Bolts
Bolts
1. Introduction
Bolted joints are among the most common type of fastening connections. Similarly to the screws, a
bolt joint requires a series of elements that forms what can be called a bolt group. The bolt group
consist in the following items:
Bolt: Similar to a screw, the bolts are design following some ASTM and ISO standards. Bolts can be
made with different type of heads; which includes hexagonal head, round head, or square head.
Nut: Bolt connections require a Nut, which also has its own ASTM and ISO standards. The nut has to
be chose of the same material that the Bolt. Special types of Nuts, such as Locking Nuts can be used
to reduce the self-opening risk during frame vibration, or fluctuating loads.
Always select a nut which proof tensile stress is equal or higher the proof tensile stress of the bolt.
Washer: Not strictly necessary for all the bolt connections; washers are recommended to be
installed under the bolt head as well as under the nut, in order to improve the effectiveness of a pre-
load connections, but also to protect the clamped material.
For pre-loaded bolted connections, specific standard washers, made of hardened steel are required.
Bolts are similar to screws, but they strictly need to be tighten using a nut; meanwhile screws are
normally threaded to a tapped hole.
Screws and bolts have normally quite different designs, as it can be seen comparing a mushroom
head bolt with a socket head cap screw. However, for the specific group of Hexagonal Bolts and
Hexagonal head cap screws, they are much alike.
-Structural bolts require to be used paired with a nut, while hexagonal bolts do not require a nut.
-The proper method to apply a bolt to a connection is fixing the head, and turn the nut; while in a
screw (even if you are using a nut), you should tight the head of the screw.
-Both may use different standards, which could be for dimensions/tolerances or for the mechanical
properties. Although not always, the bolt standards, tend to produce thicker heads than in the
hexagonal screw (for example comparing the ISO 4014 & the ISO 8765, or the ASTM vs the SAE).
To make the separation easier, I personally suggest using ASTM and ISO 4014 type of standards for
bolts, and use other type of configurations (rather than hexagonal head) when using screws. If
Hexagonal head screws are the best option to use, use them in accordance with Standards SAE J249
or ISO 8765 as shown in the screw Unit.
They are the most common type of Bolts used. They can be used in all type of bolted connections,
with good rates of torsional resistance, as well as tensile/shear strength. Although they can be
tighten thought the head (using a wrench), the design method is using the wrench to fix the head,
and tight the nut around the bolt.
Extensively used in structural and building construction (such as in bridges), this bolt has a round
head (flat from the bottom) to clamp the parts together, and a square neck to prevent the bolt of
turning during the tightening operation. Therefore, this type of bolt doesn’t require a wrench to fix
the head.
It is similar to the mushroom, but using a conical flat head, which flush with the surface of the
clamped part.
Similar to the hexagonal bolt, large flat sides make them easy to grip with a wrench and keep them
from rotating in square holes.
ISO 8765: Hexagon head bolt with fine thread. Product grades A and B
ISO 7411: Hexagon bolts for high strength structural bolting (threads according to ISO 888). Product
grades C.
Apart from the ISO, there are other metric standards such as the ASME in the USA:
ASME B18.2.6M: Hexagon heavy head bolt as well as the Nuts (after 2012).
Imperial
ASME B18.2.6 is used for structural bolts for ASTM A325 and ASTM A490.
The following ISO standards are set to determine the material and properties of the bolts and nuts:
ISO 898-1: Steel and Alloy Steel Bolts, screws, and male fastener material properties
ISO 898-2: Steel and Alloy Steel Nuts and female fasteners material properties
ISO 3506-1: Stainless Steel Bolts, screws and male fasteners material properties
ISO 3506-2: Stainless Steel nuts and female fasteners material properties
There are other metric standards equivalents to the ISO, such as:
Imperial
ASTM A563: Low and medium carbon steel, may be quenched and tempered
Although there are different types of protective coating that can be applied to bolts and nuts, the
most commonly found are the following: [6] [7]
ISO
ASTM [10]
ASTM 2329/F2329M: Standard specification for Zinc coating and Hot-dip requirements for
application to carbon steel and alloy steel fasteners
If the environment or use requires high corrosion resistance, other type of bolts are required,
normally Stainless Steel (titanium or aluminium if the weight is an important factor). When using
steel and Stainless steel bolts, be careful when they touch aluminium or galvanized steel parts. If the
conditions for galvanic corrosion are met, the more noble metal will become the cathode and the
more active the anode; this condition increases the anode’s rate of corrosion. [8] [9]
In these conditions, any element present to break the galvanic corrosion conditions will be sufficient
to stop the anode’s corrosion. This “element” could be using insulated washers, add other type of
coating (such as paint or polymer coatings)…
2. Stress Section and Bolt Strength
2.1 Stress Section
Bolts and nuts are design so they always fail first in the threaded area [3], representing the weakest
point of the joint.
