Fastener Design Manual, Part One

Rich Barrett
Summary
This manual was written for design engineers to enable them to choose
appropriate fasteners for their designs. Subject matter includes fastener
material selection, plating, lubricants, corrosion, locking methods, washers,
inserts, thread types and classes, fatigue loading, and fastener torque. A
section on design criteria covers the derivation of torque formulas, loads on a
fastener group, combining simultaneous shear and tension loads, pullout load
for tapped holes, grip length, head styles, and fastener strengths. The second
half of this manual presents general guidelines and selection criteria for rivets
and lockbolts.
Introduction
To the casual observer the selection of bolts, nuts, and rivets for a design
should be a simple task. In reality it is a difficult task, requiring careful
consideration of temperature, corrosion, vibration, fatigue, initial preload, and
many other factors. The intent of this manual is to present enough data on
bolt and rivet materials, finishes, torque, and thread lubricants to enable a
designer to make a sensible selection for a particular design. Locknuts,
washers, locking methods, inserts, rivets, and tapped holes are also covered.
General Design Information
Fastener Materials
Bolts can be made from many materials, but most bolts are made of carbon
steel, alloy steel, or stainless steel. Stainless steels include both iron- and
nickel-based chromium alloys. Titanium and aluminum bolts have limited
usage, primarily in the aerospace industry.
Carbon steel is the cheapest and most common bolt material. Most hardware
stores sell carbon steel bolts, which are usually zinc plated to resist corrosion.
The typical ultimate strength of this bolt material is 55 ksi.
An alloy steel is a high-strength carbon steel that can be heat treated up to
300 ksi. However, it is not corrosion resistant and must therefore have some
type of coating to protect it from corrosion. Aerospace alloy steel fasteners
are usually cadmium plated for corrosion protection.
Bolts of stainless steel (CRES) are available in a variety of alloys with ultimate
strength ranging from 70 to 220 ksi. The major advantage of using CRES is
that it normally requires no protective coating and has a wider service
temperature range than plain carbon or alloy steels.

A partial listing of bolt materials is given in Table 1. The following precautions

are to be noted:
1. The bolt plating material is usually the limiting factor on maximum service
temperature.
2. Carbon steel and alloy steel are unsatisfactory (become brittle) at
temperatures below -65 F.
3. Hydrogen embrittlement is a problem with most common methods of
plating, unless special procedures are used. (This subject is covered more
fully in the corrosion section.)
4. Series 400 CRES contains only 12 percent chromium and thus will corrode
in some environments.
5. The contact of dissimilar materials can create galvanic corrosion, which
can become a major problem. (Galvanic corrosion is covered later.)

Platings and Coatings

Most plating processes are electrolytic and generate hydrogen. Thus, most
plating processes require baking after plating at a temperature well below the
decomposition temperature of the plating material to prevent hydrogen
embrittlement. However, heating the plating to its decomposition temperature
can generate free hydrogen again. Thus, exceeding the safe operating
temperature of the plating can cause premature fastener failure due to
hydrogen embrittlement as well as loss of corrosion protection. (A summary of
plating and coatings is given in Table 2.)

Cadmium Plating
The most common aerospace fastener plating material is cadmium. Plating is
done by electrodeposition and is easy to accomplish. However, cadmiumplated parts must be baked at 375 F for 23 hours, within 2 hours after plating,
to prevent hydrogen embrittlement. Since cadmium melts at 600 F, its useful
service temperature limit is 450 F.
Zinc Plating

Zinc is also a common type of plating. The hot-dip method of zinc plating is
known commercially as galvanizing. Zinc can also be electrodeposited.
Because zinc plating has a dull finish, it is less pleasing in appearance than
cadmium. However, zinc is a sacrificial material. It will migrate to uncoated
areas that have had their plating scratched off, thus continuing to provide
corrosion resistance. Zinc may also be applied cold as a zinc-rich paint. Zinc
melts at 785 F but has a useful service temperature limit of 250 F. (Its
corrosion-inhibiting qualities degrade above 140 F.)
Phosphate Coatings
Steel or iron is phosphate coated by treating the material surface with a
diluted solution of phosphoric acid, usually by submerging the part in a
proprietary bath. The chemical reaction forms a mildly protective layer of
crystalline phosphate. The three principal types of phosphate coatings are
zinc, iron, and manganese. Phosphate-coated parts can be readily painted, or
they can be dipped in oil or wax to improve their corrosion resistance.
Fasteners are usually coated with either zinc or manganese phosphate.
Hydrogen embrittlement seldom is present in such parts. Phosphate coatings
start deteriorating at 225 F (for heavy zinc) to 400 F (for iron phosphate).
Nickel Plating
Nickel plating, with or without a copper strike (thin plating), is one of the oldest
methods of preventing corrosion and improving the appearance of steel and
brass. Nickel plating will tarnish unless followed by chromium plating. Nickel
plating is more expensive than cadmium or zinc plating, and also must be
baked as cadmium to prevent hydrogen embrittlement. Nickel plating is good
to an operating temperature of 1100 F, but is still not frequently used for
plating fasteners because of its cost.
Ion-Vapor-Deposited Aluminum Plating
Ion-vapor-deposited aluminum plating was developed by McDonnell-Douglas
for coating aircraft parts. It has some advantages over cadmium plating:
1. It creates no hydrogen embrittlement.
2. It insulates against galvanic corrosion of dissimilar materials.
3. The coating is acceptable up to 925 F.
4. It can also be used for coating titanium and aluminum.
5. The process forms no toxic byproducts.
It also has some disadvantages:
1. Because the process must be done in a specially designed vacuum
chamber, it is quite expensive.
2. Cadmium will outperform ion-vapor-deposited aluminum in a salt-spray
test.

