tid-bits from the bench

with excerpts from ARP's resources


ARP rod bolts...the de facto standard in fastening technology. It is with no surprise that a large number of rod makers (including us) use ARP as their "go-to" fastener to keep that cap firmly attached to the arguably, the most important fasteners in any engine are the connecting rod bolts. They are the key to the entire rotating assembly and the weakest link in the connecting rod itself. A broken bolt will lead to a grenaded engine...and will end a happy day real quick.

So *why* do we need to use a stretch gauge? Why can't we just swing the bolts to a pre-determined value with a torque wrench and call it a day? Below we will attempt to explain the importance of having the most accurate clamping load possible so that you can keep racing...not keep repairing.


In addition to using a rod bolt with sufficient strength to withstand the tremendous cyclical strains placed upon it, it is absolutely mandatory that the bolts be properly tightened. The preferred method of monitoring the correct amount of tension is through the use of a STRETCH GAUGE.

Why is it the preferred method?

One word...friction. Maybe two...frictional loss.

Friction is a challenging problem because it varies so much, and is extremely difficult to attain repeatability by using commonly known lubricants. The best way to avoid the known variables associated with different lubricants is by using the stretch method. By using the Stretch Gauge you are removing the friction variable, and the clamp load can be controlled AND repeated. The type of lubricant is actually irrelevant.

To make matters even hairier, we have also found that each time a new bolt is torqued and loosened IN AN ALUMINUM ROD, the friction value will decrease with each torque cycle. This is because of "thread burnishing". You are essentially decreasing the friction by "polishing" the threads with repeated cycles of tightening and loosening the bolt (torquing).

When this cycle happens an excessive number of times, the torque value recommended for the bolt will spread farther and farther away from the stretch value recommended for the same bolt. In other words, if your torque value is a recommended 50 ft/pnds (or .006" stretch), you could very well achieve that .006" stretch with only 35 ft/pnds! We've seen this happen many times. The relationship between torque and stretch values *should* be very similar...if not the same, and NOT spread apart to such a degree as in the above example.

What happened? Those repeated torque cycles were the culprit. Hmm...

Just think...if you didn't have a Stretch Gauge to give you such information, I would predict a guaranteed rod punt in your future. This is a SERIOUS factor to consider...and the Stretch Gauge protects you from such erroneous readings created by excessive "thread burnishing".

How do you avoid that scenario in the first place? Simple. Don't burnish the rod's threads excessively. Don't torque/loosen/torque/loosen any more than necessary to achieve your target journal clearance.

We can't speak for other rod makers, but thankfully with our rods, the rod bolts come "seasoned" or "preset" to have the bolt give a very accurate torque wrench value to stretch value...provided you are using the specified weight of motor oil and your torque wrench is calibrated. Basically, torque the bolt to 90 ft/pnds, and you're already pretty darn close to the recommended stretch value...if not, spot on.

This is more work for us, yes...but we want to make sure things go as smoothly as possible for you at assembly we realize, not everyone owns a Stretch Gauge. We do what we do.

Regardless, stretch is stretch and the ONE value in a connecting rod that needs to be closely monitored...NOT torque values. Because essentially, torque is merely measuring friction...not the true clamping load that is given by the bolt.



The rod bolts must support the primary tension loads caused by each rotation (or cycle) of the crankshaft. When the crank rotates, the big end bore of the connecting rod essentially becomes oval-shaped and the rod bolts bend. As the crankshaft continues to rotate, the big end bore becomes round again.

This imposes additional stretch on the bolt. This stretch disappears when the load is removed on each revolution or cycle. Over time, this cyclic stretching and relaxing can cause the bolt to fail due to fatigue, just like a paper clip that is repeatedly bent back and forth by hand. To prevent this condition, the bolt’s clamp load must be GREATER than the load caused by the engine. In other words, the engine load that exists when the cap hangs onto the rod for dear life as it approaches TDC. On the Exhaust stroke to be exact.

So how to combat this? A properly installed rod bolt remains stretched by its clamp load and is not subjected to those cyclic loads. A quality bolt will stay stretched this way for years without failing.

In other types of bolted joints, this careful attention to tightening is not as important. It’s not only valuable when properly setting up a rod for resizing (to maintain repeatable concentricity), but also obtaining the proper clamp load when installed.

For example, flywheel bolts need only be tightened enough to prevent them from working loose. Flywheel loads are carried either by shear pins or by side loads in the bolts; they don’t cause cyclic tension loads in the bolts. Connecting rod bolts, on the other hand, support the primary tension loads caused by engine operation and must be protected from cyclic stretching. That’s why proper tightening of connecting rod bolts is SO important.


I will begin by saying the term "stretch" can be compared to, let's say, a carburetor return spring. If you stretch this spring too far, it is past its "yield" and will not return to its original length. If the spring is NOT stretched far enough, it will not be able to return the linkage properly.

All bad.

A connecting rod bolt is much like the same thing. If it isn't stretched to the recommended length, it will not exhibit the correct amount of clamping load. Too much stretch and the bolt will not be able to return to its original length.

Simply follow the written instructions ARP recommends as detailed on the spec sheet included with your rod set.

Prior to installing the rod, measure the length of the bolt in a "relaxed" (not torqued) state with your Stretch Gauge. Set your indicator to "ZERO". Then stretch (tighten) the bolt to the recommended value. When the bolt has stretched to this specified amount, the correct clamping load has been achieved. Write this value down. A sample stretch monitoring chart is included with all of our rod sets for your convenience...or below as a downloadable PDF.

When you tear the engine down for maintenance, again measure the length of each rod bolt by setting your indicator to "ZERO" before loosening the bolt – being careful to keep everything in its proper order. Loosen the bolt, and then write down the indicator's value (it will be a negative number). If you find the length of the bolt has increased over .001” from your original “IN” measurement, not only is the bolt trashed, but that is also a good indicator that the rod itself has possibly seen stresses beyond what it can safely continue to handle. Replace the rod set. This is standard protocol with most racers' engine replacing the rods is far more cost-effective than replacing damaged billet blocks, fancy cranks, etc from a rod gone AWOL.


The Stretch Gauge is far more accurate than using a torque wrench and is one of the most important tools a serious engine builder can own. If any other method is used to attain proper clamping load, you're just flying "blind". If folks were armed with this knowledge when installing their connecting rods, there would be fewer failures... fewer pointing fingers...fewer cuss words...fewer forum get the picture.

No joke.

End of story.


Scroll to Top