tightening torque question for engineers:

[gtc]

Just use an electric imact wrench. Electric motors have infinite torque.

[vash]

Originally Posted by sammyg2

BTW,
Proper lubrication makes all the difference in the world when it comes to torque. Dry threads can require up to 180% of the twisting force to obtain the same percentage of yield as one with say, moly lube or mil spec anti-seize.
On the threads, washer (if there is one) and also on the face of the fastener and the part it's going up agin.

torque doesnt mean a thing to the fastener or the parts being clamped together. tension is king.

so lubrication on fasteners really comes into play. but, in the real world as professional and shadetree mechanics, i dont think we can insure repeatablilty.

i've done enough bolt testing to make me vomit. we get bolts that have no torque spec, but a tension spec..we do the test to determine the torque value.

[Superman]

Just get it really tight. With torque values in that range, the thing to avoid is not getting it tight enough.

[Flieger]

Originally Posted by IROC

Another fun equation is that - for most practical circumstances:

T=.2Fd

Where:

T = torque
F = axial force (clamping) that is applied by the fastener during torquing
d = the nominal diameter of the fastener

The ".2" is a friction factor which is a decent approximation most of the time...

So if the fastener you are torquing to 260 ft-lbs is a 3/4" bolt, for instance, then that fastener is generating:

260 ft-lbs (12 in) = .2(F).75

F = 20800 lbs

That's a lot.

Axial force depends on thread pitch and angle of rotation. The pressure depends on force and the area of the bolt head or washer contacting the piece being clamped. So, you can either measure bolt stretch or angle of rotation. I am not sure how friction modifies the effective angle of rotation when measuring at the bolt head.

[island911]

Denis, if you put all of your 200lbs of weight at the end of a 1.3_ft bar (~1_ft, 4") you will have the 260 ft*lbs of torque.

Originally Posted by sammyg2

BTW,
Proper lubrication makes all the difference in the world when it comes to torque. Dry threads can require up to 180% of the twisting force to obtain the same percentage of yield as one with say, moly lube or mil spec anti-seize.
On the threads, washer (if there is one) and also on the face of the fastener and the part it's going up against.

I'm going to (cautiously) disagree with you on that last (my bold) part. ...splitting hairs, perhaps...

I do agree that the threads often benefit from lube, but the contact face of the fastener has the best mechanical advantage for holding friction (larger radius) --and should be used (often) as the friction-locking surface. (meaning 'dry' is good)

For the non-engineers;
. . . a boltment is just a ramp (as simple machines go) So imagine that the lube is perfect (from the same physics closet as massless ropes and frictionless pulleys). That would mean that one could put that 200-something ft-lbs of torque on the bolt-assembly, and -if fully lubed- would simply spring back.

Another example would be a door-stop wedge. The top of the stop can be slippery, and still work, but if the top, and the bottom are both slippery the whole assembly skids along the floor, or just pops the wedge back out.. ...Friction is a necessary part of nuts & bolts, and door-stop wedges.

[sammyg2]

Originally Posted by vash

torque doesnt mean a thing to the fastener or the parts being clamped together. tension is king.

so lubrication on fasteners really comes into play. but, in the real world as professional and shadetree mechanics, i dont think we can insure repeatablilty.

i've done enough bolt testing to make me vomit. we get bolts that have no torque spec, but a tension spec..we do the test to determine the torque value.

I have an excel spreadsheet at work that I got from a hydraulic torquing company for B-7 and B-16 fasteners that is cool. If anyone wants it I can e-mail it to them. It only goes down to about 3/4" so it's no good for really small stuff, but it goes up to 4" diameter IIRC.
You select the diameter and number of threads, then plug in the desired percentage of yield, then pick the lubricant. Or you can type in a custom friction coefficient if you know it, and it tells you how much torque it takes to obtain that yield. It'll get you close.

The coolest way I ever did it was at a nuke plant. We pulled every stud on the turbine (1.25 million HP) and faced off both ends in a lathe. The studs were anywhere from 3' to 6' in length.
Then when it came time to torque them, we used an ultrasound machine that measured the length of the stud down to .001" and recorded that measurement. then we used a hytorq to tighten the studs one at a time and re-measured the length, and adjusted the torque until we nailed the desired stretch. That's about as accurate as you can get.
In other places we used an electric induction heater that looked like a cattle prod and heated the hollow studs and used a protractor to determine how much to turn the nut (by hand), then when everything cooled down it was tight.
I've used micrometers to measure stretch on smaller stuff, I was amazed at how much variation there was just using a torque wrench with a multiplier.

[sammyg2]

Originally Posted by island911

Denis, if you put all of your 200lbs of weight at the end of a 1.3_ft bar (~1_ft, 4") you will have the 260 ft*lbs of torque.


I'm going to (cautiously) disagree with you on that last (my bold) part. ...splitting hairs, perhaps...

I do agree that the threads often benefit from lube, but the contact face of the fastener has the best mechanical advantage for holding friction (larger radius) --and should be used (often) as the friction-locking surface. (meaning 'dry' is good)

For the non-engineers;
. . . a boltment is just a ramp (as simple machines go) So imagine that the lube is perfect (from the same physics closet as massless ropes and frictionless pulleys). That would mean that one could put that 200-something ft-lbs of torque on the bolt-assembly, and -if fully lubed- would simply spring back.

Another example would be a door-stop wedge. The top of the stop can be slippery, and still work, but if the top, and the bottom are both slippery the whole assembly skids along the floor, or just pops the wedge back out..
...Friction is a necessary part of nuts & bolts, and door-stop wedges.