Chromium Plating
Chromium plating is commonly used for automotive and appliance decorative
applications, but it is not common for fasteners. Chromium-plated fasteners
cost approximately as much as stainless steel fasteners. Good chromium
plating requires both copper and nickel plating prior to chromium plating.
Chromium plating also has hydrogen embrittlement problems. However, it is
acceptable for maximum operating temperatures of 800 to 1200 F.
Sermatel W and SermaGard
Sermatel W and SermaGard are proprietary coatings consisting of aluminum
particles in an inorganic binder with chromates added to inhibit corrosion. The
coating material is covered by AMS3126A, and the application procedure by
AMS2506. The coating is sprayed or dipped on the part and cured at 650 F.
(SPS Technologies 2 has tested Sermatel W-coated fasteners at 900 F
without degradation.) This coating process prevents both hydrogen
embrittlement and stress corrosion, since the fastener is completely coated.
Sermatel is about as effective as cadmium plating in resisting corrosion but
costs about 15 percent more than cadmium. Fasteners are not presently
available "off the shelf" with Sermatel W or SermaGard coating, but the
company will do small orders for fasteners or mechanical parts. These
coatings will take up to 15 disassemblies in a threaded area without serious
coating degradation.
Stalgard
Stalgard is a proprietary coating 3 process consisting of organic coatings,
inorganic-organic coatings, or both for corrosion resistance. According to
Stalgard test data their coatings are superior to either cadmium or zinc plating
in salt spray and weathering tests. Stalgard coatings also provide galvanic
corrosion protection. However, the maximum operating temperature of these
organic coatings is 475 F.
Diffused Nickel-Cadmium Plating
This process was developed by the aerospace industry to allow for higher
temperature cadmium coating. A 0.0004-in.-thick nickel coating is plated on
the substrate, followed by a 0.0002-in. thick cadmium plate (per AMS2416).
The part is then baked for 1 hour at 645 F. The resulting coating can
withstand 1000 F. However, the nickel plate must completely cover the part at
all times to avoid cadmium damage to the part. This process is expensive and
requires close control.
Silver Plating
Silver plating is cost prohibitive for most fastener applications. The big
exception is in the aerospace industry, where silver-plated nuts are used on
stainless steel bolts. The silver serves both as corrosion deterrent and dry
lubricant. Silver plating can be used to 1600 F, and thus it is a good high-

temperature lubricant. Since silver tarnishes from normal atmospheric

exposure, the silver-plated nuts are commonly coated with clear wax to
prevent tarnishing. Wax is a good room-temperature lubricant. Therefore, the
normal "dry torque" values of the torque tables should be reduced by 50
percent to allow for this lubricant.
Passivation and Preoxidation
Stainless steel fasteners will create galvanic corrosion or oxidation in a joint
unless they are passivated or preoxidized prior to assembly (ref. 1).
Passivation is the formation of a protective oxide coating on the steel by
treating it briefly with an acid. The oxide coating is almost inert. Preoxidization
is the formation of an oxide coating by exposing the fasteners to
approximately 1300 F temperature in an air furnace. The surface formed is
inert enough to prevent galling due to galvanic corrosion.
Black Oxide Coating
Black oxide coating, combined with an oil film, does little more than enhance
the appearance of carbon steel fasteners. The oil film is the only part of the
coating that prevents corrosion.
Thread Lubricants
There are many thread lubricants from which to choose, but only a few of the
most commonly used are covered here. The most common are oil, grease or
wax, graphite, and molybdenum disulfide. There are also several proprietary
lubricants such as Never-Seez and Synergistic Coatings. Some threadlocking compounds such as Loctite can also be used as lubricants for a
bolted assembly, particularly the compounds that allow the bolts to be
removed. A summary of thread lubricants is given in Table 3.