2.1.1 Metric
d: Nominal diameter of the bolt. Also is the shank (unthreaded zone) of the bolt.
H: height of fundamental thread Triangle of the thread in accordance with ISO 68-1.
P: Pitch
d2: is the basic pitch diameter of external thread in accordance with ISO 724 [5].
See ISO 724 to obtain a table showing all the previous parameters for all the nominal diameters.
( )
Meanwhile in the unthreaded zone (shank), the section can be calculated as:
2.1.2 Imperial
( )
According to the Shigley’s mechanical engineering design book [3], “The proof load is the maximum
load (force) that a bolt can withstand without acquiring a permanent set. The proof strength is the
quotient of the proof load and the tensile-stress area. The proof strength thus corresponds roughly
to the proportional limit and corresponds to 0.0001 in permanent set in the fastener (first
measurable deviation from elastic behaviour)”.
To understand its difference with respect yield strength, Dane McKinnon, from portlandbolt [4],
defines the proof strength as “Because of the difficulty of getting reliable test results for yield when
testing full size fasteners, the proof load test was developed as an alternative. In a proof load test, a
headed bolt is placed in a testing machine with a nut on the threaded end, and a wedge under the
head. The wedge varies from 4-10 degrees depending on the size and configuration of fastener, and
serves to evaluate the ductility of the bolt. In a machine specimen test, you test ductility by
measuring elongation and reduction of area, but those are impossible during a full size test, so head
deflection is used in its place. Proof load testing is typically performed at 90-93% of the expected
minimum yield strength and is a simple pass/fail test. The bolt length is measured, and after being
subjected to the published proof load value for 10 seconds, if it has not elongated more than
0.0005″, it is deemed to have passed”.
So, basically, the proof strength is an alternative elastic limit value for the bolts that typically ends up
being around 85-95% of the yield strength. [5]
Since the proof strength (Sp) is smaller and more accurate than the yield strength, we will be using it
when we do our analysis.
1.7 Bibliography
[1] https://www.fastenal.com/en/76/metric-system-and-specifications
[2] https://www.fastenal.com/content/documents/FastenalTechnicalReferenceGuide.pdf
[3] http://www.boltcouncil.org/files/2ndEditionGuide.pdf
[8] http://www.ssina.com/corrosion/galvanic.html
[9] http://www.aluminiumdesign.net/design-support/aluminium-corrosion-resistance/
[10] https://www.astm.org/Standards/fastener-standards.html
2. Non-preloaded Bolts
Bolt joints could be pre-loaded to a specific torque or non-preloaded
In this Section we are discussing about the non-preloaded bolts. Non-preloaded bolts are all the
different type of bolt joints that aren’t pre-loaded to a specific and known torque; even if the joint is
torque by a wrench. Since the torque value is not known, the bolt is considered non-preloaded [1]
[2]
2.1 Purpose
Since pre-loaded bolts require high trained personnel to install them, non-preloaded bolts made the
structure connections cheaper. However, there are some conditions when non-preloaded bolts can’t
be used due to its limitation to handle fatigue stress.
-Non-critical connections. Defining critical as if the fastener fail, it would cause a catastrophically
event or high economical event.
-Low fatigue vibration connections, in which the use of locking mechanisms (such as locking nuts or
washers are recommended).
-Pinned connections where it’s required that the bolt is able to slip around the hole.
NOTE: If during the design, you are required to follow any standard, in which they establish when to
use pre-loaded or non-preloaded bolts. Replace the information here commented for the standards’
one. Never use this information to substitute any engineering standard.
