Clamp-up versus preload 1

[koopas]

Hi all,

What's are the differences between "clamp-up" and "preload" pertaining to fastened joints?

Clamp-up: essentially how "tight" the press is in a fastened joint. Ranges from low for a high shear rivet to high for a hi-lok. Seems to me that clamp-up is good.

For a hi-lok, according to Niu: "They [hi-loks] are normally used where high clamp-up is desired for sheet pull-up or faying surface sealing requirements dictate".

What does that mean in plain English? What's "sheet pull-up"? How would clamp-up relate to faying surface requirements?


Preload: obviously not desirable. You deform the material to close a gap. This induces extra tensile (bending) stresses which 1) reduce fatigue life and 2) is conducive to stress corrosion (which occurs when a steady tensile stress is maintained in the midst of a corrosive environment).


Could someone clarify the difference between clamp-up and preload? They both seem to induce additional stresses, so how can clamp-up be beneficial?

Thanks,
Alex

 

[jetmaker]

koopas,

This is how I see the definitions:

Clamp-up: The normal force that is exerted by the faying surfaces. The higher the clamp-up force, the more load transfer that occurs through friction rather than bolt bearing and bending.

Preload: The force that is within the bolt to induce a specific clamp-up. However, a high fasteners preload can exist even with zero clamp-up. This condition could exist if there is a gap between the faying surfaces of 2 very stiff parts. Therefore as you torque the bolt, the gap never closes completely, yet high loads are inparted to the bolt. This situation results in Pull-up (see next).

Pull-Up: Occurs when a gap exists between 2 faying surfaces and the joint is tightened. The result is that the gap closes due to local bending. This increases the stress state, and results in high potential for stress corrosion fatigue.

Hope this has helped.

Trevor

 

[edbgtr]

Alex
I like Trevor's definitions above. Your question about the beneficial effects are as follows.
Simply put, the stress concentration factor(SCF) at the holes is reduced due to the clamp-up. Where clamp-up is not present, the fastener can cant in the hole, thereby increasing the local bearing stress at the edge of the hole. When the fastener is clamped up, the head and nut at either end of the fastener react the moment that was reacted in the hole. This results in a nearly uniform bearing stress in the hole, thereby reducing the SCF. This in turn improves the fatigue life of the joint. Bill McCombs deals with a bolt ultimate load condition for such a joint (D3.5a Fillers (and shims)) in his Bruhn Supplement.
Results of this effect can be seen in J Schijve's fastened joint fatigue test results in NACA TM 1395 available from the NACA report site

http://naca.larc.nasa.gov/

and is also evident from Boeing's Detail Fatigue Rating (DFR) curves for clamped-up joints.
I have yet to see a formula or design curve relating the amount of clamp-up to the reduction in the bend-bearing stress in the hole. From my attempts at deriving a general elastic method for this problem, it is a redundant solution that requires the bearing stiffness interaction between the sheet material and the fastener bending. If anyone on this forum has such a closed-form solution to this problem, I would certainly like to see it, and I'm sure so would other fatigue analysts dealing with such fastener details.
Ed.

 

[Sparweb]

Hi Alex,

Do you have a joint in tension? Clamping force plays a role in the fatigue life of a fastener in a tension joint, too. Bolts are torqued to specific torque, and the "tightness" puts a tension pre-load into the bolt. When the lugs are pulled apart, the tension in the bolt keeps the lugs together, in contact with one another, until the force pulling the lugs apart equals the pre-load in the bolt. Getting the bolt sized properly so that it can be torqued enough to exceed the tensile load expected in service increases the fatigue life of the joint over a type of joint that can "pull apart" at times. Bruhn has a good explanation of the process in C13 (I think). You should get yourself a table of bolt-torques which relates the torque to the tensile load that will be applied. I think there's a mil-standard that adresses this, and I can look it up for you later if you want.(try

http://astimage.daps.dla.mil/quicksearch/)

If all you wanted was sheet metal joint clamp ups, listen to Ed and Jetmaker. I haven't much to add, because I've never tried to derive a solution to the bearing stiffness interaction like Ed has.

I hope you have received the stuff I sent you earlier.


STF

 

[Sparweb]

Oops,

Old link. This one works:

http://assist.daps.dla.mil/quicksearch/

STF

 

[koopas]

Jetmaker:

Thanks for responding. My initial post lacked the mention that the joint would be in shear, not tension.

A few additional questions regarding your definitions:

-Clamp-up. When you say "The higher the clamp-up force, the more load transfer that occurs through friction rather than bolt bearing and bending.", what do you mean by bolt "bearing and bending"? Do you simply mean "the shear load transfer between the portion of the fastener shank that's in contact with the sheet metal AND the latter itself?"

-Preload: If I understand you correctly, in an ideal, proper joint, the preload force will be equal to the clamp-up force. However, in joints with gaps, the preload force will be much higher than the clamp-up force, thus resulting in higher stresses and susceptibility to stress corrosion fatigue. Correct?