Oil and Grease

Although oil and grease are the most common types of thread lubricants, they
are limited to an operating temperature not much greater than 250 F. (Above
this temperature the oil or grease will melt or boil off.) In addition, oil cannot
be used in a vacuum environment. However, oil and grease are good for both
lubrication and corrosion prevention as long as these precautions are
observed.
Graphite
"Dry" graphite is really not dry. It is fine carbon powder that needs moisture
(usually oil or water) to become a lubricant. Therefore, its maximum operating
temperature is limited to the boiling point of the oil or water. It also cannot be
used in a vacuum environment without losing its moisture. Because dry
graphite is an abrasive, its use is detrimental to the bolted joint if the
preceding limitations are exceeded.
Molybdenum Disulfide
Molybdenum disulfide is one of the most popular dry lubricants. It can be
used in a vacuum environment but turns to molybdenum trisulfide at

approximately 750 F. Molybdenum trisulfide is an abrasive rather than a

lubricant.
Synergistic Coatings
These proprietary coatings are a type of fluorocarbon injected and baked into
a porous metal-matrix coating to provide corrosion prevention and lubrication.
However, the maximum operating temperature given in their sales literature is
500 F. Synergistic Coatings will also operate in a vacuum environment.
Neverseez
This proprietary compound is a petroleum-base lubricant and anticorrodant
that is satisfactory as a one-time lubricant up to 2200 F, according to the
manufacturer. The oil boils off, but the compound leaves nongalling oxides of
nickel, copper, and zinc between the threads. This allows the fastener to be
removed, but a new application is required each time the fastener is installed.
NASA Lewis personnel tested this compound and found it to be satisfactory.
Silver Goop
Silver Goop is a proprietary compound containing 20 to 30 percent silver.
Silver Goop can be used to 1500 F on materials other than aluminum or
magnesium. It is extremely expensive because of its silver content.
Thread-Locking Compounds
Some of the removable thread-locking compounds (such as Loctite) also
serve as antigalling and lubricating substances. However, they are epoxies,
which have a maximum operating temperature of approximately 275 F.
Corrosion
Galvanic Corrosion
Galvanic corrosion is set up when two dissimilar metals are in the presence of
an electrolyte, such as moisture. A galvanic cell is created and the most active
(anode) of the two materials is eroded and deposited on the least active
(cathode). Note that the farther apart two materials are in the following list, the
greater the galvanic action between them.
According to reference 2 the galvanic ranking of some common engineering
materials is as follows:

Galvanic Ranking of Engineering Materials

Rank
1
2

Material
Magnesium
(most active)

Rank
11
12

Material
Cast iron
Ni-Resist

Rank
21
22

Material
Yellow
brass

Rank
31
32

Material
Monel
Type 304

3
4
5
6
7
8
9
10

Magnesium
alloys
Zinc
Aluminum
5056
Aluminum
5052
Aluminum
1100
Cadmium
Aluminum
2024
Aluminum
7075
Mild steel

13
14
15
16
17
18
19
20

Type 410
stainless
(active)
Type 304
stainless
(active)
Type 316
stainless
(active)
Lead
Tin
Muntz
Metal
Nickel
(active)
Inconel
(active)

23
24
25
26
27
28
29
30

Admiralty
brass
Aluminum
brass
Red brass
Copper
Silicon
bronze
70-30
Coppernickel
Nickel
(passive)
Inconel
(passive)
Titanium

33
34
35
36

stainless
(passive)
Type 316
stainless
(passive)
Silver
Graphite
Gold (least
active)

Note the difference between active and passive 304 and 316 stainless steels.
The difference here is that passivation of stainless steels is done either by
oxidizing in an air furnace or treating the surface with an acid to cause an
oxide to form. This oxide surface is quite inert in both cases and deters
galvanic activity.
Because the anode is eroded in a galvanic cell, it should be the larger mass
in the cell. Therefore, it is poor design practice to use carbon steel fasteners
in a stainless steel or copper assembly. Stainless steel fasteners can be used
in carbon steel assemblies, since the carbon steel mass is the anode.
Magnesium is frequently used in lightweight designs because of its high
strength to weight ratio. However, it must be totally insulated from fasteners
by an inert coating such as zinc chromate primer to prevent extreme galvanic
corrosion. Cadmium- or zinc-plated fasteners are closest to magnesium in the
galvanic series and would be the most compatible if the insulation coating
were damaged.
Stress Corrosion
Stress corrosion occurs when a tensile-stressed part is placed in a corrosive
environment. An otherwise ductile part will fail at a stress much lower than its
yield strength because of surface imperfections (usually pits or cracks)
created by the corrosive environment. In general, the higher the heat-treating
temperature of the material (and the lower the ductility), the more susceptible
it is to stress corrosion cracking.
Material manufacturers have been forced to develop alloys that are less
sensitive to stress corrosion. Of the stainless steels, A286 is the best fastener
material for aerospace usage. It is not susceptible to stress corrosion but
usually is produced only up to 160-ksi strength (220-ksi A286 fasteners are
available on special order). The higher strength stainless steel fasteners (180