2.2 Static Strength
2.2.1 Tension
Notification
F: Tensile Load per bolt. Actual Force that is applied axially to the bolt (N)
As: Bolt section in the threaded area (As,nom in the ISO 898-1) (mm2)
When some force is acting axially on a bolt, it produces a stress on the threaded zone. That stress
can be obtained as:
That safety factor accounts for all the irregularities in order to calculate the force component and
value, as well as other unknowns. However, since the bolt material and properties may be very well
tested based on the standards ISO or ASTM, the Safety factor used could be different than the one
used to calculate manufactured parts.
The allowable tensile stress shall be compared with the proof stress of the Standards (i.e. ISO 898-1).
General Notification
Fs: Shear Force per bolt. Actual Force that is applied perpendicular to the length of the bolt (N)
As: Bolt section in the threaded area (As,nom in the ISO 898-1) (mm2)
In the case that the joint is loaded purely in shear, the bolt is going to act similarly to a pin. As it can
be seen in the previous image, in a non-preloaded bolt, the “clamped” parts slip and end up
pressuring the bolt. The stress produced on the parts is called “bearing stress” (that’s why this type
of connections are called bearing type connections), and on the bolt shear stress. The bolted joint
can be loaded in single shear, double (like in the image), or in multiple shear; making the connection
“stronger”.
Using the safety factor that we consider, we can obtain the allowable shear stress:
Using the Von mises failure theory, a 0,577 factor should be applied to the proof tensile stress to
obtain the shear proof stress [6]. However, except for lightweight designs, a factor of 0,5 will be
used, for safety purposes.
t: plate thickness
Due to the pressure between the shank/thread of the bolt and the hole surface, the “clamped”
material can have an excessive stress/deformation, enough to yield the material, or break it.
Bearing stress is therefore, the stress that is being acted over the parts that are joined by the
bolt/nut.
There are mainly three failures possible due to bolt acting in shear [11]:
Bearing deformation
Notation
f: Area Factor
The Bearing deformation is produced by the surface of the bolt that is in contact to the hole.
Depending of the size of the hole clearance (loose, normal, or close), the amount of surface of the
bolt in contact with it varies.
The area in contact “Ab” can be found in the material mechanics books as:
Where “d” is the diameter of the bolt, and “t” is the thickness of the plate. It is importance to clarify
that unless the hole diameter is very close to the bolts nominal diameter, the area in contact is
smaller than the “Ab” wrote previously.
Once the hole clearance grows respect to “d”, the pressure angle is reduced. Therefore, the bearing
stress is determined in the following equations depending on the type of clearance hole:
Close Fit:
Once the Stress area is calculated, we proceed to find out the bearing stress:
Again, we multiply the stress by the Safety factor to obtain the Allowable bearing stress:
The Allowable bearing stress produced by the bolt is compared with the allowable yield bearing
stress of the plate. The reason to use the yield stress is because even if the plate doesn’t fail, it will
be deformed once the yield point of the material is crossed.
A failure in a bearing stress commonly elongates the bolt hole, producing a slot as is represented in
the picture (c).
Tear-out/shear-out
The second most typical bearing failure is the tear-out of the plate between the hole and the edge
perpendicular to the shear force vector. Although similar to the bearing stress commented before,
the tear-out is represented by two high stress lines, that if fail, would tear apart the material
between the hole and the edge. It is also common that the tear-out failure is preceded by a bearing
plastic deformation.
Notation
e1: Distance from bolt centreline to the end of the plate that is perpendicular to the shear force
component.
To calculate if the joint is strong enough, we need to find out what is the area affected. That area is
determined by the distance between the hole and the edge “e1”, and the thickness of the plate “t”.
Normally, to avoid this type of failures, design engineers use a thumb rule of leaving a distance “e1”
equal of 1,5 times the diameter of the hole “d”, and a distance between holes equal to 3 times the
hole diameter. This value can be also found in some construction standards such as the AISC.
Net-tension
This is the third most habitual bearing failure for bolt connections. Although it’s a failure less
common than the other two previously commented, it can occur if the distance of the hole and the
edge parallel to the force vector is small.
Notation
e2: Distance from the hole to the end of the plate that is parallel to the shear force component.
As shown in the picture (a), net-tension failure tends to create cracks at both sides of the hole that
grow enough to separate the plate in two. The crack is always initiated at the hole, due to the higher
concentration stress factor; propagating to both edges.
