Mathematical Model for Tapping Simulation to

Predict Radial Pitch Diameter Difference of threads

Jie Ren (  botao12330@[Link] )
Taiyuan University of Science and Technology
Xianguo Yan
Taiyuan University of Science and Technology

Original Research

Keywords:

Posted Date: February 12th, 2021

DOI: [Link]

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
Noname manuscript No.
(will be inserted by the editor)

Mathematical model for tapping simulation to predict Radial

Pitch Diameter Difference of threads
Jie Ren1 · Xianguo Yan1

Received: date / Accepted: date

Abstract As the most important component of parts, 1 Introduction

thread has a great influence on the mechanical proper-
ties and service performance of parts. In order to ensure In modern industry, threaded connections are often used
the quality of the thread, the thread quality inspec- when parts and pipe connections need to be assem-
tion standard involves 11 main thread characteristic- bled in a non-destructive manner. The performance of
s, of which the surface roughness of the thread is the threaded connections is critical to many fields such as
most studied, and the research on Radial Pitch Diam- petroleum, aerospace, shipbuilding, high-speed rail, nu-
eter Difference (RPDD) is still blank. In this paper, a clear energy, and automobile manufacturing[1–3].Accor-
quasi-static model of the tapping process is develope- ding to statistics, threaded connections generally ac-
d based on the roundness error mechanism of the hole, count for 60% of the total mechanical components in
which includes cutting force and cutting damping force. each machinery and equipment[4].Therefore, it is of great
Due to the regenerative nature of cutting, the force on significance to improve thread processing quality[5].
each cutting edge depends on both the tool’s current Poor thread quality will affect the mechanical prop-
position and previous position. According to the eigen- erties and service performance of the components, such
values and eigenvectors of the discrete state-transition as tensile strength, torsion strength, vibration resis-
matrix, RPDD is finally determined, and the influence tance, connection reliability, pipeline tightness, etc.[2, 3,
of the chamfer length and the spindle speed on RPDD is 6].The factor that has the greatest influence on the me-
simulated by this model. The results demonstrate that chanical properties and service performance of threaded
the chamfer length and spindle speed will affect RPDD, components is the geometry and dimensional accura-
and the RPDD is the smallest when the chamfer length cy of the thread[2].Dong[2] and Leon[7] studied the in-
is 2 threads and the spindle speed is 1400 rev/min. The fluence of thread dimensional conformance on the self-
development of this model not only provides a cheap loosening resistance and static strength of the threaded
and effective method for the study of RPDD, but also connection which will be greatly reduced due to the
lays a foundation for further experimental research. unqualified outer diameter, middle diameter or inner
diameter of bolts and nuts. Nassar et al.[8] showed that
unqualified thread root radius will adversely affect the
fatigue performance of pre-tightened threaded fasten-
Keywords Mathematical model · Tapping · Radial
ers. In order to ensure the quality of thread, the thread
Pitch Diameter Difference · Simulation
quality inspection standard involves 11 main thread
characteristics, and RPDD is one of them. RPDD is
defined as the maximum difference among the pitch di-
ameters in all radial directions within a lead. When
Jie Ren ·
Xianguo Yan( )
a thread with an excessive RPDD is matched with a
E-mail: botao12330@[Link] qualified thread, the sharply changing contact area be-
1
School of Mechanical Engineering, Taiyuan University of tween the internal and external threads may cause un-
Science and Technology even load distribution on the thread surface, and severe
2 Jie Ren1 , Xianguo Yan1

