ARP bolts - Yield / Elastic limit - via stretch gauge

[tscompusa]

Has anyone on here ever done an experiment with an ARP 2000 fastener or 625+ or any fastener of ARP for that matter to see how far you could
go past the recommended stretch values until the bolt stops returning to its normal length and reaches its max yield / limits?

I'm just curious how conservative the recommended stretch values really are in ARPs' paper.

For example, an ARP 2000 SBC rod bolt is recommended between .0058-.0062

How far past these stretch values has anyone taken these and had them still return to their original size?

I've taken a ARP 2000 SBC bolt to .0066" that was rated at the above stretch and it returned back to its original
length. I'm just curious on how conservative or safe ARP has their values listed. One article states its around 75%.

I have a bunch of old 625+ bolts for testing that im not using also if anyone would be interested to see this if no one has done it on the forum.

Thanks!
Tom

[Carnut1]

I would like to know. I think fasteners that are well designed are amazing. That extra stretch before yield is like a built in safety factor for those who do not torque properly. Thanks, Charlie

[modok]

If the bolt were torqued such that it was exactly at it's elastic limit,
then the force of running it in the engine would push it past it's elastic limit.

[Sparksalot]

^^^^ This is not true.

Most people do not understand how high performance bolted connections are designed or operate under external loads, even many mechanical engineers do not understand them after being exposed to the basics in Machine Design 101 as undergraduates. I can't do a ME class here, bolted connections are not the simple devices they appear to be in mechanical or engineering terms.

They are in no way intuitive either. You can't look at it and understand it. It takes quite a bit of education to know what is really going on, more to correctly design them.

In a well designed bolted joint for a highly cyclically loaded joint such as a typical connecting rod connection with the bolt properly preloaded (the joint is a system of springs, that's how it is analyzed during design, the bolt is one of the springs) the bolt sees next to nothing in added stress beyond its preload from external loads during operation. That's why the bolts survive and it's the result of good design by folks who know what they're doing.

And that's why most people don't understand them.

As for using bolts in high performance bolted connections installed in a past yield stress level state, that's how to do it if one wants a real high performance connection. OEMs have done it for decades. That's why lots of stressed fasteners in modern engines are one use only.

[swampbuggy]

Sparksalot: Heres a question for you. If you owned say a very expensive Drag Race engine built from all new parts and you installed say CARR bolts that Carrillo markets, at what point would you replace the bolts. In other words exactly what criteria would you use to determine the time for R & R of the Con-Rod bolts. Thanks Mark .

[Kevin Johnson]

viewtopic.php?p=37181#p37178

[zums]

The stretch values they list are typically 75% of yield, if you wanted to take it to 100% yield just divide your maximum stretch by .75, in your case .0062/ .75=.0082 which isnt recommended but if you are experimenting you could start there, i got real deep into this 28yrs ago when i was stretching a friends 3/8 bolt to spec on his 327 and it took 90lbs, i thought he was going to pass out while he was watching me read torque/stretch

[Nick Campagna]

We can talk stretch, but at what length ? viz .004 on a bolt that is 1 inch from head to nut is a whole lot different that .004 on a bolt 2 inches from head to nut. We need to think x stretch / y working length. imo

[novadude]

We can talk stretch, but at what length ? viz .004 on a bolt that is 1 inch from head to nut is a whole lot different that .004 on a bolt 2 inches from head to nut. We need to think x stretch / y working length. imo

Called "strain" in engineering speak with in/in as units.

[modok]

^^^^ This is not true.

Most people do not understand how high performance bolted connections are designed or operate under external loads, even many mechanical engineers do not understand them after being exposed to the basics in Machine Design 101 as undergraduates. I can't do a ME class here, bolted connections are not the simple devices they appear to be in mechanical or engineering terms.

They are in no way intuitive either. You can't look at it and understand it. It takes quite a bit of education to know what is really going on, more to correctly design them.

In a well designed bolted joint for a highly cyclically loaded joint such as a typical connecting rod connection with the bolt properly preloaded (the joint is a system of springs, that's how it is analyzed during design, the bolt is one of the springs) the bolt sees next to nothing in added stress beyond its preload from external loads during operation. That's why the bolts survive and it's the result of good design by folks who know what they're doing.

And that's why most people don't understand them.

If a bolt is a spring and the rod is a spring, how can you pull on the bolt not cause it to become longer? That is the definition of a spring.
I believe a persons understanding of a concept IS related to their ability to describe it, so you may know quite a bit but you are not done learning.

[peejay]

It doesn't stress the bolt any more because you're not putting enough additional force on it to overcome that preload.

Think of pulling the engine from your 60s car... the front end is now sitting on its top out bumpstops. The springs are going to be under the same amount of load whether the car's weight is on the tires or not... the car's weight isn't enough to compress the spring from its preload, so the load on the spring isn't changing.

Same thing - you're stressing the bolts to be under more tension than the highest loads you can expect from the running engine. That's the preload on the spring.

[modok]

The additional force certainly is small, but it is not zero.
Speaking of car springs, you bring up creep.
It is well known some springs to lose force over time, or some combination of time, heat, and cycling. The excuse for everybody backing off their torque wrenches, standing the cranks upright, backing off the rockers ect ect.
Certainly many valve springs never do lose any pressure, and some cars never sag, but other do. It's hard to predict. But surely, the closer you go to the limits of the material, the more heat, the more cycles. the more creep will happen. Rod bolts are not immune to it.

[clshore]

We can talk stretch, but at what length ? viz .004 on a bolt that is 1 inch from head to nut is a whole lot different that .004 on a bolt 2 inches from head to nut. We need to think x stretch / y working length. imo

This^^
Without knowing the length, amount of stretch is meaningless.

The load on a joint will only create more tension in the bolt if the joint actually separates at the gap and stretches the bolt more.
The joint gap only separates if the external applied load exceeds the bolt tension.
(consider the force you must apply to an installed valve to overcome the spring and lift it off the seat, up to that point, there is no movement)

This is unlike a coil or leaf spring, where the flexing movement can eventually result in creep or sag.
The bolt only flexes and moves a very tiny bit, due to thermal cycling, and due to flexure of the entire
bolted joint due to external applied loads.

[modok]

Creep does not require any movement at all to occur. Wikipedia agrees
"In materials science, creep (sometimes called cold flow) is the tendency of a solid material to move slowly or deform permanently under the influence of mechanical stresses. It can occur as a result of long-term exposure to high levels of stress that are still below the yield strength of the material. Creep is more severe in materials that are subjected to heat for long periods, and generally increases as they near their melting point."

The rod could creep also, and they do, but the bolt is more likely to creep due to being under greater strain.

[digger]

^^^^ This is not true.

In a well designed bolted joint for a highly cyclically loaded joint such as a typical connecting rod connection with the bolt properly preloaded (the joint is a system of springs, that's how it is analyzed during design, the bolt is one of the springs) the bolt sees next to nothing in added stress beyond its preload from external loads during operation. That's why the bolts survive and it's the result of good design by folks who know what they're doing.

And that's why most people don't understand them.

sorry they dont see "next to nothing" the bolt still sees a portion of the external load proportional to the stiffness of the bolt in relation to the total stiffness of the joint. This can amount to the bolt being subjected to 30% of the applied external load depending on exact geometry and materials used

So the cyclical part is still significant and you must consider the preload plus a portion of the applied load when doing the fatigue calculation.