to 220 ksi) are usually made of 17-7PH or 17-4PH, which are stress corrosion
susceptible. Fasteners made of superalloys such as Inconel 718 or MP35N
are available if cost and schedule are not restricted.
An alternative is to use a high-strength carbon steel (such as H-11 tool steel
with an ultimate tensile strength of 300 ksi) and provide corrosion protection.
However, it is preferable to use more fasteners of the ordinary variety and
strength, if possible, than to use a few high-strength fasteners. High-strength
fasteners (greater than 180 ksi) bring on problems such as brittleness, critical
flaws, forged heads, cold rolling of threads, and the necessity for stringent
quality control procedures. Quality control procedures such as x-ray, dye
penetrant, magnetic particle, thread radius, and head radius inspections are
commonly used for high-strength fasteners.
Hydrogen Embrittlement
Hydrogen embrittlement occurs whenever there is free hydrogen in close
association with the metal. Since most plating processes are the electrolytic
bath type, free hydrogen is present. There are three types of hydrogen-metal
problems: * Hydrogen chemical reaction: Hydrogen reacts with the carbon in
steel to form methane gas, which can lead to crack development and strength
reduction. Hydrogen can also react with alloying elements such as titanium,
niobium, or tantalum to form hydrides. Because the hydrides are not as strong
as the parent alloy, they reduce the overall strength of the part. * Internal
hydrogen embrittlement: Hydrogen can remain in solution interstitially
(between lattices in the grain structure) and can cause delayed failures after
proof testing. There is no external indication that the hydrogen is present. *
Hydrogen environment embrittlement: This problem is only present in a highpressure hydrogen environment such as a hydrogen storage tank. Unless a
fastener was under stress inside such a pressure vessel, this condition would
not be present.
Most plating specifications now state that a plated carbon steel fastener "shall
be baked for not less than 23 hours at 375 + 25 F within 2 hours after plating
to provide hydrogen embrittlement relief" (per MIL-N-25027D). In the past the
plating specifications required baking at 375 + 25 F for only 3 hours within 4
hours after plating. This treatment was found to be inadequate, and most
plating specifications were revised in 1981-82 to reflect the longer baking
time. Hydrogen embrittlement problems also increase as the fastener strength
increases.
Cadmium Embrittlement
Although hydrogen embrittlement failure of materials is well documented (ref.
3), the effects of cadmium embrittlement are not. In general, hydrogen
embrittlement failure of cadmium-plated parts can start as low as 325 F, but
cadmium embrittlement can start around 400 F. Since both elements are
normally present in elevated-temperature failure of cadmium-plated parts, the
combined effect of the two can be disastrous. However, the individual effect of
each is indeterminate.

Locking Methods
Tapped Holes

Figure 1. Spiralock Thread.

In a tapped hole the locking technique is normally on the fastener. One

notable exception is the Spiralock 7 tap shown in Figure 1. The Spiralock
thread form has a 30' wedge ram at its root. Under clamp load the crests of
the male thread are wedged tightly against the ramp. This makes lateral
movement, which causes loosening under vibration, nearly impossible.
Independent tests by some of the aerospace companies have indicated that
this type of thread is satisfactory for moderate resistance to vibration. The bolt
can have standard thread, since the tapped hole does all the locking.
[Spiralock is distributed by Detroit Tap & Tool Company, Detroit, Michigan,
through license from H.D. Holmes.]
Locknuts

Figure 2. Split-beam locknut.

There are various types of locking elements, with the common principle being
to bind (or wedge) the nut thread to the bolt threads. Some of the more
common locknuts are covered here.
Split beam. The split-beam locknut (Figure 2) has slots in the top, and the
thread diameter is undersized in the slotted portion. The nut spins freely until
the bolt threads get to the slotted area. The split "beam" segments are
deflected outward by the bolt, and a friction load results from binding of the
mating threads.
Deformed thread. The deformed-thread locknut (Figure 3) is a common
locknut, particularly in the aerospace industry. Its advantages are as follows:

Figure 3. Deformed-thread locknut.

1. The nut can be formed in one operation.

2. The temperature range is limited only by the parent metal, its plating, or
both.
3. The nut can be reused approximately 10 times before it has to be
discarded for loss of locking capability.

Figure 4. Nylok pellet locknut.

Nylok pellet. The Nylok pellet (of nylon) is usually installed in the nut threads
as shown in Figure 4. A pellet or patch projects from the threads. When
mating threads engage, compression creates a counter-force that results in
locking contact. The main drawback of this pellet is that its maximum
operating temperature is approximately 250 F. The nylon pellet will also be
damaged quickly by reassembly.

Figure 5. Locking collar.

Locking collar and seal. A fiber or nylon washer is mounted in the top of the
nut as shown in Figure 5. The collar has an interference fit such that it binds
on the bolt threads. It also provides some sealing action from gas and
moisture leakage. Once again the limiting feature of this nut is the
approximate 250 F temperature limit of the locking collar.
A cost-saving method sometimes used instead of a collar or nylon pellet is to
bond a nylon patch on the threads of either the nut or the bolt to get some
locking action. This method is also used on short thread lengths, where a
drilled hole for a locking pellet could cause severe stress concentration.

Figure 6. Castellated nut.

