Bolt strength in tension?

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
Wasn’t exactly sure which sub forum this belonged in so here it is. Also, I know bolts in tension are sub-optimal but I’m willing to live with it.

My question is actually about the tensile strength of the threads as they are surely not as strong as the shank. One of my too many projects requires the manufacture of around a hundred custom U-bolts of the same width but slightly different lengths. Off the top of my head the longest one is probably no more than 3/4” longer than the shortest. I’d love to just mass produce these things so I have a big box of U-bolts standing by when I need them and for that it would be great if they were all identical and cut to length on installation.

To do that, a bunch would end up with up to 3/4” of threads on the shank side of the nut (they won’t see any shear) and I’m curious if this would cause any issues of strength or if one thread or twenty are the same and just limited to whichever one is the weakest.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
17,642
Location
Memphis, TN
I think what happens is more threads equals more stretch. They won’t be as strong as the shorter ones. Strong enough is a different question. You have to answer that one.
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,294
Location
Saline Michigan
Bolt strength is a big topic. My immediate suggestion is to find and read the chapter of Shigley on Design of Screws, Fasteners, and Connections. Chapter number varies with the edition. For those not knowing, Shigley was a Univ of Michigan Mechanical Engineering Professor who wrote what is considered THE book on mechanical engineering design. Widely translated or blatantly copied. We list it in the reference materials.

Now to specifics. Bolted joints are designed to stiffness. Bolts have threads that are made to order for fatigue. Make the clamp load exceed the live load and also prevent the joint from opening and closing, and the load on the bolts becomes essentially static - cyclic part of the load is tiny, and fatigue is pretty much eliminated. Shigley discusses this stuff, as well as strength of bolted joints, threaded fasteners, etc.

Billski
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,294
Location
Saline Michigan
I think what happens is more threads equals more stretch. They won’t be as strong as the shorter ones. Strong enough is a different question. You have to answer that one.
See my response with recommendation on reading - Shigley.

General theory is that the bolt is elongating and the nut is shortening - both elastically - under tightening load and that results in almost all the load being carried in the first three threads... When the load in the first thread exceeds the shear yield strength of the thread, the thread root deforms plastically, moving the active three threads up the bolt. When that progresses to the last thread it is stripped.

Billski
 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
General theory is that the bolt is elongating and the nut is shortening - both elastically - under tightening load and that results in almost all the load being carried in the first three threads... When the load in the first thread exceeds the shear yield strength of the thread, the thread root deforms plastically, moving the active three threads up the bolt. When that progresses to the last thread it is stripped.
That’s the good stuff, thanks Bill. If I understand correctly the shank is expected to be stronger than the threads which are stronger than the threads in the nut.
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,294
Location
Saline Michigan
Wasn’t exactly sure which sub forum this belonged in so here it is. Also, I know bolts in tension are sub-optimal but I’m willing to live with it.

My question is actually about the tensile strength of the threads as they are surely not as strong as the shank.
Distortion energy theory says that pure shear yield is about 58% of pure tensile yield strength. We trust the heck out of that relationship in design of mechanical elements...

Usual recommendation is to torque bolts to achieve 90% of yield stress for several reasons discussed in Shigley. Then we usually design the amount of clamp load to keep the joined elements from slipping and then keep the joint from opening and closing under load cycles to prevent fatigue of the those threads. If instead the bolt is used as a pin in shear, and you need a significant part of the bolt's shear strength, the nut holding it had better be lightly torqued and locked by any of the common methods, but is not clamping the joint.

Tensile stress area (computed with the mean of pitch and minor thread diameters) is smaller than the major diameter, so the strength of the threaded portion is usually less than the stress for a rod at major diameter with the same material. But have you noticed that the shank of many bolts is smaller than the nominal major diameter. AN bolt bodies are more often used as pins in shear joints, and they tend to run close to nominal size, so they will tend to yield and fail in the threaded portion between the nut and body. Look up Asubt in Shigley and compare to area of the bolt body.

Strength of the threads? In many applications where nut runners and the like are used, torque to yield is utilized and these fastener sets are not reused ever. If you have to loosen them, throw them away... They are not stripped in this process, usually the threaded portion is what yields. Again see Shigley for material strength, stress areas, and torque limits.