stress concentration occurs at the position with a s- hole is largely caused by the vibration of the drill. The
maller pitch diameter, which accelerates wear and tear, tap usually shows transverse vibration, torsional vibra-
shorten the service life, and severely cause ”looseness” tion and axial vibration during tapping, but there is no
or even ”slip buckle”[3]. coupling relationship between the transverse vibration
Compared with external threads, it is more diffi- and axial/torsional vibration of the tap[23]. Because the
cult to ensure the quality of internal threads, especial- torsional stiffness and axial stiffness are much greater
ly small diameter internal threads. The machining of than the bending stiffness, the torsion and axial dis-
internal threads is very complicated, which is usual- placement of a cantilever tool (such as a tap) with a
ly the last process of workpiece manufacturing. Any large length-to-diameter ratio is much smaller than the
machining failure or reduced accuracy can not achieve transverse displacement under a given force. Therefore,
the perfect assembly of components without gaps, and torsional vibration and axial vibration are ignored, that
even lead to huge economic losses[9, 10]. Tapping is a is, the pitch error generated in tapping is ignored. The
process widely used in the manufacturing of internal pitch diameter of the thread is only related to the nom-
threads. Although in the past few years, other process- inal diameter of the thread, which means that the R-
ing methods have been used to manufacture internal PDD is caused by the roundness error of the root cy-
threads with great success, such as thread milling and cle. Bayly et al.[24] established a quasi-static reaming
turning[11–16], tapping is almost the only method to model to explain the vibration of the tool during the
manufacture small diameter internal threads[17]. How- cutting process and the resulting roundness errors in
ever, there are relatively few literatures on thread qual- reamed holes, and process parameters such as the di-
ity in tapping. Fromentin et al.[18] showed that the use rection and the angular frequency of rotation of the tool
of suitable oil-based lubricants in tapping can improve will affect the dimensional of the roundness error. Deng
the mechanical properties and surface integrity of inter- et al.[25] believed that the dynamic excitation force will
nal threads. Piska et al.[19] compared taps with PVD cause the deflection and the vibration of the tool, lead-
and TiN composite coatings and uncoated taps to pro- ing the roundness error of the hole, and developed a
cess C45, and the results showed that the use of coat- system control equation composed of exciting force to
ings can effectively reduce tool damage and improve study mechanism of roundness error of the hole. This
the surface finish of threads. Hsu et al.[20] used the will provide great help for this research.
Taguchi method to test and analyze the influence of Since the previous research in this area is still blank,
tool parameters and cutting conditions on thread qual- the blind cutting experiments are not only difficult to
ity during tapping, and found that the larger the helix determine the factors that affect the dimensional of R-
angle, the worse the quality of the thread. Bratan et PDD in tapping, but the cost of the experiment is also
al.[21] developed a tool to improve the threading pro- very high. Based on previous studies, this paper will es-
cess considering the combined deforming-cutting tap. tablish a quasi-static model of vibration with reference
The results have shown that utilising proposed com- to the analysis method of roundness error of the hole,
bined deforming-cutting taps in the processing of inter- and finally determine the dimensional of RPDD, lay-
nal threads with a small diameter for M3-M6 in alu- ing a solid foundation for the next step of experimental
minium alloys enhanced the accuracy and surface qual- research.
ity of threads. It can be seen that the current literatures
on the thread quality obtained by tapping is mostly fo- 2 Quasi-static model
cused on the influence of various factors on the thread
surface quality, and the research on the RPDD of the During the tapping process, as the tap enters the pre-
thread is almost blank. Therefore, it is very necessary drilled hole at an axial rate synchronized with the thread
to analyze the factors that affect the dimensional of R- lead, each tooth cuts a layer of material from the sur-
PDD by establishing a mathematical model of RPDD. face formed by the previous tooth. In order to ensure
For a standard internal thread, its pitch diameter the smooth execution of the cutting process, the thread
can be determined by the nominal diameter and pitch of end of the tap is made into a tapered surface with κr as
the thread, that is, pitch diameter = nominal diameter- a chamfer to truncated the full thread, and a number
pitch × [Link], it can be considered that R- of straight flutes for chip evacuation evenly distribut-
PDD of the thread is the result of the combined effect ed over the entire thread length divide the thread into
of the roundness error of the root circle and the pitch several continuously distributed cutting teeth. The ma-
error. The mechanism of roundness error of the root jor cutting edges of the chamfered section of the tap
cycle is similar to that of the hole. From the litera- are formed by the intersection of the tapered surface
ture[22], it is known that the roundness error of the and the straight flute surfaces, and the minor cutting
Mathematical model for tapping simulation to predict Radial Pitch Diameter Difference of threads 3

