Talking about Torque

An article by George Lorimer, Retired;
formly of the G.M. Powertrain Fastener Lab.

George Lorimer

There is a TV preacher that exhorts,"Tough Times Never Last, Tough People Do!" This is an excellent philosophy to live by because it means if you have faith and knowledge you can overcome many of lifeís obstacles. Iím not going to preach a sermon, I just want to point out that we, collectively, cause some of our own tough times by resisting change, even when itís for the better.

In the engineering community, change is a way of life and for the most part it is well accepted. We have design changes, parts changes, procedure changes and people changes. When people change jobs on the way through their career, they often are bombarded with change that interferes with their productiveness... or so they think.

When procedures have been established for many years and the feeling is, "thatís the way we have always done it," I tend to get just a little suspicious of these procedures. They may be in place because Ďitís cheaper to do it that wayí, or Ďmy boss prefers to have this done this wayí. Does any of this ring bells? Probably!

Now letís get to the meat of this article. The name of the game is "customer satisfaction", anyone disagree? We all are here to do our respective jobs with several end goals in mind. Customer satisfaction , cost savings, deproliferation of parts, robust designs and reliability of product, and of course retiring to a summer cottage with a clear conscience and a bundle of money.

One of the changes Iíve been directly involved in is the use of new tightening procedures for the new 4-cylinder engine and chassis and body assemblies. These procedures involve the use of computer controlled tightening such as torque threshold plus an angle of rotation to achieve a given clampload with an acceptable range of deviation. Others are torque control with angle monitor and/or shut off. Some are just advanced torque control that require the use of a computer for monitoring the torque rate based on time. Others are procedures that contain algorithms that can predict the tension in the fastener to some degree based on several constants and derived information gathered during rundown of the fastener.


So, what does it all add up to? good question and I hope I can explain where Iím coming from!!

One change deals directly with one of the oldest and most accepted assembly tools.......torque. It is easy to implement and measure and control. It is also fairly easy to inspect. Oops, I hope Dr. Deming isnít listening. "Donít inspect in quality."

Well, Iíve broken the ice so now I guess I have to either jump in or over.. Letís begin with Ďwhat is torque?í Webster defines torque a couple of different ways (no pun), the first is a torture device that was worn around the neck, either an iron band or chain. The second is a force that tends to produce a rotation or twisting.

Now, if you apply torque to a bolt or screw the bolt or screw turns and if the bolt or screw is in a hole with matching threads the bolt/screw will be threaded into the hole. Yeah..We all know that. So what? O.K. We can now utilize the effect of the torque to some extent. We can cause the bolt/screw to become tight and exert force that can be used to hold parts together.. makes sense, Iíll go on.

The use that the bolt/screw can now be put to is unlimited for assembling pieces. Some pieces wonít be fussy and can be fastened together by just twisting the screw with so much twisting force. Other joints need to be fastened together to a given load so it can withstand the service loads and external loads as well. External loads can be generated by such phenomena as temperature change, either hot or cold or both. The loads can be varied by using dissimilar materials together during a temperature change. A thermostat can be made of dissimilar metals and is called a bi-metal temperature switch because as it is heated or cooled the movement caused by forces generated by the bi-metal either breaks or makes contact in an electric circuit. In a joint of different materials the materials will move relative to each other by the difference of their coefficients of thermal expansion. These joints require a clampload that is more precise than most joints require. A bolt has to stretch to be utilized as a spring, and a bolt is a very stiff spring, it has to be tightened in a manner that will give repeatable clampload and be fairly evenly distributed around the joint if it has multiple bolts. As a joint is moved by temperature forces the parts scrub against each other. The parts are being pressed together with a large force from the bolt and are generally under high pressure. The shearing forces generated by the temperature excursions are resisted by the shear load in the joint. These can be calculated by shear load equals friction times the normal load times the number of fasteners in the joint. Iíve broken some more ice, and now we have to talk about one of the biggest variables that affects the proper tightening of a joint...friction...