To do that, a bunch would end up with up to 3/4” of threads on the shank side of the nut (they won’t see any shear) and I’m curious if this would cause any issues of strength or if one thread or twenty are the same and just limited to whichever one is the weakest.
Strength will be about the same as it is the threaded portion that usually yields, and you have some of that no matter what. What will change is if you get to yield stress in the threaded portion, and you have longer threaded portion in some than in others, the length change when undergoing yield will be greater with the longer threaded portion involved. Will that be enough to matter? You will have to figure that out.

The biggest concern I have for your U-bolt joints is designing so that the bolts do not fatigue. Typically, the U shaped region of your U-bolts is slightly larger bend radius than the piece being wrapped and clamped - it then has both tension and bending on it and tends to be deformed plastically - it permanently changes shape as you draw down the nuts. You must apply enough torque over enough turns to make the bolt conform to the round element, then apply enough more torque to approach tensile yield in the threaded portion. During the U-bolt settling phase, torque will be at one level and turn fairly far, then as the bolt draws up, free space around the U portion disappears, and torque with turning the nuts will rise. Unless you have large prevailing torques in these fasteners, the mechanic can easily feel the transition. If the mechanic stops here, thinking the bolt is tight, inadequate bolt tension could be left on the part, and the joint is susceptible to fatigue. This is where the torque wrench comes in. Get both nuts on the U-bolt drawn to the transition, then tighten both nuts step-wise up to whatever torque you have determined you need to adequately clamp the joint, and it will be solid.

Those bolts with 3/4" of threaded portion will undergo more elastic deformation during tightening than those with one thread under tension. If you are not torqueing to yield, you will probably not notice a difference. If you are torqueing to number then a designated number of degrees more rotation, you will usually need a larger rotation value with more thread under tension. Torque to yield processes require testing of each joint design with exemplar bolts. I doubt you are doing any of this for anything less than high production volume products.

Again, read the chapter in Shigley. It is a gold mine to understanding bolted joint design.

Billski
 
Last edited:

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
Do you mean 'not as strong' in tension as in shear? Might want to get a 2nd opinion. I was surprised by the answer too, when discussing the subject with an engineer buddy.
Fair enough, I just often see bolts in double shear treated as best practice. In any case I wanted to curb thread drift before it started, you could come on here asking how to change a light bulb and end up with nineteen replies about how much better skylights are.
 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
What are the U bolts used for?
It’s a little weird.
To clamp a round thing to a flat thing.
Weirder than that.

I haven’t started building anything yet but I’d still like to do a Bleriot XI. I’ve gathered two sets of period drawings and between those and existing examples I ought to be able to kludge one together. Anyways the fuselage is held together by brace wires anchored to U-bolts at every cluster. You can sort of make them out here, top and bottom of that vertical piece:
7B8AAF50-97DA-4184-BD12-C712525D2DE2.jpeg
 

mcrae0104

Well-Known Member
Supporting Member
Joined
Oct 27, 2009
Messages
4,240
Location
CO
It’s a little weird.

Weirder than that.

I haven’t started building anything yet but I’d still like to do a Bleriot XI. I’ve gathered two sets of period drawings and between those and existing examples I ought to be able to kludge one together. Anyways the fuselage is held together by brace wires anchored to U-bolts at every cluster. You can sort of make them out here, top and bottom of that vertical piece:
View attachment 128758
Is there a dowel or some kind of locating pin at each end of that vertical member? Just curious.
 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
Is there a dowel or some kind of locating pin at each end of that vertical member? Just curious.
Nope, just held in there by hopes and dreams I guess. Seems to have worked okay on the hundreds(?) of Bleriot XIs built from 1909 to present.
 

Tiger Tim

Well-Known Member
Joined
Apr 26, 2013
Messages
4,853
Location
Thunder Bay
Weird, but elegant in it's own way.
Very elegant in a time when wood glue was probably not very trustworthy for something like this. Two world wars later they were still building airplanes in a similar way in Belgium.

I bet the free body diagram of what’s happening in each of those clusters is wild
 
Top