Axial loads produced by fine & course pitch threads with equal torques 2

[John2004]

Hi everyone,

I'm hoping someone can clear up this topic a little for me. I've seen this discussed elsewhere and I'm interested in obtaining further clarification.

It seems to me that if you have one fine pitch and one course pitch threaded bolt, both having the same diameters, grades, and finishes, and you put a nut on each bolt and torque the nuts to equal torques, the fine pitch nut should exert a greater axial load than the course pitch nut.

This seems plausible because the fine pitch nut should have a greater mechanical advantage than the course pitch nut, due to the thread pitch being less steep.

However, is this really the case ?

I have read that even identical bolts, when tightened to identical torque values, can vary in their actual tensions from plus / minus 25% to plus / minus 50%, because 85% of the torque is consumed by friction. This friction varies on each bolt. I read the information here...

http://www.surebolt.com/

(link was posted in another eng-tips thread)

If this information is reliable, then it seems almost impossible to do a real world test to accurately measure differences in axial loads of fine and course pitch bolts torqued to the same values. A variation of plus / minus 25% to plus / minus 50% on identical bolts seems huge.

The preload formula at the following link seems to take thread pitch into account in the "K" factor...

http://euler9.tripod.com/fasteners/preload.html

1. Is it generally assumed that a fine pitch thread will exert a greater axial force than a course pitch thread of the same diameter torqued to equal values ? Strictly from a theoretical standpoint, everything else being equal, this should be the case, correct ?

2. Is the mechanical advantage difference between a fine and course pitch thread of equal diameter generally very slight ? Anyone know what the difference generally is ? If it is slight, then I supposed any differences would be nullified by the fairly large friction variations given above.

3. Would any differences in axial force of fine and course pitch bolts be more prevalent on small or larger bolts ?

This seems like a case where practical application counters theory and common sense reasoning, mainly due to friction.

I would appreciate any feedback,

Thanks
John

 

[handleman]

What causes the nut to stop turning? Friction.

What are the factors involved with friction? Normal force and coefficient.

Conclusion is left as an exercise.

-handleman, CSWP (The new, easy test)

 

[CoryPad]

1. Yes, it is generally assumed that a fine pitch thread will exert a greater axial force than a coarse pitch thread of the same diameter torqued to equal values.

2. Yes, the difference is slight. Using typical values for engineered joints, the difference is < 5 % for a pitch step (e.g. M10 x 1.5 changed to M10 x 1.25). The friction variation definitely overwhelms this contribution.

3. I did calculations for M10 (pitches 1.5, 1.25, & 1) and M20 (2.5, 2, & 1.5). The differences were the same.

The way to take advantage of the larger stress area provided by a finer pitch is to use a preloading method that is not dependent on torque. Also, the loosening resistance and fatigue strength of the screw, not the static strength, are the more important reasons why fine pitch is selected.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[John2004]

Thanks for your replies guys.

>Corypad wrote:
>2. Yes, the difference is slight. Using typical values >for engineered joints, the difference is < 5 % for a pitch >step (e.g. M10 x 1.5 changed to M10 x 1.25). The friction >variation definitely overwhelms this contribution.

Similar size imperial threads seem to compare to a 1 mm pitch versus a 1.5 mm pitch, i.e., 3/8-16 threads versus 3/8-24 threads,. This would be a thread spacing of approximately .0625" versus .042" with a difference of .0205". Basically the same as comparing a 1 mm versus a 1.5 mm pitch difference.

With mechanical advantage difference being slight, and negated by friction, it seems you could torque a fine pith and course pitch bolt of the same OD, and either bolt could produce the higher or lower axial force of the two.

So many things are specified with a torque rating, torque an engine head bolt down too much or in the wrong sequence, and you can get distorted cylinders and/or heads. About every nut and bolt on a vehicle has a torque specificaiton. Not to mention all the industrial machinery and tools that have torque specifications.

My point being that the plus / minus 25% to 50% variation in axial load mentioned above (if reliable) would seem to have a significant impact on the use of a precision torque wrench. You take the time to keep your top of the line torque wrench calibrated, but then identical bolt's vary enough so that you can get a +/- 50% variation in axial load even with a perfect torque wrench.

Even if the point of torquing is more to generate friction than axial load, axial load could effect many applications, i.e., engine cylinder head bolts, too much axial tension there would seem to distort cylinders or heads.

I did not know there was that much variation in identical bolts, +/- 25% to 50% seems like quite a bit to me.