Castellated nut. The castellated nut normally has six slots as shown in Figure
6 (a). The bolt has a single hole through its threaded end. The nut is torqued
to its desired torque value. It is then rotated forward or backward (depending
on the user's preference) to the nearest slot that aligns with the drilled hole in
the bolt. A cotter pin is then installed to lock the nut in place as shown in
Figure 6 (b). This nut works extremely well for low-torque applications such as
holding a wheel bearing in place.

Figure 7. Jam nut.

Jam nuts. These nuts are normally "jammed" together as shown in Figure 7,
although the "experts" cannot agree on which nut should be on the bottom.
However, this type of assembly is too unpredictable to be reliable. If the inner
nut is torqued more tightly than the outer nut, the inner nut will yield before
the outer nut can pick up its full load. On the other hand, if the outer nut is
tightened more than the inner nut, the inner nut unloads. Then the outer nut
will yield before the inner nut can pick up its full load. It would be rare to get
the correct amount of torque on each nut. A locknut is a much more practical
choice than a regular nut and a jam nut. However, a jam nut can be used on a
turnbuckle, where it does not carry any of the tension load.

Figure 8. Durlock nut.

Serrated-face nut (or bolthead). The serrated face of this nut (shown in Figure
8) digs into the bearing surface during final tightening. This means that it
cannot be used with a washer or on surfaces where scratches or corrosion
could be a problem.
According to SPS Technologies, their serrated-face bolts (Durlock 180)

require I 10 percent of tightening torque to loosen them. Their tests on these

bolts have shown them to have excellent vibration resistance.

Figure 9. Lockwiring.

Lockwiring. Although lockwiring is a laborious method of preventing bolt or nut

rotation, it is still used in critical applications, particularly in the aerospace
field. The nuts usually have drilled corners, and the bolts either have throughholes in the head or drilled corners to thread the lockwire through. A typical
bolthead lockwiring assembly is shown in Figure 9 (a), and a typical nut
lockwiring assembly is shown in Figure 9 (b). Direct interfering thread. A direct
interfering thread has an oversized root diameter that gives a slight
interference fit between the mating threads. It is commonly used on threaded
studs for semi-permanent installations, rather than on bolts and nuts, since
the interference fit does damage the threads.

Figure 10. Tapered thread.

Tapered thread. The tapered thread is a variation of the direct interfering

thread, but the difference is that the minor diameter is tapered to interfere on
the last three or four threads of a nut or bolt as shown in Figure 10,
Nutplates. A nutplate (Figure 11) is normally used as a blind nut. They can be
fixed or floating. In addition, they can have most of the locking and sealing
features of a regular nut. Nutplates are usually used on materials too thin to
tap. Primarily the aerospace companies use them, since their installation is
expensive. At least three drilled holes and two rivets are required for each
nutplate installation.

Figure 11. Nutplate.

Locking Adhesives
Many manufacturers make locking adhesives (or epoxies) for locking threads.
Most major manufacturers make several grades of locking adhesive, so that
the frequency of disassembly can be matched to the locking capability of the
adhesive. For example, Loctite 242 is for removable fasteners, and Loctite
2719 is for tamper-proof fasteners. Other manufacturers such as Bostik, ND
Industries, Nylock, 3M, and Permaloc make similar products.
Most of these adhesives work in one of two ways. They are either a single
mixture that hardens when it becomes a thin layer in the absence of air or an
epoxy in two layers that does not harden until it is mixed and compressed
between the mating threads. Note that the two-layer adhesives are usually
put on the fastener as a "ribbon" or ring by the manufacturer. These ribbons
or rings do have some shelf life, as long as they are not inadvertently mixed
or damaged.
These adhesives are usually effective as thread sealers as well. However,
none of them will take high temperatures. The best adhesives will function at
450 F; the worst only 200 F.

Fastener Design Manual, Part Two

Rich Barrett
Washers
Belleville Washers

Figure 12. Types of Belleville washers. (a) smooth; (b) serrated

Belleville washers (Figure 12) are conical washers used more for maintaining
a uniform tension load on a bolt than for locking. If they are not completely
flattened out, they serve as a spring in the bolt joint. However, unless they
have serrated surfaces, they have no significant locking capability. These
surfaces, on the other hand, will damage the mating surfaces under them.
These washers can be stacked in combinations as shown in Figure 13 to
either increase the total spring length (Figure 13(a) and (c)) or increase the
spring constant (Figure 13(b)).

Figure 13. Combinations of Belleville washers. (a) series; (b) parallel; (c) parallel-series

Lockwashers
The typical helical spring washer shown in Figure 14 is made of slightly
trapezoidal wire formed into a helix of one coil so that the free height is
approximately twice the thickness of the washer cross section. Lockwashers
are usually made of hardened carbon steel, but they are also available in
aluminum, silicon, bronze, phosphor-bronze alloy, stainless steel, and KMonel.
The lockwasher serves as a spring while the bolt is being tightened. However,
the washer is normally flat when the bolt is fully torqued. At this time it is
equivalent to a solid flat washer, and its locking ability is nonexistent. In
summary, a lockwasher of this type is useless for locking.