-Finally, does "sheet pull-up" simply describe the condition whereby there's a gap in a joint? And if sheet pull-up is to be corrected without a shim, a HIGHER preload would exist?

Thanks!
Alex

 

[koopas]

Hi Sparweb,

Negative, I still haven't received the stuff. Did you email it to my work or hotmail account?

Regarding this current post, I was referring to a joint in shear, not tension. I am still processing your explanation of lugs in tension. A coworker was explaining to me that the tension preload (for lugs in tension) that's introduced by simply torquing the bolt/nut tends to "cancel out" some of the tensile force trying to pull the lugs apart. In other words, the preload force can be essentially subtracted from the externally applied tensile force, with the joint's strength being ultimately dictated by say, Ftu of the bolt. Throwing some random numbers, say the bolt is rated at Ftu=130 ksi and the preload from torquing is 10 ksi (I have no clue whether this is realistic), the actual allowable external tensile stress will be 130 + 10 or 140 ksi.

Does this somewhat agree with your explanation?

Thanks for clarifying,
Alex

 

[jetmaker]

koopas,

To further clarify my perspective on joints:

I will simplify things here, but basically it involves doing a FBD of the joint. Assume you have a standard 1/4" bolt in a Class II fit hole that is installed finger-tight (i.e. no joint clamp-up). As you load the joint in shear, the bolt transfers the applied joint load to the other part through shear within the bolt. However, the eccentricity of the bolt shear load at the faying surface of the parts relative to the applied load, introduces a moment within the bolt. This moment is reacted out by the head of the bolt. Hence you have some bending within the bolt. The tighter the bolt fit within the hole, the less the bolt bending. The same thing applies with gapped joints, except that you are increasing the eccenticity. Now, if you apply a high clamp-up force, this force acts downwards through the bolt head. So, a friction force is now added to the FBD which is proportional to the amount of downward/normal force. So, increasing the clamp-up increases the friction force. Now, if the bolt fit is loose enough, the friction force must be overcome before the bolt can act in bearing. It is important to keep in mind that this effect is beneficial primarily for fatgiue, where the loads do not overcome the friction force. Hope that helped.

Next, in a zero gap joint, the preload = clamp-up. Now, if you have a gap... the preload is higher because you have to apply a certain load to overcome the gap first. That is why it is possible to break a bolt while there is still a gap between parts.

Finally, it is unwise to not shim a gap. If the gap is large, you might want to use a structural shim reather than a free one. Unshimmed parts that are "pull-up" during assembly will result in residual stresses. The amount can be predicted using formulas from Roark. These residual stresses increase bolt stress and part stress... and can lead to premature fatigue, stress corrosion cracking, and bolt overload.

Hope all this helps. BTW I think this is way more than 2 cents worth.

Later.

 

[koopas]

Jetmaker,

Thanks again for replying. I've got a few more questions about what you wrote:

-I understand that there's a moment induced on the bolt due to the natural eccentricity that exists between the two metal sheets. Why is that moment reacted by the bolt's HEAD? I would think that the bolt's SHANK would react the moment, acting as a beam. I don't even consider the bolt head or nut as being relevant to the problem, only the shank as a straight pin taking the shear loads as well as that moment. Do you agree with me?

-Are you saying that applying a clamp-up force induces a normal force and thus a frictional force (normal force times coefficient of friction) that aids in keeping the sheets together (i.e. prevents them from moving)?

Could you please also shed some light on my previous questions in my post dated Sept. 14? Sorry to beat a dead horse here.

Thanks for your support.
Alex

 

[jetmaker]

koopas,

In response to your latest response, you are correct in that the shank does carry bending. The best example is a pin, which does not have a head. So in that case, you carry/transfer all the bending through the shank and into the part by bearing... hence a non-uniform through thickness bearing distribution. It is important to not that pins are generally used in tight fit holes. For bolts, the head reacts a portion of the bearing load if it is in a tight fit hole. If it is in a loose fit hole, it reacts a significant portion more. What the portions are... I have no specific data available.

Correct on the clamp-up force being the normal force which induces friction. If you want to prove it to yourself, connect 2 plates using a bolt in class 2 fit holes. Do not tighten the bolt. Now load the plates in light tension and then reverse the load... observe the max deflection. Now repeat with the bolt torque up. The max deflection should be zero.

With regards to your Sept 14 question... check out the Welding, Bonding, & Fastener Engineering forums. I know this issue has been addressed in several threads there. However, I will add this. The max ultimate load on a tension joint will be dictated by the bolt strength and is independant of bolt preload. Preload is really a fatigue enhancement option... also good for vibration. The theory is this... that a preloaded bolt contains a certain tensile force... similarily, the joined parts contain an equal compressive force. As you load a part to separate it from the other, the applied force is really just being used to aliviate the compressive load in the parts. Drawing a picture here helps to understand this. In reality, the slight alivation of load in the parts requires a small delta elongation in the bolt. This elongation does add some additional strain to the bolt, which in turn is stress... and hence load. So a slight increase in bolt load is observed.

Hope that helps.

