[Vicegrip]

Originally Posted by Charlie Stylianos

Upon further pondering, I don't know if this is correct. The material accepts said force, then what happens, it just disappears? Another suspicion is that the material will transfer its 'internal' torque to the lugs. So depending on the elastic properties of the material, it could just be a time difference. Softer material will take longer to deform, then transfer, and vise versa. I'll get to the bottom of this. Yes, I'm neurotic.

IMHO There is no change in torque when using a soft socket. To think so you have to assume fixed power input and speed. When the torque wrench reaches the set point it releases. The input torque then drops as the head releases and the rotation backs off from the head releasing. Click! The deformation of everything is being taken up as the power is input.
Think of it this way. As the torque is building everything is giving some regardless of a soft socket or not. If the strain builds up in a lineal fashion then the give does so as well. When the torque wrench setting limit is reached and the head releas ses the strain on everything is released a bit as well. The nut gets turned until it reaches the point that it's resistance to rotation is equal to the torque setting on the wrench and drive line give is being taken up as this is happening. If the socket were so soft that it twisted 1/2 way around before torque was met then all that means is you have to turn the wrench in a longer arc in order to reach then same torque. Your arm has the ability to add arc and doing so causes the strain to build until the release is met. (OK no more talk about arms, nuts and release)

To extrapolate on this example look at torque limiting extensions. Torque limiting extenders only work with impact guns. They do not limit torque at all by hand. Why? To answer this you have to look at how they work. They are torsion bars. When you apply a twisting motion to one end and the other is fixed they twist more and more as torque increases. The thiner the extension bar the more twist rotation for a given torque.
OK. Now how does this limit the guns ability to deliver power? To answer this you have to know how the gun works.
Over simplified answer. An air or electric motor spins a shaft that has one or two hammers attached to it. Lets look at a two hammer gun. As the hammers spin they swing out to apogee and push on anvils (bumps) moulded inside a drum that transfers the rotating force to the part that sticks out the front. Go figure the stub that sticks out the front of an impact gun and holds the socket is called an anvil. As the hammers spin they are held out by centrifugal force, push on the bumps, turn the whole mess and the socket spins the nut in. When the resistance to rotation exceeds the power that the anvil is delivering in free spin the anvil stops rotating freely. Now we are getting to the above mentioned answer. When the anvil stops the hammers do not! This is the rat a tat sound and why we call them impact guns to start with. The hammers spin around and beat on the internal anvils. Bang bang bang, each blow sends power down the shaft and into the socket and nut. Impact guns are rated in torque and BPM. Beats Per Minute. Kinda like turning a nut by hitting the side of a wrench.
Stay awake, I'm still getting there. Unlike you arm, each beat of the hammers on the anvils can turn the drive line but so far. They are limited in travel unless the anvil drum is in free spin. (They make up for the limited travel by being plentiful) The torque limiting extension will transfer twisting energy until the amount of resistance by the fastener is equal to the amount of twisting energy and degree of rotation delivered. Once this point is reached each hammer beats worth of energy is merely absorbed by the extension. Kinda like beating on a wrench with a long spring on the end. Boing, boing, after a while nothing is getting through the spring but if you grab the spring and turn it with you hand, it does bend but you just add more rotation arc to increase the torque input.

So, if the range of input motion is limited then drive line give CAN limit or change a torque value, if not it rises until the planned set point is met.

Drive line materal deformation energy is returned via springback or converted into heat. (Or as my boy Calvin would say "It was Transmogrifed")

BTW. All extensions will twist some and will reduce the power delivered by an impact gun. Using extensions on a hand held torque wrench can have a profound effect on a torque value by skewing the input load on the nut and bolt but that is a long, boring, rambling post for another day...

[VaSteve]

Uhhhh.... My head hurts....

[ras911]

Kurt
Did you have your meds today?

[William Miller]

How about the concept of lubricant on the threads or bearing surface on the lug nuts? Does the fastner get tighter if the friction is reduced?
I recall a debate on Pelican about locktight effecting torque values because of it's initial lubricating properties. I think the example was flywheel bolts.

What Kurt said is why impact wrenches don't work well on something you hold in your hand, or something mounted in rubber. Works great on rear axle nuts because of the mass of the axle etc. counteracting the hammer blows.