The stress area of this kind of failure is determined as the section at both sides of the hole, as shown
in the picture below (in green).
Area that is determined by the width and the tensile shown in the picture:
Where the “2” is because the green are is localized at both sides of the hole .
The stress is therefore:
The Ks is the concentration stress factor. This factor can be calculated following the empirical
graphics in any material mechanics book [14]. Depending on the ratio between the hole diameter
and the plate/bar width, the tensional stress at the hole could be 2 or even 3 times higher than the
nominal stress.
Multiplying the stress by the safety factor, we obtain the allowable stress:
Differently than the previous failures of “bearing stresses”, the Net-tension stress is compared with
the tensile stress of the material. It still a thread due to the high concentrations that a hole next to
an edge can produce.
Bearing Stregth
Once the bearing stress is calculated (for any of the failure options besides Net-tension), it is
necessary to compare it to the allowable yield or ultimate (depending on the failure case) bearing
stress of the material. The relation to the allowable stress and the yield/ultimate is the safety factor.
There are two options to calculate the yield/ultimate bearing stress of a material:
-Material specialist: Sometimes, you can obtain the bearing yield strength or bearing ultimate
strength in some material websites or books. Also, you can contact your material supplier who
should probably have more information regarding to their materials.
As an example, in matweb, for the Aluminium 6061-O, they show the bearing strength
-Bearing strength related to the tensile: Since it is very difficult or hard to find the bearing strength
of a material, it’s common that engineers use half of the ultimate and yield strength (or a third if we
are designing using plastics) [12]. This value, although not accurate or empirical based, tends to be
conservative, increasing the safety factor of the design. The same thing could be done to calculate
hypothetical endurance limit strength for the steel or aluminium.
The two most extended criteria are maximum principal stresses and Von-mises. Although there is
high complexity in order to choose between one over the other, typically, the Von mises failure
criteria is used in “ductile” materials (such as steel and aluminium), and maximum principals stresses
to “brittle” materials (such as cast iron). The reality is that the real stress could be found in between
the both criteria, especially when using very hard but more brittle steel bolts (such as the class 12.9).
In case of not being sure which failure theory use, calculate the stress using both and use the highest
one. [12]
Von-Mises
Using Von mises failure criteria:
Maximum principal
Using maximum principal stresses [6]:
√
2.3 Fatigue
Although, as commented at the beginning of the chapter, non-preloaded bolt connections might be
working in a dynamic or fatigue application. We recommended to use preloaded connections
whenever you are using bolts in a fatigue or dynamic condition, but specifically for critical
connections. At the same time, there could be multiple applications where the forces applied
fluctuate in non-critical connections.
Being non-critical, the usage of specific fatigue equations may not be strictly necessary, but could
reduce the cost of the hardware, and extend the maintenance programmed schedule.
2.3.1 Tension
A fatigue connection can be separated in two groups due to the number of cycles expected during its
operative life:
This case is a little easier to solve, using some fatigue “factors” to reduce the ultimate strength of
the bolt [16]. In this case, we will be using 0.75 as followed:
Medium cycle fatigue (more than 103 and less than 106)
Medium fatigue analyses are not as easy to obtain as the long life ones. The main way to solve it is to
obtain a S-N (Wohler) curve for the exact material used.
If the S-N curve is not available, follow item 3.5.1 using C=1 and Fl = 0; calculating the Goodman,
Gerber, or ASME-Elliptic, and compare the safety factor with the yielding factor. You can also obtain
the number of cycles following the Unit 6 in the Shigley’s Mechanical and design book.
Extremely Long life analyses are difficult to obtain, since they might change drastically depending on
the conditions, environment, surface, load application, and material.
For steel and iron alloys, it is common to use the endurance limit strength instead of the proof, yield,
or ultimate strength. Commonly, the endurance strength is lower than the yield or proof for a bolt.
For the rest of materials (copper, aluminium, and titanium), the endurance limit is usually less clear
than in the steel material. Empirical values can be used, as well as obtaining the endurance strength
like in the previous chapter.
In any of the long life cycle designs, it is very important to check for concentration stresses, using
concentration factors; since the cracks usually start in them.