Trevor

 

[koopas]

Hi Trevor,

My three Sept. 14 questions were definitional in nature (see below). I'd like to absolutely understand the terms before proceeding further. Could you please answer or comment on them separately? Thanks for your patience.

1. Clamp-up. When you say "The higher the clamp-up force, the more load transfer that occurs through friction rather than bolt bearing and bending.", what do you mean by bolt "bearing and bending"? Do you simply mean "the shear load transfer between the portion of the fastener shank that's in contact with the sheet metal AND the latter itself?"

2. Preload: If I understand you correctly, in an ideal, proper joint, the preload force will be equal to the clamp-up force. However, in joints with gaps, the preload force will be much higher than the clamp-up force, thus resulting in higher stresses and susceptibility to stress corrosion fatigue. Correct?

3. Finally, does "sheet pull-up" simply describe the condition whereby there's a gap in a joint? And if sheet pull-up is to be corrected without a shim, a HIGHER preload would exist?

Alex

 

[jetmaker]

koopas,

Let's see if these answer the question.

1) Clamp-up: Joint load is transfered from one sheet to another by first going into a fastener through bearing (projected fastener diameter times sheet thickness). The bearing load in the fastener is converted to shear and bending, which reacts the load through the bolt and out into the other sheet through bearing again. So... if you have high clamp-up (implying high friction force between sheets), the load transfer mechanism becomes... load transfers directly from one sheet to another through shear... it does not load the bolt at all.

2) Preload: Correct.

3) Sheet Pull-Up: As I define it.. yes.. it means that one part is pulled towards another, and results from a gap prior to fastening. A higher preload ONLY would exist if you were not using a torque wrench and the goal was to close the gap. If you are using a torque wrench, the preload is a function of the applied torque. Therefore the load in the bolt is the same whether or not there is a gap present if you are torquing to a specific value.

Hope this helped.

 

[koopas]

Trevor:

Ok, thanks for the response. Yes, it helped. I've got preload and sheet pull-up understood. However, when you describe "clamp-up" in (1), what do you mean when you write:

"So... if you have high clamp-up (implying high friction force between sheets), the load transfer mechanism becomes... load transfers directly from one sheet to another through shear... it does not load the bolt at all."

I thought that high joint clamp-up would only translate into the resulting frictional force having to be overcome before the bolt can act in bearing. In other words, the frictional force due to clamp-up is providing some beneficial "third-party" shear resistance that lessens the shear load acting on the bolt. Is that what you meant?

Regards,
Alex

 

[Somber]

I'll let jetmaker respond to your question, but as I see it - when your "third party" shear resistance is larger than the load (as is sometimes the case, especially in fatigue loading), then it isn't "third party" any more - it is primary... if you don't get to the point at which you overcome load capacity of the clamp up - then the fasterner doesn't need to carry any more than its preload that induces that clamp up...

I just wanted to comment on something in your first post koopas. You said: "Preload: obviously not desirable." I beg to differ. I understand that there are situations in which you need to work to elminate and/or minimize preload (wires, cabling, hydraulic lines, structures), but preload also is essential to make a lot of things work - in fastener applications as well as a huge number of other mechanical applications.

 

[Sparweb]

Alex,

This stuff all came to mind because I recently had the topic bonk me on the head when I neglected bolt torque in a preliminary analysis. I had a feeling you were talking about shear all along, but thought I should bring this up because the concepts of both shear and tension loads go hand in hand.

Some references on the topic: Mil-Hdbk-60, and page C13.18 of Bruhn.

Look through the references because I don't think I explained it very well. The bolt is always good for its ultimate strength. The pre-tenion that you get from torque-ing up the bolt prevents the two members of the fitting from SEPARATING if you pull on them.

I have given the e-mail a second try.

STF

 

[jetmaker]

Koopas,

Somber answered the question better than I could have. Excellent work Somber.

An addition to what we are talking about with Clamp-Up... it is poor engineering practice to rely upon clamp-up for load transfer in an ultimate analysis. In the real world, fasteners loosen in service, or are not torqued properly during installation. One should only use clamp-up in ultimate analysis with due dilligence, and certain precautions must be taken during installation to ensure that the fasteners are actually installed as required.

 

[koopas]

So if I understand Somber and Jetmaker correctly, taking your explanations to the extreme, in the case of a joint with extremely high clamp-up, the proportionally high frictional force (resulting from the clamp-up) between the metal sheets will act as the primary and sole shear transfer medium? In other words, the contact surface area common to the two sheets in the area adjacent to the fastener hole is "clamped-up" (tight) enough to transfer the joint's shear loads. And the only role of the bolt is to provide that clamp-up.

The now "unloaded" bolt sees no direct applied shear loads from the sheets. The only load the bolt experiences is the preload during torquing?

You can tell I am a slow learner :/
Alex

 

[CoryPad]

koopas,

Everything you said (except maybe the slow learner part) in your last post is correct. In fact, this joint style (shear-friction) is the main type used in most industries.

Regards,

Cory

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