Figure 14. Helical spring washers.

Tooth (or Star) Lockwashers

Tooth lockwashers (Figure 15) are used with screws and nuts for some spring
action but mostly for locking action. The teeth are formed in a twisted
configuration with sharp edges. One edge bites into the bolthead (or nut)
while the other edge bites into the mating surface.

Figure 15. Tooth lockwashers. (a) flat; (b) countersunk

Although this washer does provide some locking action, it damages the
mating surfaces. These scratches can cause crack formation in highly
stressed fasteners, in mating parts, or both, as well as decreased corrosion
resistance.

Figure 16. Self-aligning washer.

Self-Aligning Washers
A self-aligning washer is used with a mating nut that has conical faces as
shown in Figure 16. Because there is both a weight penalty and a severe cost
penalty for using this nut, it should be used only as a last resort. Maintaining
parallel mating surfaces within acceptable limits (two feet, per SAE Handbook
(ref. 4)) is normally the better alternative.

Figure 17. Wire thread insert installation.

Inserts
An insert is a special type of device that is threaded on inside diameter and
locked with threads or protrusions on outside diameter in a drilled, molded, or
tapped hole. It is used to provide a strong, wear-resistant tapped hole in a soft
material (e.g. plastics, nonferrous metals), as well as to repair stripped
threads in a tapped hole.
The aerospace industry uses inserts in tapped holes in soft materials in order
to take advantage of weight savings provided by small high-strength
fasteners. The bigger external thread of the insert (nominally 1/8 in. bigger in
diameter than the internal thread) gives, for example, a 10-32 bolt in an
equivalent 5/16-18 nut.
In general, there are two types of inserts: those that are threaded externally,
and those that are locked by some method other than threads (knurls,
serrated surfaces, grooves, or interference fit). Within the threaded inserts
there are three types: the wire thread, the self-tapping, and the solid bushing.

Figure 18. Wire thread insert types. (a) free running; (b) locking

Threaded Inserts
Wire thread. The wire thread type of insert (Heli-coil 10) (a) is a precision coil
of diamond-shaped CRES wire that forms both external and internal threads
as shown in Figure 17. The coil is made slightly oversize so that it will have
an interference fit in the tapped hole. In addition, this insert is available with a
deformed coil (Figure 18) for additional locking. The tang is broken off at the
notch after installation.
The wire thread insert is the most popular type for repair of a tapped hole with
stripped threads, since it requires the least amount of hole enlargement.
However, the solid bushing insert is preferred if space permits.

Figure 19. Self-tapping inserts. (a) Slotted; (b) Nylok

Self-tapping. Most of the self-tapping inserts are the solid bushing type made
with a tapered external thread similar to a self-tapping screw (Figure 19).
There are several different locking combinations, such as the Nylok plug
(Figure 19(b)) or the thread-forming Speedsert deformed thread (Figure 20).
An additional advantage of the thread-forming insert is that it generates no
cutting chips, since it does not cut the threads. It can only be used, however,
in softer materials. [Rexnord Specialty Fasteners Division, Torrance,
California.]

Figure 20. Speedsert.

Solid bushing. Solid bushing inserts have conventional threads both internally
and externally. A popular type is the Keensert [Rexnord] shown in Figure 21.
The locking keys are driven in after the insert is in place. Another
manufacturer uses a two-prong ring for locking. These inserts are also
available with distorted external thread or Nylok plugs for locking.

Figure 21. Keensert.

Nonthreaded Inserts
Plastic expandable. The most familiar of the nonthreaded inserts is the plastic
expandable type shown in figure 22. This insert has barbs on the outside and
longitudinal slits that allow it to expand outward as the threaded fastener is
installed, pushing the barbs into the wall of the drilled hole. (See ref. 5.)
Molded in place. This type of insert (Figure 23) is knurled or seffated to resist
both pullout and rotation. It is commonly used with ceramics, rubber, and
plastics, since it can develop higher resistance to both pullout and rotation in
these materials than self-tapping or conventionally threaded inserts. (See ref.
5.)

Figure 22. Plastic expandable insert.

Ultrasonic. Ultrasonic inserts (Figure 24) have grooves in various directions to

give them locking strength. They are installed in a prepared hole by pushing
them in while they are being ultrasonically vibrated. The ultrasonic vibration
melts the wall of the hole locally so that the insert grooves are "welded" in
place. Since the area melted is small, these inserts do not have the holding
power of those that are molded in place. Ultrasonic inserts are limited to use
in thermoplastics. (See ref. 5.)

Figure 23. Molded in-place insert.

Threads
Types of Threads Since complete information on most threads can be found
in the ANSI standards (ref. 6), the SAE Handbook (ref. 4), and the National
Institute of Standards and Technology Handbook H-28 (ref. 7) no thread
standards will be included here. The goal here is to explain the common
thread types, along with their advantages and disadvantages. The common
thread types are unified national coarse (UNC), unified national fine (UNF),
unified national extra fine (UNEF), UNJC, UNJF, UNR, UNK, and constantpitch threads.