2.3.2 Shear
Similar to the tension fatigue case, bolts in shear can be also used in fatigue applications (non-
critical). The equations and options are the same as in tension, except that we use the Shear proof,
yield, and ultimate strength.
2.3.3 Combination
Similarly to the static equations, a von-mises or principle stresses formulas could be applied to
calculate the combination of shear and tension on a bolt.
It is important to mention than an impact bolt resistance could be calculated in conjunction with a
fatigue analysis; if the load case is repeated during N number of cycles.
Hypothetically, a sudden reaction can be considered as double of a static reaction [17]. This factor
could be considered quite conservative, and can be modified based on testing and FE analyses.
2.4.2 Impacts
Impacts are much harder to calculate than static loads. Using the equations available, you can
calculate the “impact Force” [16] [17], and use it to obtain the bolt strength required. Theoretical
dynamic calculations are generally very conservative, being reduced by materials/designs that can
absorb a lot of energy without yielding.
Part of the explanation is that dynamic calculations don’t typically consider the energy lost during
the process. For that reason, a possible “reduction” factor could be applied in conjunction with tests
or FE analyses.
2.5 Bibliography
[2] http://www.boltcouncil.org/files/2ndEditionGuide.pdf
[4] http://www.portlandbolt.com/technical/faqs/proof-vs-yield-strength/
[5] http://www.smartbolts.com/fundamentals/
[6] http://www.roymech.co.uk/Useful_Tables/Matter/shear_tensile.htm
[7] http://fgg-web.fgg.uni-lj.si/~/pmoze/esdep/master/wg11/l0310.htm#SEC_7_2
[9] https://arc.aiaa.org/doi/abs/10.2514/8.11197?journalCode=jans
[10] http://www.calqlata.com/productpages/00064-help.html
[11] http://pubs.sciepub.com/ajme/3/4/5/figure/5
[12] http://www.eng-tips.com/viewthread.cfm?qid=155036
[13] http://www.bgstructuralengineering.com/BGSCM13/BGSCM003/BGSCM00306.htm
[14] http://www.fracturemechanics.org/hole.html
[15] https://en.wikipedia.org/wiki/Bearing_pressure
[16] http://www.roymech.co.uk/Useful_Tables/Fatigue/Fatigue.html
The main advantage is when the connection is under fatigue type of stress, with fluctuating forces
acting on the connection; this type of bolt method reduces the fatigue failure compare with the
regular method, both in tension and shear.
However, it is important to know that a pre-loaded method of connecting the bolt has some
limitations, and once leaving the operational range of force resistance, it acts like a regular bolt.
Other limitations come with the knowledge of the mechanics who install the bolt group, if they over-
tight the bolt to the nut, they can produce permanent deformation to the threads, which may
produce a failure on the connection under an external force; if under-tight, the connection reduces
the operational range, and will act as a regular bolt connection (which could be catastrophically if
the engineer calculated the fatigue of the group considering that the bolts are correctly preloaded).
Comparing with the conventional method, the use of preloaded bolts is carried out in the following
scenarios:
-Critical connections, especially where the forces applied to the bolt group are fluctuating.
-Any scenario typified in the standard used during the design of any equipment/construction (such
as Eurocode or AISC for steel construction).
Sp: Proof load if the bolt, around 85% of the yield tensile force
K: Friction coefficient
In order to make practical the use of the friction effect, a high tensile strength is applied on the bolt,
clamping the materials together. The tensile force applied torqueing the bolt/nut changes depending
the type of pre-loaded needed.
Friction coefficient is an important factor that affects the torque needed during the pre-load.
Normally the material and surface texture is going to determine the K coefficient [2]. However, it is
very common for steel bolts to have some sort of protecting coating, which will be in contact with
the clamped surface or the washer.
In the following chart, there are some of the typical steel coatings found in the bolts [1]:
Bolt coating K
Zinc-plated 0.20
Cadmium-plated 0.18
There are other conditions or elements that will affect the friction coefficient, such as [1]:
Bolt condition K
Following Shigley’s Mechanical Design handbook page 423 [1], The torque equation can be defines
as:
That equation permits calculating the torque needed to set the bolt to the tension required based
on the permanent/non-permanent coefficients.
l = ld + lt
Fl = preload force
To find out the members stiffness, we use a half-apex angle of 30° as explained in Shigley’s design
book [1]:
Where:
A and B are constants related to the material; see Shigley’s design book [1] page 416.