R
ky cy
R(i+1)j
O2
T

cx (x(t),y(t)) fi Rij
fi
O1 X O1 X
kx

Y Y
a b c

Fig. 1 Dynamic model of tapping process by considering the transverse vibrations

edges lie on the flank surfaces of the tap threads. As As shown in Figure 1b, the dynamic model of the
can be seen from the above description, the geometry tap only considers the two orthogonal degrees of free-
of the tap is very complicated. In order to convenience dom of the tool in the radial direction. The differential
the description of the tapping process, some coordinate equation of the two-degree-of-freedom transverse vibra-
systems first need to be established. The global coor- tion of the dynamic tapping system in the global coor-
dinate system X-Y-Z whose origin O1 is at the center dinate system is:
of the hole is fixed on the workpiece surface, and the
M Ẍ + C Ẋ + KX = F (3)
Z axis coincides with the hole axis. The cutting edge
coordinate system is U-V-W with the origin O2 at any among them, M, C, K are the mass matrix, damping
designated point of the elementary cutting edge, and matrix and stiffness matrix of the tap respectively, and
the W axis coincides with the major cutting edge. The they are all second-order matrices.
local rotating coordinate system T-R-A is the transi- Whitehead[26] and Towfighian[27] mentioned that
tion coordinate system between the global coordinate the vibrations to cause roundness errors are divided into
system and the cutting edge coordinate system. Ro- two categories: chatter and low frequency vibration. In
tating the global coordinate system around the Z axis tapping process, the speed of the tap is usually low, and
through φi can be converted into the T-R-A coordinate the frequency is much lower than the lowest natural
system that changes with the position of the elementary frequency of the tool. At this time, the stiffness term
cutting edge, and then rotating this coordinate system in the equation plays a leading role, while the inertia
around the X and Y axes through λ and κr respectively term and damping term can be ignored. Therefore the
to convert to cutting edge coordinates system, as shown above equation becomes the following form:
in Figures 1a and 2a. The three coordinate systems can KX = F (4)
be transformed into each other through the transfor-
mation matrix. The relationship between them can be Equation 4 eliminated the inertia and damping terms in
expressed as: the transverse vibration differential equation is called a
      quasi-static model. The exciting force F includes regen-
X T U erative cutting force and cutting damping force due to
 Y  = T 3  R  = T 1 T2 T 3  V  (1) the waviness on the pre-drilled surface. For the conve-
Z A W nience of research, let the number of tap flutes be Nf ,
where and the tap are evenly dispersed into Nz slices with
  length dz along the axial direction, then the force of
cos κr 0 sin κr the entire tap is distributed to the Nf × Nz elemen-
T1 =  0 1 0  tary cutting edges. According to the cutting principle
− sin κr 0 cos κr of the tap, each elementary cutting edge is regarded as
an oblique cutting model.
 
1 0 1
T2 =  0 cos λ − sin λ  (2)
0 sin λ cos λ 2.1 Cutting force
 
cos φi − sin φi 0
T3 =  sin φi cos φi 0  The cutting force of oblique cutting is proportional to
0 0 1 the cross-sectional area of the chip to be removed. There-
4 Jie Ren1 , Xianguo Yan1

fore, the cutting force Fc can be calculated by the fol-

Workpiece
lowing formula:

Fc = kc A = kc hb (5)
Tap
where kc is the cutting force coefficient, h is the cutting

s
hks
thickness, and b is the cutting width. hke

Fig. 3 Axial position of each cutting edge on the straight

flute tap
∆Rij

Tap
∆hij
As the tapping depth increases, each cutting edge
workpiece Actual
position Nominal
will experience partial and full engagement status. There-
position
fore, the cutting force is constantly changing with the
r

a b
effective cutting edge length involved in cutting dur-
ing the continuous tapping process. It can be seen from
Fig. 2 Position change of the tap axis under transverse vi- the geometric relationship in Figure 3 that the length
bration of the cutting edge can be determined by its start and
end positions in the axial direction. The start and end
positions of each cutting edge in the axial direction can
When the tool’s actual trajectory (shown by the
be expressed as:
dashed line in Figure 2b) deviates from the nominal one
due to transverse vibration (shown by the solid line in
Figure 2b), an additional dynamic uncut chip thickness
√
will attach to the desired static uncut chip thickness. 3P (k − Nf ) cos κr
Since the static uncut chip thickness does not cause hks =
2Nf cos( π6 + κr )
vibration, only the load caused by the dynamic uncut
2sNf sin(2κr + π3 )
chip thickness remains in the cutting force. Therefore, hke = (10)
Nf (2 cos(2κr + π3 ) − 1)
the radial, tangential and axial components of the cut- √
ting force of elementary cutting edge in the cutting edge 3kP sin 2κr + 2sNf cos(2κr + π6 )
−
coordinate system U-V-W can be expressed as: Nf (2 cos(2κr + π3 ) − 1)
   