John

 

[CoryPad]

Friction variation can be large, but the auto industry does combat it with multiple methods. For example, new fastener coatings have specified friction coefficient ranges that are tighter than those of older coatings. Also, for your example of cylinder head bolts, a torque + angle specification or a yield-point tightening method is used. These largely eliminate the dependence of preload on friction.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[HeviiGuy]

Hello John,

I've read your comments with interest and note your well-founded concerns.

Indeed, there could be huge axial load variation in practical applications if you're relying on even a "calibrated" torque wrench to apply the load.

Calibration of torque wrenches is widely misunderstood: Torque wrenches are only calibrated insofar as they will indicate when a certain "torque" has been achieved. Since "torque" is merely a measurement of the resistance to input force, such calibration has nothing to do with the axial load. Well, in some cases it might, in others it won't. And that's the crux of the matter: One can never tell what the actual friction is going to be in the joint before the bolt is tightened.Yet this is what the typical torque procedure asks us to assume. I've used this analogy before: it's akin to being asked how your soup tastes before you've even put a spoonful into your mouth.

So, unless you've got absolute perfect and repeatable conditions, "calibrating" torque wrenches is totally redundant if in doing so the intent is to apply accurate and consistent axial loads.

If somebody really needs to know what your soup tastes like, they need to ask you after it's hit your tongue ;-)

Ciao,

HevïGuy

http://www.heviitech.com

 

[CoryPad]

Hev[&iuml;]Guy,

I am not exactly sure what your reason is for mentioning torque wrench calibration. Whether using an inline torque transducer (such as this one), or an internal one provided by a power tool manufacturer like Atlas Copco, the applied torque to a fastener can be measured with 1 % or better accuracy. If the friction coefficient range is known, and the part geometry is known, then calculations and/or measurements can provide a preload range.

Using your analogy, if I use sodium chloride in the soup, I know it will be salty. If I use sucrose, I know it will be sweet. I don't need the soup on my tongue to characterize it.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[HeviiGuy]

Hello Cory,

You'd be basing your assumption on what the soup tasted like the last time that you had some. What if the cook had decided to change the recipe a bit this time? You know: Maybe a litle more sugar, a pinch less salt. Perhaps a smaller dab of lubrication, or a bigger glob. Maybe a dinged thread or two. Maybe even a hole that's out of tolerance.

Your "calibrated" preconception will tell you that everything's going to be just tasty, yet your tongue's about to tell you to vomit ;-)

Continuing on with the analogy: So what if it's got more sugar or more salt than usual, right? We'll, if you're a diabetic or you have hypertension it's certainly very important. Same thing goes for bolted joints: On some applications it probably doesn't matter how accurate the preload is. However, on certain critical applications where loss of joint integrity could lead to loss of equipment, loss of production or even loss of life, you bet that you're going to want to know that you've got the proper preload.

There's just something wrong when placing blind (well, okay: nearsighted) faith into critical bolted joints by refusing to do any kind of verification whatsoever because somebody's prediction or gut-feel is deemed to be okay :-(

Ciao,

HevïGuy

http://www.heviitech.com

 

[CoryPad]

I can't argue that critical joints require the most scrutiny. Some joints need strain gage precision, some only torque to fail testing, some calculation only, some just a standard torque from a wall chart.

The cooking analogy is interesting to me. You mention the cook changing the recipe - but what if the the recipe is tightly controlled and inspected? Geometry, materials, tools are not allowed to vary beyond narrow ranges. Then, the extra end-of-line inspection, testing, etc. are not as necessary.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[HeviiGuy]

I can't argue at all with your suggestion that if everything is tightly controlled, one would get exactly what one expects. Perhaps this degree of control can be had in a lab or on a production line where everything is shiny and new. However, once the widget or pressure vessel or reactor gets out into the field and people start to open it up, all bets are off.

With the greatest apologies, I'm going to use the soup analogy again ;-) ...

Let's say the chef gives the recipe to another and tells the latter not to waver at all from his instructions. He can't be there to see what the new guy actually does. Thus, we're left to assume that everything will be fine. We'd be taking a bit of a risk, right? Now, if you had an aversion to say, horse radish or high-pressure hydrogen-service flanges letting go, wouldn't you feel safer if somebody checked the soup before it left the kitchen to make sure that it was allright? This is exactly what we'd be doing by verifying the preload of "tightened" bolts before they're put into service.


Ciao,

HevïGuy

http://www.heviitech.com

 

[Tmoose]

Not too long ago I finally noticed Clamping load is related to pitch and torque, not fastener diameter. Big bolts just can tolerate higher axial force, thus permitting higher torque

 

[plasgears]

The finer pitch favors more net tensile area in the bolt. Another factor is that high strength bolts are generally roll formed, which favors fine pitch.