Figure 24. Ultrasonic inserts.

Unified national coarse. UNC is the most commonly used thread on generalpurpose fasteners. Coarse threads are deeper than fine threads and are
easier to assemble without cross threading. The manufacturing tolerances
can be larger than for finer threads, allowing for higher plating tolerances.
UNC threads are normally easier to remove when corroded, owing to their
sloppy fit. However, a UNC fastener can be procured with a class 3 (tighter) fit
if needed (classes to be covered later).
Unified national fine. UNF thread has a larger minor diameter than UNC
thread, which gives UNF fasteners slightly higher load-carrying and better
torque-locking capabilities than UNC fasteners of the same identical material
and outside diameter. The fine threads have tighter manufacturing tolerances
than UNC threads, and the smaller lead angle allows for finer tension
adjustment. UNF threads are the most widely used threads in the aerospace
industry.
Unified national extra fine. UNEF is a still finer type of thread than UNF and is
common to the aerospace field. This thread is particularly advantageous for

tapped holes in hard materials and for thin threaded walls, as well as for
tapped holes in thin materials.
UNJC and UNJF threads. "J" threads are made in both external and internal
forms. The external thread has a much larger root radius than the
corresponding UNC, UNR, UNK, or UNF threads. This radius is mandatory
and its inspection is required, whereas no root radius is required on UNC,
UNF, or UNEF threads. Since the larger root radius increases the minor
diameter, a UNJF or UNJC fastener has a larger net tensile area than a
corresponding UNF or UNC fastener. This root radius also gives a smaller
stress concentration factor in the threaded section. Therefore, high-strength
(180 ksi or more) bolts usually have "J" threads.
UNR threads. The UNR external thread is a rolled UN thread in all respects
except that the root radius must be rounded. However, the root radius and the
minor diameter are not checked or toleranced. There is no internal UNR
thread.
UNK threads. The UNK external thread is similar to UNR, except that the root
radius and the minor diameter are toleranced and inspected. There is no
internal UNK thread.
According to a survey of manufacturers conducted by the Industrial Fasteners
Institute, nearly all manufacturers of externally threaded fasteners make UNR
rolled threads rather than plain UN. The only exception is for ground or cut
threads.
Constant-pitch threads. These threads offer a selection of pitches that can be
matched with various diameters to fit a particular design. This is a common
practice for bolts of 1-in. diameter and above, with pitches of 8, 12, or 16
threads per inch being the most common.
A graphical and tabular explanation Of UN, UNR, UNK, and UNJ threads is
given on page M-6 of reference 8. A copy (Figure 25) is enclosed here for
reference. According to the Industrial Fasteners Institute, "[the following] is
not a screw thread standard, should not be used as a working sheet, and
should only refer the reader to the proper ANSI Standards document wherein
the full thread details on working data are contained."

Figure 25. Explanation of UN, UNR, UNK, and UNJ threads. Reprinted with
permission of Industrial Fasteners Institute.
Classes of Threads
Thread classes are distinguished from each other by the amounts of
tolerance and allowance. The designations run from IA to 3A and IB to 3B for
external and internal threads, respectively. A class I is a looser fitting, generalpurpose thread; a class 3 is the aerospace standard thread, and has a tighter
tolerance. (The individual tolerances and sizes for the various classes are
given in the SAE Handbook (ref 4).)
Forming of Threads
Threads may be cut, hot rolled, or cold rolled. The most common

manufacturing method is to cold form both the head and the threads for bolts
up to one inch in diameter. For bolts of larger diameter and high-strength
smaller bolts, the heads are hot forged. The threads are still cold rolled until
the bolt size prohibits the material displacement necessary to form the
threads (up to a constant pitch of eight threads per inch). Threads are cut only
at assembly with taps and dies or by lathe cutting.
Cold rolling has the additional advantage of increasing the strength of the bolt
threads through the high compressive surface stresses, similar to the effects
of shot peening. This process makes the threads more resistant to fatigue
cracking.
Fatigue-Resistant Bolts
If a bolt is cycled in tension, it will normally break near the end of the threaded
portion because this is the area of maximum stress concentration. In order to
lessen the stress concentration factor, the bolt shank can be machined down
to the root diameter of the threads. Then it will survive tensile cyclic loading
much longer than a standard bolt with the shank diameter equal to the thread
outside diameter.
Fatigue (Cyclic) Loading of Bolts