∆Fcu,ij kcu
 ∆Fcv,ij  =  kcv  ∆hij dz (6) where k represents the serial number of the cutting
∆Fcw,ij kcw edge, hks is the start position of the k th cutting edge,
hke is the end position of the k th cutting edge, and P is
where ∆hij represents the dynamic uncut chip thick- the pitch. Since the tap cutting edge is not continuous-
ness of the j th elementary cutting edge corresponding ly distributed, the window function g(z) is introduced
to the i th flute. If the transverse displacement of the to identify whether the elementary cutting edge corre-
tap axis at time t is represented by x(t) and y(t), the sponding to a certain flute engages with the workpiece
transverse displacement ∆Rij of the elementary cutting at time t. The elementary cutting force can be expressed
edge can be determined by the displacement of the tap in the U-V-W coordinate system as:
axis and the tooth angle φi (Figure 1c):

∆Rij = Rij − R(i−1)j = ∆x cos φi + ∆y sin φi (7)

 
∆Fcu,ij
where  ∆Fcv,ij 
∆Fcw,ij
∆x = x(t) − x(t − τ0 )   (11)
kcu [∆x cos φi + ∆y sin φi ] cos κr
(8)
∆y = y(t) − y(t − τ0 ) =  kcv [∆x cos φi + ∆y sin φi ] cos κr  g(z)dz
kcw [∆x cos φi + ∆y sin φi ] cos κr
where τ0 is the time interval between adjacent teeth.
The dynamic uncut chip thickness can be expressed as:
The elementary cutting force in the T-R-A coordi-
∆hij = ∆Rij cos κr nate system can be expressed as follows based on coor-
(9)
= [∆x cos φi + ∆y sin φi ] cos κr dinate transformation:
Mathematical model for tapping simulation to predict Radial Pitch Diameter Difference of threads 5

and ∆Fcz in the tangential, radial and axial directions

  of the tap can be calculated as follows:
∆Fct,ij
 ∆Fcr,ij   
∆Fcx
∆Fca,ij  ∆Fcy 
  (12)
kcu [∆x cos φi + ∆y sin φi ] cos κr ∆Fcz
= T1 T2  kcv [∆x cos φi + ∆y sin φi ] cos κr  g(z)dz  
Nf
X Z kcu   
kcw [∆x cos φi + ∆y sin φi ] cos κr
l0  ∆x
= T3 T1 T2  kcv  cos φi sin φi cos κr g(z)dz
l ∆y
i kcw
where
(18)

1, z ∈ (hks , hke ), k = 0, 1, 2, ... through simplification, Equation 18 can be expressed
g(z) = (13)
0, others as:
   