 

[unclesyd]

Here is the Dose paper that figures the strip out or bolt breaking that includes the Pitch of the fastener.

http://www.hexagon.de/dose/dose-1e.pdf

There are some good papers at RS Technologies, like the paper on VDI 2230 and M-Alpha.

http://www.rstechltd.com/Technical_articles.htm

 

[tbuelna]

A fine pitch thread will have a slightly larger pitch diameter than a coarse pitch thread of the same fastener size. All other things being equal, this will result in a slightly higher friction moment with the fine thread, thus producing less effective axial force for a given input torque.

Also, a fine pitch thread will have more thread pitches engaged over a fixed axial length than a coarse pitch thread of the same nominal diameter. The accumulated tolerance of the increased number of engaged thread pitches with the fine thread will usually result in more interference, and thus a higher prevailing torque in the fine thread vs. the coarse thread.

Finally, the coarse pitch thread has a higher effective helix angle than the fine pitch thread. So the axial component will naturally be higher for a given input torque.

 

[CoryPad]

tbuelna,

Your conclusions are contrary to published documents like VDI 2230 and MIL-HDBK-60.

While a fine pitch thread does have a slightly larger pitch diameter, the higher thread friction torque is negated by the pitch torque change. This is a simple relationship:

F = 2[&middot;][&pi;][&middot;]M/p

Where

F = force
[&pi;] = 3.141 592 654
M = pitch torque
p = pitch

While more engaged threads may increase prevailing torque, this component is small compared to total applied torque. For free spinning fasteners, there is essentially zero difference between coarse pitch and fine pitch.

Coarse pitch threads have a higher lead angle than fine pitch threads. The helix angle is the complement to the lead angle, so coarse pitch threads have a smaller helix angle than fine pitch threads. As stated above, a smaller pitch develops more force for a given torque. The benefit of a coarse pitch thread is that it develops more linear displacement for a given angular displacement.

The simplification of threaded fastener assembly to a torque-tension relationship, with no mention of rotation angle, displacements, or strains is not the best practice. As shown in a previous post of mine, the differences between fine and coarse pitch threads are on the order of 5 % or less, which makes this discussion moot.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Tmoose]

Hi HevïGuy,

How do you tighten your lug bolts??

Dan T

 

[GregLocock]

Tmoose, I award you the /evil/ red star. Exactly.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Screwman]

The thing to remember is that in the commercial (read automotive) world, most joints are designed with massive overkill to allow for a lot of assembly variation. You can pisk up a car with a single M6 PC10.9 bolt and yet we use 3 M10 bolts to holt down the battery tray.
When choosing an assembly method you have to look on it a selecting an assembly strategy. You need to balance criticallity vs. cost. If I'm putting speaker grills on, I'm not going to need anything more than a quick drive to strip test and torque control. If I putting together the combustion chamber on a rocket motor I better do a much more in depth joint design and then use stretch control for assembly.
Back to original post. Other than the slight increase in tensile and fatigue with fine pitch threads we have never seen any measurable difference in loosening. We have seen a greater tendency towards cross threading with fine pitch, but every japanese car is built with fine pitch, so that is no big deal. If I am trying to get a most out of a joint I'll use fine pitch. If I'm building down and dirty assemblies I'll use coarse pitch.
One of the things that keeps Cory and I employed is the fact that there are no universal solutions for fastening. You have to look at the application and select your design based on that. Trying to apply universal solutions over the years has ended up keeping us very busy undoing the resulting problems down the road. About 15 years ago, GM was going to install every bolt on an engine line using yield control; everything from rod bolts to alternator brackets. It was a disaster and they had to go back to torque / angle for many of the joints. You have to use the correct tool for the job at hand. One of our old timers says "you can install our screws with a hammer, but they work better with a screwdriver".

 

[HeviiGuy]

Hello Dan and Greg,

It seems that you've mis-posted. The proper thread is at: thread404-229718. Nevertheless, here is my answer (which I had thought I already answered)

In the absence of something that addresses this obvious problem, I tighten my lug nuts on a wing and a prayer - with anti-seize.

*Why the animosity? Rather than burying one's Luddite head in the sand, why not accept the fundamental flaw in present practices and work towards a solution. As was mentioned by someone earlier, if it saves just one life, is this not worthwhile?

Ciao,

HevïGuy

http://www.heviitech.com