Hold on there turbo, by properly lubricate those surfaces I meant lubricate them as per design. If the design calls for no lubrication, then no lubrication is what should be used. It'll just require a great deal more torque to get the same stretch.

But if a fastener is intended to be lubricated to obtain the proper torque and you do not lubricate the faces, then you will get nowhere near the desired stretch from the fastener.

It's the stretch of the bolt or stud that does the clamping, just like a really stiff spring under tension.
If a typical fastener relies just on friction from the faces to keep it from coming loose something is probably wrong.

Disclaimer: if it's possible to engineer something no matter how screwy, someone will do it. There is always an exception. That's why some fasteners need loctite, because they are not engineered well (IMO). It's sometimes necessary to compensate for a design deficiency.

Generally, if the threads are supposed to be lubricated then the faces should be lubricated too. Otherwise you will not get it tight enough even if you use the recommended torque.
I've seen large nuts gall on the faces and even though they were pulled up to 2400 ft. lubs, they were still barely snug.
It's a beotch cutting them with an air-arc off without trashing the stud.

Like Vash, I've spent too many years playing with this stuff.

[sammyg2]

Here's a quick and simple example of how friction factor (K) affects torque and clamping:

AN6 (3/8-24) bolt tightened to 75% of bolt proof strength.
Camping force: 5,788 pounds. Required torque:

Cadmium plated and dry 28.9 lb Ft.
Moly anti-seize 19.9 Lb Ft.

[island911]

excellent! ...indeed there are differing goals (locking/holding torque, clamping force and/or desired fastener strain.)

[vash]

Originally Posted by Flieger

Axial force depends on thread pitch and angle of rotation. The pressure depends on force and the area of the bolt head or washer contacting the piece being clamped. So, you can either measure bolt stretch or angle of rotation. I am not sure how friction modifies the effective angle of rotation when measuring at the bolt head.

some of you guys are damn smart and nerdy. awesome

angle of rotation is a # you need to determine to achieve a axial tension load. two bolts, same size, different thread pitch..will need the same tension to clamp something together. (i think of bolts as springs with miniscule elongation). but with different thread pitch, you need different rotation.

tension is tension is tension.

thanks everyone for the refresher.

[GH85Carrera]

Wow, two pages and 30 posts about how to torque a bolt. With charts & graphs, and 8x10 color glossies with arrows and captions. Cool.

[id10t]

Originally Posted by GH85Carrera

Wow, two pages and 30 posts about how to torque a bolt. With charts & graphs, and 8x10 color glossies with arrows and captions. Cool.

All to tighten something that some big German dude named Hans did with a hand tool 45+ years ago...

[Flieger]

By the way, I suspect that a larger diameter fastener will have a larger torque spec than a smaller one used in the same application. The friction of the threads has a larger lever arm with a large diameter bolt or nut.

Center-lock wheels and axles and things like that have a high torque spec because they need more axial clamping force since they have to take the 1 G lateral acceleration of a ~3000 pound GT3 along with all the braking and acceleration force.



For the other nerds who want to know more about fasteners and clamping force:



If you know the thread pitch, you can finger-snug the bolt so that the bolt head is in contact with the piece being clamped. Then, if you measure the rotations of the head from that point and relative to the nut, you can work out how much clamping force you have.

Multiply the # of rotations of the bolt head (or gland nut) times the thread pitch (given as mm per thread). Then divide by 1000 for meters.

Now, the formula for this is dL=(PL)/(EA)

dL = meters = difference in length (what we just calculated with pitch and rotations)
L = meters = total initial length of fastener (between the nut and the bottom of the bolt head)
A = meters^2 = (pi/4)*bolt diameter^2

P is what we are solving for

E is the Young's Modulus. You can look that up online for your particular material.
Young Modulus of Elasticity for Metals and Alloys (English units)
Young's modulus - Wikipedia, the free encyclopedia
Aluminum is about 69 GPa and Steel is about 200 GPa

GPa is Giga-Pascal, so 1x10^9 Newton-meters/second^2 This means you need to multiply by 10^9 those numbers I gave you.

Plug these numbers into your calculator and use the formula:

(dL*E*A)/L = P

Just remember to use meters everywhere.




Dang, I just realized that the metal being clamped will compress and affect dL. Particularly if it is Aluminum. In that case you need to do:

P = (dL*E steel*A bolt*E aluminum*A washer)/(L* (E steel*A bolt) + (E aluminum*A washer))

A washer is the area of the washer you are using. pi*(r^2 outer - r^2 inner).

[RWebb]

Originally Posted by rick-l

You should stand with the middle of your foot at 1 foot.

he should stand on one foot and wear shoes with a sharp angular wdege down the middle...

[DARISC]

Originally Posted by RWebb

he should stand on one foot and wear shoes with a sharp angular wdege down the middle...

Good luck with that.
Cutting edge shoes are hard to find in a work-boots store.

[Amail]

Why not use a 130' bar with one pound on the end?

[DARISC]

Originally Posted by Amail

Why not use a 130' bar with one pound on the end?

Or a bar so long that you don't need any weight on the end?

[Amail]

Originally Posted by DARISC

Or a bar so long that you don't need any weight on the end?

Well then that would be zero foot-pounds. Don't be silly.

[TimT]

Well then that would be zero foot-pounds. Don't be silly.

Not really, consider the weight of the bar

[turbo6bar]

Originally Posted by RWebb

he should stand on one foot and wear shoes with a sharp angular wdege down the middle...

Hehe, pretty sad, but I was thinking the same thing. Denis is pretty big dude, right. How does he get those size 14 feet at the precise point on the bar?

I recommend a 2 foot bar and application of a force equal to 4 pimp slaps.