Except for perhaps the valvesprings on a 10,000-rpm race engine, the most highly stressed engine components in any performance engine are the rod bolts. When the piston passes across top dead center (TDC) on the exhaust stroke, the crankshaft yanks hard on the connecting rod, and high engine speeds add exponential stress to the process. There is no cylinder pressure on this rotation which means the crank is trying very hard to peel the cap right off the end of the connecting rod. The only thing keeping all this together is a pair of small fasteners.
It’s a thankless job, repeated 50 times a second at 6,000 rpm. So if that gives you an inkling of the stress that the rod bolts must withstand, perhaps it will also deliver a sense of importance around the proper technique for ensuring those fasteners will live for as long as you care to run them. Just like creating an arm bar in Jiujitsu or cooking a steak to perfection on your backyard bar-b-que, there’s a technique to this that works perfectly if you just follow the steps.
Before we get into the technique, let’s first look at why putting a torque wrench to a fastener is not as precise a process as you might think. Our information comes from the professionals at ARP who have decades of experience in manufacturing and testing high performance fasteners. A bolt can be viewed like a coil spring. If we compress and then expand a coil spring within its accepted range of motion, it will perform this task thousands of times. Tighten a bolt to the point just shy of where it fails and it can be used over and over again just like a coil spring pulled apart (think of this like a Slinky toy). But torque a fastener beyond what is called its “yield point” and the bolt will quickly fail, much like a spring that is pulled too far apart and will not contract normally.
When torque is applied to a bolt, the torque spec is designed to stretch the fastener to just short of its yield point. The difficulty with the application of torque is there are multiple variables that can affect how much of the applied torque is used to overcome friction and how much actually stretches the bolt. The overwhelming majority of applied torque (around 75- to 80-percent) on a bolt is actually used just to overcome friction. The small remaining torque is applied to accurately stretch the fastener to just below its yield point.
The number of variables that affect tension on a bolt are both numerous and nefarious. Friction between the threads, the condition of the threads, thread overlap, the smoothness of the interface between the underside of the bolt head and the connecting rod, the material strength of the fastener, what kind of lubricant used, and the accuracy of the torque wrench are just a few of the main variables.
ARP has performed all these tests. They know, within a very specific range, exactly how much load is necessary for a given ARP rod bolt to achieve its ideal clamp load. With regard to torque specs, the ARP approach is to establish a torque that will stretch the bolt to 75-percent of the fastener’s yield strength. This is also the rod bolt stretch number.
As an example, for an ARP 8720-steel, 3/8-inch small-block Chevy connecting rod bolt (P/N: 134-6003), the torque spec is 55 lb-ft using ARP’s Ultra-Torque lubricant, and that produces a rod bolt stretch figure of 0.0055 to 0.0060-inch. The only way to ensure the rod bolt has achieved this proper clamp load is to measure it with a rod bolt stretch gauge. The gauge fits over the bolt to measure its free length. Then the bolt is tightened until the stretch spec is achieved. We will look at how a variable like the kind of lubricant used can have a drastic affect on friction and therefore the amount of stretch actually imparted into the rod bolt when only a torque wrench is used.
To hammer this lesson home, we lost a 355ci small-block Chevy on the dyno when a rod bolt failed because it had not been properly stretched. The under-torqued rod bolt nut eventually backed off after several hours of dyno testing, completely destroying the engine in the process. The cause was traced to a torque wrench that fell short by 10 lb-ft combined with new rod bolts and using engine oil as a lubricant. The resulting tolerance stack caused the failure. When we disassembled the engine, the rod bolt nut was laying in the bottom of the oil pan. All of the remaining rod bolts were badly under-stretched as well. Lesson learned.
We will not sugar coat this. Yes, performing a stretch measurement on each of 16 rod bolts on a V8 engine does require more time. But the insurance factor weighs heavily in your favor. One technique we’ve found is to establish a minimum torque on the rod bolts using your torque wrench and a stretch gauge before they are put into the engine. Determine a torque that will create a stretch within 0.001-inch of the final stretch spec. This demands installing the stretch gauge on each rod bolt several times to achieve the spec, but it makes the final assembly go a little faster.
Stock rod bolts do not come with a rod bolt stretch number and in this situation, torque is your only option. ARP also recommends torquing new bolts or studs with nuts several times ahead of the final assembly. This creates a wear pattern underneath the bolt head or in the area underneath the rod bolt nut. This will reduce the overall friction in this area and help stabilize the torque numbers. Follow the manufacturer’s recommendation and seriously consider certifying your torque wrench for accuracy. You might be surprised at what you find.
Rod Bolt Stretch Test
The following test was performed on a 7/16-inch ARP 8740 through bolt used on a Manley connecting rod. The specs supplied by Manley call for 0.0060 – 0.0065-inch of stretch. Manley also specs 95 lb-ft of torque to be used with the Manley lube and 90-100 lb-ft when using 30wt engine oil.
The point of this test was to measure the effect of lubrication (or lack thereof) on bolt stretch. This will show that as torque remains the same, as better lube is used the bolt stretch increases because more torque is used to stretch the bolt instead of overcoming friction. One column also shows the effect of tightening the bolt five times in a row and what affect that has on stretch.
The test reveals that 30wt oil is inconsistent when torque is applied multiple times and how the UItra Torque maintained a more consistent bolt stretch after five consecutive applications. Also note that none of these applications of torque achieved the minimum spec of 0.0055-inch. We had to increase torque to 110 ft-lbs to achieve 0.0061-inch of bolt stretch.
This illustrates why relying on simple torque – even when using the proper lubricant is not sufficiently accurate. Our torque wrench was not calibrated, but the odds are good that it is not accurate at 90 to 100 ft-lbs so we are not certain what torque was actually applied. So this test reinforces the point that the only accurate way to establish bolt preload is with a bolt stretch gauge.
Bolt Condition | Rod Bolt Stretch | Torque Bolt 5 Times |
Dry | 0.0035 | 0.0035 |
30wt Oil | 0.0040 | 0.0035 |
Ultra-Torque | 0.0050 | 0.0049 |
Proper bolt stretch range: 0.0060 to 0.0065-inch