When you tighten with a torque wrench, you are in effect measuring the amount of friction being generated by the mating surfaces in the joint and as the clampload is increased so is the torque. You can see that there is some relationship between torque and clampload. This relationship is determined by the amount of friction in the joint as well as the clampload being developed... The really big problem is that you cannot measure or predict what the absolute value of the friction is in the joint. If we could , there would be no need for tightening procedures other than torque. We can however, minimize the effect of friction by using lubricants, and zillions have been tried with varying degrees of success. among these have been grease, chicken fat, oleo, mayonnaise, graphite, teflon, plastics and encapsulated formulas that smash and liquify under pressure. however sophisticated they have all fallen short of the goal of absolute friction control. What can we do if we canít control this continuous variable thing called friction?? The simplest is to utilize the fastener itself to obtain the clampload by rotating a given angle from some starting point called a threshold. The threshold is usually a low level of torque chosen because the clampload variation is less at low torque. the bolt is then turned to an angle and the result is an acceptable level of clampload. And you say ....why? And I say, because now that we are turning through an angle we no longer care about friction and its effect. Remember the thread around the bolt is a continuous spiral and if one turn is straightened out you have an inclined plane that has the height of one pitch of the thread in 360 degrees or one turn.


If we turn the bolt ľ turn then the threads will engage the mating threads and stretch the bolt ľ of the value of the pitch, almost. The stretch is actually somewhat less than the ľ pitch because the joint compresses and uses some of the pitch value, so if the pitch was 0.060" then ľ pitch would be 0.015" and the joint might use up 0.005" and the bolt would only stretch 0.010". This 0.010" stretch represents a finite clampload and will be repeatable every time the bolt is stretched that amount. This simple fact is the basis for the tightening procedure we call Ďtorque-angle Ďand it is also the basis of the sophisticated tightening procedure called ĎLRM Ď (logarithmic rate method) used to tighten some fasteners during final assembly.

We started out talking about change and whether or not we can accept it and get on with the job. Even though the concept of torque angle tightening is well known there are people that are not ready to accept this change. Letís look at some reasons why!

Q. How can I inspect the results of a joint that has been tightened using torque angle? I canít measure the threshold torque after itís tight and I canít measure the angle after the fact.

A. First, you donít inspect the tightening process after the joint is tight; Rather, you monitor the process during the tightening procedure by testing the threshold torque and measuring the angle as it is being turned. This is accomplished by using Ďsmart toolsí or tools that have transducers incorporated in their design and are controlled by process controllers.

Q. Arenít these tools very expensive?

A. I can answer that with a question. What is the cost of quality? And the answer to that is, the cost of doing it wrong.

Q. Isnít there some way of verifying the tightening process other than having a computer printing out miles of paper?

A. Yes; When a tightening procedure is recommended, a specific range of clampload is desired. This clampload can be measured in the bolt ultrasonically thus verifying the tightening process.

This is a good place to talk a little about the Raymond bolt gage. the bolt gage is really a glorified fish finder. you drop the transducer overboard and if there are fish in range you will see a blip on the screen or tracing depending on the type. This blip tells you about how far below you the fish are and if they are big enough you may see several blips that indicate individual fish. The Raymond bolt gage works the same way, by sending out a signal from the transducer down the length of the bolt and receiving it back to the transducer. The time is measured and if you know the speed of sound through a particular substance then you can relate the time to distance or the length of a bolt. The bolt gage has an additional feature that allows us to measure a change in length of the bolt due to load.(Modified by the Stress Factor of the material.) This is accomplished by the bolt gage Ďrememberingí the original length of the bolt and subtracting it from the length is reads when the bolt is tightened. A bolt stretches linear and the change is proportional to the load applied.(Mr. Hookeís law) so now we have a means to relate the change in length of the bolt to the clampload in the joint.

We canít allay all the fears you may have about the changes in tightening procedures but you can help yourself to a great extent by becoming more knowledgeable about joints and fasteners. There is a wealth of knowledge available in engineering through your fastener engineers. This article is not to say that no joint can be tightened using torque only, but that due to a changing workplace that demands quality builds the first time and every time so that we can meet the customer head on with a quality product, allows us the tools to do so, and many times this calls for change, radical change sometimes, and the tough knowledgeable people will win the battle.

Iím with Webster, sometimes torque is a pain in the neck.

George Lorimer, Retired.
G.M. Powertrain Fastener Lab.