[matt de maria]

A lot has been written on lubrication and torquing. Using lube on dry thread torquing will usually increase torque value about 20%. Maufacturers will usually spec out whether the particular application is supposed to be clean and dry, or lightly lubed with light oil, or coated with never seize, etc.
What about tolerance. The normal torque values have a tolerance of about 10%. 8mm bolts (8g hardness) are torqued to 14 to 18 ft-lbs. That would be 16 +/- 2 ft-lb. 6mm bolts are set to 6 to 9 ft-lb. That is 7.5 ft-lb +/- 1.5 ft-lb. (+/- 20%). There is a lot of slop for most applications. Note that one of the famous exceptions is the 8mm cv bolts. They are 12g hardness and the torque value is 33 ft-lb. That is near the limit of failure for this bolt.
And all these values are based on the 2% proof limits for the elastic modulus of the steel. If you look at the values listed on the sheets that come with your torque wrench, they are based on the 1% proof load and are quite a bit higher than the common euro specs.
Age affects the torque value on threads. Smooth shiney threads will increase torque. But really worn threads will roll over on the edges (peaks) and will decrease torque.
All threaded fasteners will have a running torque value and a static torque value. That is why when torquing a fastener, a smooth continuous motion is applied to the torque wrench. When a fastener is loosened it will usually snap and then release. That is due to the static torque value being exceeded. Static torque can be 20 to 50% more than the running torque. This is easily noticed when a factory torqued fastener is loosened for the first time. (eg. lug nuts) . When there has been a lot of traffic on the threads, the static torque decreases a lot. (eg old lug nuts and studs) . Professional racing teams go through a lot of hardware in order to keep this safety factor. Nylocs get thrown out after one torque.
All in all for a street car there is a lot of safety factor built into the fasteners. However for a track car a lot of care has to be exercised. The design values for the car can be exceeded. Follow the manufacturers specs. If there is any doubt about a fastener, replace it.

[aubreym]

OK Truk, you're cut off! I think you've blown your spring center and the thingamajij is stripped where it connects to the wampuskitty. This is causing a pronounced lack of traction! :P

[Parenn911]

Originally Posted by ras911

Kurt
Did you have your meds today?

ROFL

[Vicegrip]

One thing to remember that the torque is not the real end value that the app designer is looking for, fastener tension and or clamping force is. We read the torque but it is only an indirect way to calculate the tension (stretch) applied to the fastener (bolt). Stretch is the true value as it is, calulated with the fastner build, directly responsible for the clamping force and preload of the fastener. (The stretch and clamping force calulation design can be inverted as well) Some critical fastener installations are not measured by torque but by elongation. You measure the no load length of the fastener and then tighten until it elongates to a predetermined length gain. That is a better way to yet again, indirectly, measure the real clamping force of a fastener as it bypasses many real world variables. Obviously this method of measurement is not possible in many apps nor is it required for most.

If you have a fresh, greased up bolt and a clean smoothly threaded hole the bolt will produce a higher clamping force for a given torque value when compared to a dry bolt and hole. Higher clamping force is not always desired as it pre loads (= added strain to) the fastener and threaded portion of the mounting material. "Too tight" might break the bolt or pull the threads out right away or the assembly might prematurely fail from excess tension over time. Just as torsion bars or springs bend and relax so do bolts. Unlike spring steel, most fastener alloys are not formulated to tension and relax many cycles. Some bolt alloys stretch and only partly spring back. This is one of many reasons some bolts in some apps are not to be reused. The above mentioned CV bolts that are installed close to full tension and clamping value and are a good example of this. Flywheel bolts are another.

Matt brings up another excellent point. Static torque. Grid line checking an already installed lug nut by putting a torque wrench on it and tightening until the wrench setting is met is useless to read fastener torque and get even a rough reading of clamping force. It will only find the really loose lugs or nuts which is a good thing by itself. To truly test you have to loosen and retighten which is in a way not a test if you think about it. How do you test the test?

[aubreym]

An intervention may be required.

You guys take Truk into the house and I'll guard his garage.

[Spike]

Kurt, if torquing properly is stretching to induce an acceptable clamping force - could we test by measuring the length of the torqued fastener,
assuming we knew the original length?

If so, we just need you to design a laser/optical or other type measuring
device. Or, perhaps some type of seismic device which can calculate
length based upon echo results so you don't have to be able to 'see'
the far end. ?

Steve, I have a soft socket and I shattered it after just a few uses.
Yes, I realize I may be a special case with my torqueing disability
and all, but beware.