Ron Carte, via CarCraft.com: I have noted a great deal of rhetoric lately about using a dial indicator to measure bolt stretch rather than a torque wrench to secure engine bolts. Being an old-timer with many builds under my belt, I feel compelled to offer my two cents' worth along with my questions.

 

 

This classified ad for an original-paint '70 Hemi 'Cuda four-speed was clipped from a Massachusetts-based auto trader back in 1983. The price is $4,800 firm. This 'Cuda was offered just before the first musclecar collector boom hit in 1984. At the time, $4,800 was a stiff figure; by 1985 it could have sold for four times that amount. Today this one-of-652-built '70 Hemi 'Cuda would easily trade for a quarter-million. Nuts, huh?

 

First, a torque wrench is employed to be sure head bolts, intake manifolds, and exhaust manifolds are equally tightened for a good mating surface with an engine block. Second, application of a torque wrench on rod and main-bearing bolts is to make certain that the caps of each were put to manufacturer's specifications, so that the bearings have the proper attitude of roundness for the assembly to spin freely.

 

Why would you throw away a perfectly good bolt system in favor of a stud? The only way you could measure bolt stretch on head and manifold applications would be to convert to a stud-fastening system. Also, how do we know if the bolt-stretch method on rod bolts and main-bearing studs puts them into a concentric attitude? I am perplexed.

 

 

What's this, Hemi wheels? Yes sir. The specific 15x6 stamped steel wheels that were factory-fitted to many Street Hemi B-Bodies between 1968 and 1970 have become highly sought- after and can trade for big dollars if they bear the correct stampings and date codes. Look for part number 2944169 stamped near the valve stem or on the inside of the hoop.

 

Jeff Smith: Let's see if we can eliminate some of your confusion, Ron. The whole process of attaching components with a fastener is really about creating a sufficient clamp load to produce either a proper seal, as between a cylinder head and the block, or to secure two components, as with a connecting rod and cap. To accomplish our task, we need a fastener-either a bolt or a stud-that is strong enough to secure these components without failing under load. This involves tightening the fastener with sufficient torque to create the necessary clamp load. We all know that insufficient tightening may create leaks or failed parts when stuff grenades. But overtightening can also cause fastener failure, which is just as bad.

 

 

Here's what happened when an ARP bolt was ground to clear the camshaft on a 383ci small-block Chevy stoker motor. This bolt was tightened to 0.0061-inch bolt stretch when it failed because too much material had been removed from the bolt head. Luckily, this happened on the workbench.

 

Frankly, anyone who works on mechanical devices that rely on fasteners to hold them together should have a working knowledge of bolts, nuts, and studs. The most common method of measuring clamp load is with bolt torque, so let's look at what it really measures. Let's take the process of applying a measured amount of torque, say 70 lb-ft, to a head bolt. The applied load that is measured by the torque wrench is actually the combination of several factors. First and foremost, it is a measurement of the resistance to movement created by the friction between the threads in the block and the bolt. But at the same time it measures the even greater friction between the bolt head and the cylinder head. Finally, we're also measuring a load applied to the bolt required to stretch the fastener a given amount. When it comes to creating sufficient clamp load to ensure that the head gasket does not fail, the only variable among the three we just mentioned that is of any value to create gasket sealing is the amount of stretch applied to the bolt. But because one end of the bolt is buried deep in the block where we cannot access it, we cannot measure the amount of bolt (or stud) stretch necessary to create the clamp load we need, so we're forced to substitute torque as a unit of measurement.

 

 

We spotted this freak sitting on a workbench. The oddball carrier-bearing pinch-bolt setup looks superweak compared to the beefy cap-and-bolt design used on most modern diffs. This weirdo is from a '64 Buick Riviera.

