More Thoughts on Fastening

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

 

If I tighten a bolt/screw to yield, can I use it again or is it broken??

This question is on the minds of a lot of fastener engineers and people who use fasteners all the time. Lets examine what happens when you tighten a bolt into the yield area of a steel material. First of all lets look at the fastener close up and personal!

Fig 1

A threaded fastener is usually a straightened piece of steel wire of a given diameter and some length. It is then sent through a sizing die and drawn to the final diameter and then sent to a header that shapes and forms the head by cold heading. Now we have a cylinder with a hex form on one end for driving the finished fastener to tighten it. It next goes to a thread rolling station where the threads are rolled on the opposite end. It then goes to heat treat for hardening (or the bar may have been hardened prior to thread rolling and the threads are rolled on the hardened stock.) When the threads are rolled they physically yield and upset the steel to the final thread form to the measurements designed.


This method of making threads is known as cold forming and makes for very strong threads. The resulting dimensions are the finished outside diameter (OD) of the fastener, the pitch diameter (PD) and the thread root diameter (RD). The OD is self-explanatory, the PD is the diameter where the pitch of the threads is measured and also the theoretical point of contact with the mating threads of the nut or threaded hole. The RD is the smallest diameter on the shank of a standard bolt. (Fig.3 below)

Fig 2

OK lets now look at a thread on the bolt and a look at the yield process. Unwrap a single thread form a bolt and you end up with a right triangle. (Fig 2)The shortest side will represent the pitch, the hypotenuse will roughly represent the PD and the side adjacent to the pitch will roughly represent the RD. Remembering that the RD is the smallest diameter on the bolt it is logical to assume that the bolt will yield in the RD (which it will).

Fig 3

If we mount the bolt in a fixture and pull the bolt in tension until it yields you will see the yield stresses show on the surface of the RD and the yield value will be some load number. The yield point is defined as the point where the load / elongation curve leaves the linear portion and enters the plastic range (Hookes law - Fig 1). The load is noted to be some number in lbf (pounds force). This point, an average, found after several samples have been loaded to the yield point, is known as the tensile yield strength of the fastener. If we then use a fixture that measures the clampload during tightening of the fastener with a nut or tightening into a block we will find the yield load to be somewhat lower than the tensile test. This load will be typically about 5-10% lower, depending on several factors including friction. The main reason for the difference is the addition of the shear forces induced by the rotation of the bolt during tightening and the friction in the joint. All this points to the fact that a bolt that is tightened to the yield point does not develop the tensile yield load. The load combination of torsion, creating shear stress and the tensile load developed by stretching the bolt will cause the bolt to yield in the thread root at a load lesser than the tensile yield. (When I worked in the fastener lab, this is one of the factors that never correlated and my equipment was always blamed for the discrepancy even though I had proven the calibration).

Fig 4

Now let us look at the thread root diameter and find out how the yield stress is distributed. The stressed length (with concern for yield) of the bolt is perceived as the length of the bolt from the first fully formed thread to the first fully engaged thread of the nut plus ½ thread due to thread bending under load. (see Fig 4). Actually, because the first yield occurs on the surface of the root diameter, the length is much longer that the above-mentioned length. Looking at the triangle of the unwrapped thread we can see the length of the root diameter exposed is longer than the pitch. So, in reality the yield length becomes the root diameter circumference times the number of pitches from the first fully formed thread to the first thread engaged by the nut. I know that this explanation sounds like drivel, but consider a coil spring being stretched out. Each coil is being stressed equally therefore the spring can stretch much farther than a solid bar with the same starting length. It is not quite the same thing happening because the coils are free and the threads are not but I think you can visualize the idea.

At the beginning I mentioned thread rolling after heat-treating the bolt. This calls for special thread forming dies and results in very strong threads due to the residual stress left due to cold working. These types of rolled threads are stronger than the regular rolled threads that are heat-treated after rolling and are used for special applications that require strength and smaller size. For example connecting rods in engines. These fasteners will not usually fail in the threads rather the shank or transition area of the first formed thread, when tested in a tensile machine.

Another note to consider is the difference in shear stresses induced by tightening the bolt from the head end as opposed to tightening the nut. I expect this is where the variation of installed yield strength comes from.


I ran some tests years ago and found that if I set up a test where I would yield a bolt by driving the head and then turn an additional 20 degrees I could find out whether or not the bolt was reusable. I set up my equipment to automatically tighten to some load that was the average of several tests to yield and then turn an extra 20 degrees, I then loosened the bolt and repeated the test again using the same bolt until it failed. The average times of tightening the bolts were 29 times to failure. There fore I concluded that I could reuse the same bolt at least 3 to 4 times before throwing it out. Of course the decision to reuse the bolt depends a lot on the type of joint it is used in, for example connecting rod bolts are a safety problem so I would not reuse them more than twice. Head screws or headbolts perhaps 2-3 times. Other places where bolts are tightened to yield could be cam bearing straps, oil pump drive gear fastener, flywheel to clutch plate assembly and crankshaft to clutch assembly as well as the vibration dampener on the crankshaft and the crankshaft bearing cap to block bolts.

Home work assignment: Battery cable connector end bolts. Have fun!


George Lorimer, RETIRED GM POWERTRAIN