperron

Trusted Information Resource
There have been many discussions concerning torque tools and their care and feeding and calibration.
We are having a different discussion, so if this has been covered please point me in the right direction.
Otherwise, here we go...

We have assembly processes here that have a requirement to properly attach fasteners with the correct amount of torque.
The drawings will give a tolerance for the torque setting value for the fastener, let's say ±10% for the sake of discussion.
We use various tools that are calibrated with accuracy ranging between 2% and 6%.
This kind of an accuracy ratio is not optimal, but this is the real world.

My question regards the ±10% tolerance on the drawing for torque values. Does this kind of a tolerance make sense to the mechanical engineers and metallurgists out there? Are there any studies to show how much variability you can have in the torque settings for fasteners without affecting the security of the joint? How can you tell how much is "too loose" or "too tight".
 

Jen Kirley

Quality and Auditing Expert
Leader
Admin
Good day dwperron,

My experience with prescribed torques comes from submarine mechanical assemblies. I do not remember the allowable error we had, but based on end application (generally water tight seals on saltwater, oxygen or high temperature, high pressure steam boundaries) the allowable error depends on the fastener size, material type and heat treatment, the size and material type/heat treatment of the housing/flange, the type and size of gasket (if any) or surface finish on machined surfaces, and the cost of getting it wrong.

Based on these variables it is impossible to guess here what is the appropriate error in torque. I would defer to provided specifications.
 

GRP

Involved In Discussions
There have been many discussions concerning torque tools and their care and feeding and calibration.
We are having a different discussion, so if this has been covered please point me in the right direction.
Otherwise, here we go...

We have assembly processes here that have a requirement to properly attach fasteners with the correct amount of torque.
The drawings will give a tolerance for the torque setting value for the fastener, let's say ±10% for the sake of discussion.
We use various tools that are calibrated with accuracy ranging between 2% and 6%.
This kind of an accuracy ratio is not optimal, but this is the real world.

My question regards the ±10% tolerance on the drawing for torque values. Does this kind of a tolerance make sense to the mechanical engineers and metallurgists out there? Are there any studies to show how much variability you can have in the torque settings for fasteners without affecting the security of the joint? How can you tell how much is "too loose" or "too tight".

If I am being too obvious kindly disregard, but torque leads to a tension stress on the fastener. Given the material properties and the geometry of the fastener you want an upper bound stress below yield, with a safety factor dictated by the industry and/or application. So at least, the upper specification limit should be constrained by this.

I have read a technical paper from a fastener manufacturer that stated that only about 10-30% of the effort expended in tightening a screw ends up as "tightening force" of the joint, the rest being absorbed to overcome thread friction and screw head to workpiece friction.

At any rate, too tight is breaking the fastener, too loose is any torque which does not "use" or take advantage of the fastener properties to withstand torque-induced stress to the effect of tightening the joint. There must be plenty of standards and tables for this.
 

Marc

Fully vaccinated are you?
Leader
too tight is breaking the fastener
I think the issue is that "too tight" will not necessarily actually break the fastener at the point/time of assembly. It may partially strip the threads on the fastener, or the threads in the threaded hole it is being tightened into which may not be obvious to the assembler. Or it may stress the fastener its self causing it to fail (e.g.: through fatigue) at a time when the finished product is in use.
 

John Predmore

Trusted Information Resource
There is no simple answer because a torque specification is a multi-dimensional multi-faceted dynamic reliability problem. While threaded fasteners have been around for hundreds of years, our modern understanding of fastener joints has changed since I was in engineering school. For example, split ring lockwashers, which were commonly used 40 years ago, are now discouraged because they can make a fastener joint less reliable. Every fastener application is a unique design study. The good news is, with the Internet, there are many resources including chapters of design textbooks available to a mechanical designer. There is much more information online than is possible to copy and paste here. But be wary of simplistic answers to a complex question.
 

Miner

Forum Moderator
Leader
Admin
I think the issue is that "too tight" will not necessarily actually break the fastener at the point/time of assembly. It may partially strip the threads on the fastener, or the threads in the threaded hole it is being tightened into which may not be obvious to the assembler. Or it may stress the fastener its self causing it to fail (e.g.: through fatigue) at a time when the finished product is in use.
Too tight also becomes a concern with a gasketed joint as the gasket may be deformed too much or damaged when over torqued.
 

dwperron

Trusted Information Resource
Good day dwperron,
Based on these variables it is impossible to guess here what is the appropriate error in torque. I would defer to provided specifications.

I understand, and I did not do a good job at describing what I am searching for.
I am trying to determine how much torque is recommended for a joint (and how that number is determined) and to see if there are any "industry standards" regarding what might be considered an acceptable range of torque values for that joint.
 

Miner

Forum Moderator
Leader
Admin
I am trying to determine how much torque is recommended for a joint (and how that number is determined) and to see if there are any "industry standards" regarding what might be considered an acceptable range of torque values for that joint.
Establishing a torque specification correctly is a fairly complicated process. The first step is determining the clamp force required for the joint. Then you determine the number and size for the fasteners. This will establish the axial force required for each fastener, which can then be translated into a torque specification. The allowable range of torque must be determined by the allowable range in clamp force.

See Engineering Fundamentals of Threaded Fastener Design and Analysis and POCKET GUIDE TO TIGHTENING TECHNIQUE for more detail.
 
Top Bottom