3.1.5.3. Fraction of external load carried by bolt
Based on the stiffness of the bolt / member, we can calculate the fraction of external load carried by
the bolt as:
Knowing all the previous variables, we can determine the proof stress needed in order to decide the
bolt size and material:
This stress needs to be multiplied by the safety factor (if higher than 1), to obtain the Allowable
proof stress:
Replacing all the previous equations of pre-load torque and stress, we obtain the final proof stress
equation:
Once the Proof strength is calculated, it can be compared with the standard proof stress of the bolt
(for example ISO 898-1).
Notation
n: Number of bolts
Preloaded bolt joints are also called HSFG (High Strength Friction Grip). They are design so the bolt
doesn’t slip due a shear force applied, preventing the bolt to work in bearing; instead, the shear
force is transmitted between members due to the friction force.
The main advantages [3] are the fact that HSFG connections can handle fluctuating forces much
better than the non-preloaded ones. As a disadvantage, the difficulty, training and cost of preload
correctly the bolts, force them to be used only when they are necessary.
Based on the chapter 4 of the ASIC bolt council guide [4], the slip force resistance of a preloaded
connection is given by:
Based on these equations, if the external shear force applied through the members is smaller than
the Pslip, the bolt connection will be able to transmit the force without slipping. On the other hand,
if the external force exceeds the Pslip value, the connection will slip. The amount of slippage will be
determined by the difference between the external load and the Pslip. As extreme case, the bolt will
contact the hole wall, becoming a conventional bearing type of connection.
Based on different testing, large bolted joints have an average slip coefficient of 0.33, with a
deviation of 0.07. For small joints, the average was around 0.34 with a deviation of 0.07.
The material plays a very important role in the slip coefficient, depending on the surface treatment,
steel composition, and surface condition. Based on [4], page 82, there is a summary chart of the slip
coefficients obtained in different tests.
The ESDEP course shows how to calculate the slip resistance if a external tensile load is apllied to a
preloaded bolt joint (Lecture WG11) [3].
3.5 Fatigue
P = Maximum force
3.5.1 Tension
Based on the Shigley’s Mechanical Engineering Book, we are going to analysed the fatigue action
over a preloaded bolt joint based on the regular fatigue methods commented in the same book,
chapter 6.
Since the use of roller threads is the predominant method of threading forming, we are going to use
the “rolled thread” concentration stress factor of 2,1-2,3. In the shigley’s page 433 (8th edition), we
can also see a chart showing some examples for the ISO different grades endurance Strength.
Using the Fmax and Fmin as the manixum and minimum fluctuating forces during the operative life
of the joint, we can obtain the amplitude stress, as well as the average stress:
The next problem is to find the strength components Sa and Sm of the fatigue failure line. These
depend on the failure criteria:
Goodman:
Gerber:
( )
Or the ASME-Elliptic:
( ) ( )
As Sm = Sa + σl,
Gerber:
Asme-Elliptic:
Shear
A preload type of connection doesn’t experience fatigue cycles in the same way as the non-
preloaded one. As long as the preload prevents the bolt of slipping, the bolt won’t be loaded as a
bearing type of connection. While the bolt isn’t slipping, fatigue will not appreciably affect the life of
the bolt joint.
The connection, however, may start loosening as a consequence of the fluctuation of forces. If the
nut starts loosening its preload tension, the connection starts losing the grip, and therefore the bolt
joint could be considered a non-preloaded type of joint.
A hardened steel washer is placed under the nut and the bolt head. This leads to a more uniform and
not too high friction between the rotated part and the underlying plates. When the bolt axis is not
normal to the contact surface (difference more than 3°), and appropriate taper washer shall be
applied. [3]
Based on the document in boltscience [5], flange bolts should replace regular bolts with washer
underneath.
Bibliography
[2] http://www.roymech.co.uk/Useful_Tables/Tribology/co_of_frict.htm
[3] http://fgg-web.fgg.uni-lj.si/~/pmoze/esdep/master/wg11/l0320.htm
[4] http://www.boltcouncil.org/files/2ndEditionGuide.pdf
[5] http://www.boltscience.com/pages/why-use-washers.pdf