 

You may already have figured out that if we apply oil to the threads of the bolt and underneath the bolt head, it will reduce the amount of friction created as we tighten it. That means that more of the effort put into the torque wrench will be applied to stretching the bolt, which will increase the clamp load. But wait, what if we use that slick moly paste instead of oil? The paste offers far more lubricity, so that the stretch value increases slightly. In the ARP catalog, the generic torque spec for a 71/416-inch bolt using 30-weight motor oil is 82 lb-ft, while substituting ARP moly paste changes the spec to 65 lb-ft. Thus the paste removes 17 lb-ft worth of friction to create the same tension or stretch on the bolt.

 

Just to scare you even more, what makes you think that your torque wrench is actually creating the torque value that it claims? We've seen torque wrenches off by 10 lb-ft and more. And even if you have your wrench calibrated, that can only be spec'd at one load setting, such as 65 lb-ft. As you can see, there are dozens of variables involved in the process of tightening a fastener. This is why any bolt manufacturer will publish specific torque recommendations based on the tensile strength of the bolt and the kind of lubricant used on the fastener. But even these are not completely accurate. There are too many other variables such as whether the bolt is brand-new or has been used, since creating a wear pattern in the area between the bolt head and the part affects the frictional load. ARP calls for burnishing the underside of the bolt head by applying full torque with lube a minimum of five times to establish a common wear pattern before the friction is normalized. This is also why most head-bolt manufacturers like ARP supply precision-ground head-bolt washers to improve this interface.

 

 

Did you know that you can advance the timing on a Chevy by grabbing the vacuum advance and pulling? This rotates the distributor counterclockwise.

 

As to the question of bolts versus studs, let's stick with the cylinder-head application. When bolts are torqued, a twisting motion is created between the bolt head and the threads. With a stud and nut, the twisting motion is eliminated, assuming that the stud was properly installed into a blind hole. All studs should only be installed hand-tight; never torque a stud into a blind hole. By eliminating the twisting motion, the full torque applied to a stud is used to create motion in one direction-stretch. This increases the clamp load. There isn't a great deal of difference in actual torque values between studs and bolts, but the difference is definitely there. Also, using a stud exerts less wear on the block threads, which improves block life. This is especially important in aluminum- or older iron-blocks where the head-bolt threads may be exposed to corrosion from engine coolant. This is also why professional engine builders use ARP-style thread-chasing taps to clean head-bolt holes in the block. These special taps merely clean the threads without removing metal, and more material is left for thread engagement between the bolt/stud and the block. Insufficient thread engagement is a common cause of pulled threads.

 

Now let's deal with the question of rod bolts and their effect on rod big-end concentricity. First, we can't emphasize enough the idea of using a rod-bolt stretch gauge as opposed to a torque wrench for tightening connecting rod bolts. We lost an engine several years ago when a rod bolt loosened up after being properly installed with a certified torque wrench with the proper lube. The rod cap came off (and destroyed the engine) because the bolt was not preloaded with sufficient stretch. A rod bolt is designed to stretch a certain amount in order to maintain sufficient tension (load) to keep the rod cap on the rod. We just performed this test again using brand-new ARP rod bolts for an engine we are building. The published ARP spec using ARP's moly paste for a small-block Chevy rod bolt is 50 lb-ft of torque. We'll admit that we have not had this particular clicker-style torque wrench calibrated, although it is only one year old. The proper rod-bolt stretch figure for this ARP 190,000-psi bolt is 0.0063 inch. When we torqued four different ARP rod bolts to 50 lb-ft using ARP's thread paste, we averaged 0.005-inch stretch for all four fasteners. We had to raise the torque to 59 lb-ft in order to achieve an average of 0.0061 inch for these same bolts. We did not measure the effect of torque load on connecting-rod concentricity because it's really not an issue. It's possible that overtightening the fastener might squeeze the big end of the rod, but that also means you're in danger of failing the bolt because it's overtightened. This is why measuring connecting-rod bolt stretch is so important-these bolts are the most highly stressed fasteners in the engine.

 

If we haven't bored you to death with all this technical stuff about bolts, it really is important information you need in order to properly assemble engines and suspension components. That connection is only as good as the fastener you use to bolt it together.