Figure 26. Fatigue loading of bolts. (a) bolted flanges with external load; (b)
free body with no external load; (c) free body with external load
The bolted joint in Figure 26 (from ref. 9) is preloaded with an initial load F i,
which equals the clamping load FC before the external load FE is applied. The
equation (from ref. 11) for this assembly is
FB = Fi + FE KB / (KB + KC)
where FB is the total bolt load. In this equation KB is the spring constant of the
bolt and KC is the spring constant of the clamped faces. To see the effects of
the relative spring constants, let R = KC / KB. Then (from ref. 10)
FB = Fi + FE / (1 + R)
In a normal clamped joint KC is much larger than KB (R = 5.0 for steel bolt and

flanges), so that the bolt load does not increase much as the initial external
load is applied. (Note that the bolt load does not increase significantly until F E
exceeds Fi.)
A series of triangular diagrams (Figure 27, from ref. 11) can be used to clarify
the effect of externally applied loads. Triangle OAB is identical in all four
diagrams. The slope of OA represents the bolt stiffness; the slope of AB
represents the joint stiffness (joint is stiffer than bolt by ratio OC/CB.) In
Figure 27(a) the externally applied load F E (a) does not load the bolt to its
yield point. In Figure 27 (b) the bolt is loaded by F E (b) to its yield point, with
the corresponding decrease in clamping load to F CL. In Figure 27 (c) the
external load (c) has caused the bolt to take a permanent elongation such
that the clamping force will be less than F i when FE (c) is removed. In Figure
27 (d) the joint has completely separated on its way to bolt failure. Note that
the flatter the slope of OA (or the larger the ratio OC/OB becomes), the
smaller the effect Fi has on bolt load. Therefore, using more small-diameter
fasteners rather than a few large-diameter fasteners will give a more fatigueresistant joint. Referring to Figure 27 (a), note that the cyclic (alternating) load
is that portion above Fi. This is the alternating load (stress) to be used on a
stress-versus-load-cycles diagram of the bolt material to predict the fatigue
life of the bolts. Note that an initial preload F i near the bolt yields minimizes
cyclic loading.

Figure 27. Bolt external loading.

Thermal Cyclic Loading of Bolts
If the bolt and joint are of different materials, an operating temperature higher
or lower than the installation temperature can cause problems. Differential
contraction can cause the joint to unload (or separate); differential expansion
can cause overloading of the fasteners. In these cases it is common practice
to use conical washers (see washer section of this manual) to give additional
adjustments in fastener and joint loading.

Table 4. Coefficients of static and sliding friction.

On coefficients of static and sliding friction (from

reference 12)
1. Campbell, Trans. ASME, 1939.

a. Oleic acid

2. Clarke, Lincoln, and Sterrett, Proc.

API, 1935.

b. Atlantic spindle oil (light

mineral)

3. Beare and Bowden, Phil. Trans. Roy.

Soc., 1985.

c. Castor oil

4. Dokos, Trans. ASME, 1946.

d. Lard oil
e. Atlantic spindle oil plus

5. Boyd and Robertson, Trans. ASME,

1945.

2% oleic acid

6. Sachs, zeit f. angew. Math. And Mech.,

1924.

f. Medium mineral oil

7. Honda and Yamada, Jour. I of M,

1925.

g. Medium mineral oil plus

0.5% oleic acid

8. Tomlinson, Phil. Mag., 1929.

h. Stearic acid

9. Morin, Acad. Roy. des Sciences, 1838.

i. Grease (zinc oxide

base)

10. Claypoole, Trans. ASME, 1943.

11. Tabor, Jour. Applied Phys., 1945.
12. Eyssen, General Discussion on
Lubrication, ASME, 1937.
13. Brazier and Holland-Bowyer, General
Discussion on Lubrication, ASME,
1937.
14. Burwell, Jour. SAE, 1942.
15. Stanton, "Friction", Longmans.
16. Ernst and Merchant, Conference on
Friction and Surface Finish, MIT,
1940.
17. Gongwer, Conference on Friction and
Surface Finish, MIT, 1940.
18. Hardy and Bircumshaw, Proc. Roy.
Soc., 1925.

j. Graphite
k. Turbine oil plus 1%
graphite
l. Turbine oil plus 1%
stearic acid
m.Turbine oil (medium
mineral)
n. Olive oil
o.
p. Palmitic acid
q. Ricinoleic acid
r. Dry soap
s. Lard
t. Water

19. Hardy and Hardy, Phil. Mag., 1919.

u. Rape oil

20. Bowden and Young, Proc. Roy. Soc.,

1951.

v. 3-in-1 Oil

21. Hardy and Doubleday, Proc. Roy.

Soc., 1923.

x. Triolein

22. Bowden and Tabor, "The Friction and

Lubrication of Solids," Oxford.
23. Shooter, Research, 4, 1951.

w. Octyl alcohol

y. 1% lauric acid in
paraffin oil

Fastener Torque
Determining the proper torque for a fastener is the biggest problem in
fastener installation. Some of the many variables causing problems are
1. The coefficient of friction between mating threads
2. The coefficient of friction between the bolthead (or nut) and its mating
surface
3. The effect of bolt coatings and lubricants on the friction coefficients
4. The percentage of bolt tensile strength to be used for preload
5. Once agreement is reached on item 4, how to accurately determine this
value
6. Relative spring rates of the structure and the bolts
7. Interaction formulas to be used for combining simultaneous shear and
tension loads on a bolt (should friction loads due to bolt clamping action be
included in the interaction calculations?)
8. Whether "running torque" for a locking device should be added to the
normal torque