∆Fcx ∆x
= Kc (19)
∆Fcy ∆y
τs cos(βn − γn ) + tan λ tan η sin βn
kcu = p
sin ψn cos2 (ψn + βn − γn ) + tan2 η sin2 βn where
τs sin(βn − γn ) Kc =
kcv = p
sin ψn cos λ cos2 (ψn + βn − γn ) + tan2 η sin2 βn  
Nf Z kcu 
X l0  (20)
τs cos(βn − γn ) tan λ + tan η sin βn T3 T1 T2  kcv  cos φi sin φi cos κr g(z)dz
kcw = p l
sin ψn cos2 (ψn + βn − γn ) + tan2 η sin2 βn i kcw
(14)
2.2 Cutting damping force
where τs is the shear stress, ψn is the normal shear an-
gle, λ is the inclination angle, βn is the normal friction The actual cutting edge is not sharp, but is a small arc,
angle, γn is the normal rake angle, and η is the chip as shown in Figure [Link], the edge force and the
flow angle, and these values can be obtained according cutting damping force are generated because the mate-
to the method in literature [28, 29].In the T-R-A coordi- rial under the flank face of the cutting edge is extrud-
nate system, the cutting force of all elementary cutting ed[30].Since the edge force does not cause vibration,
edges corresponding to a certain flute can be expressed only the cutting damping force remains. The cutting
as: damping force is caused by the interference between the
uneven workpiece surface caused by the tool vibration
  and the tool flank surface, which causes the effective
∆Fct,i
 ∆Fcr,i  flank angle αef f (as shown in Figure 4a) to change. Lit-
∆Fca,i erature [31] mentioned that the cutting damping force
  (15)
Z kcu    is composed of the normal force Fdv and the friction
l0  ∆x
= T1 T2  kcv  cos φi sin φi cos κr g(z)dz force Fdu , and the normal force Fdv is proportional to
l ∆y
kcw
the volume V of the workpiece material extruded under
where l0 is the start cutting position, l is the current the flank face of the cutting edge:
cutting position, they can be expressed as: Fdv = Ksp V (21)
D h − d0 where Ksp is the specific contact force, which depends
l0 =
2 tan κr on the material and the geometry of the cutting edge.
(16)
nP t The volume V is related to the geometry of the cut-
l=
2π ting edge, the vibration speed ċ perpendicular to the
machined surface and the cutting speed vc (Figure 4a):
where Dh is the diameter of the pre-drilled hole with a
value of 8.5mm, d is the tap nominal diameter, and d0 1 2 ċ
V = l b (22)
is the chamfer point diameter. 2 w vc
where lw is the interference length between the work-
d0 = d − 1.2P (17) piece surface and the flank face of the tool. According
to the model proposed by Ahmadi et al.[30], the inter-
By converting the cutting forces of all flutes from the
ference length lw can be written as:
local coordinate system to the global coordinate system
and then adding them, the cutting forces ∆Fcx , ∆Fcy l w = lh + lc + lf (23)
6 Jie Ren1 , Xianguo Yan1

Workpiece surface

a eff
rh

a
vc θ a
S
. lh lc lf
c

Fig. 4 Tool path and cutting damping force geometry

where T-R-A coordinate system can be determined as follows:

 
lh = rh sin θ ∆Fdt,ij
 ∆Fdr,ij 
lc = rh sin α (24)
∆Fda,ij
lf = rh [cos α − cos α] cot α  µkv [∆x cos φi +∆y sin φi ] cos κr  (30)
τ vc
 kv [∆x cos φi +∆y sin φi ] cos κr 
where rh is the hone radius, α is the flank angle and θ is = T1 T2   g(z)dz
τ vc
the separation angle which defines the position where 0
the workpiece material is no longer removed as chips
Like the method of obtaining the cutting force, the
through the rake face, but is extruded under the flank
tangential ∆Fdx , radial ∆Fdy and axial ∆Fdz compo-
face of the tool. Therefore, the normal component Fdv
nents of the cutting damping force can be expressed
of the cutting damping force can be written as:
as:
ċ  
Fdv = kv b (25) ∆Fdx
vc  ∆Fdy 
where ∆Fdz
 µkv

Nf
X Z  
1 2
l0
 τ vc
kv   ∆x
kv = Ksp lw (26) = T3 T1 T2   cos φi sin φi cos κr g(z)dz
2 l
τ vc ∆y
i 0
The friction component Fdu of the cutting damping (31)
force is considered to be proportional to the normal
component Fdv , so it can be written as: through simplification, Equation 31 can be expressed
as:
Fdu = µFdv (27)    
∆Fdx ∆x
= Kv (32)
where µ is the Coulomb friction coefficient, which is ∆Fdy ∆y
taken as 0.3 according to literature [32]. Therefore, the
radial and tangential components of the cutting damp- where
ing force of the elementary cutting edge in the U-V-W Kv =
coordinate system can be expressed as:  µkv

Nf Z
X l0
 τ vc
kv   (33)
T3 T1 T2   cos φi sin φi cos κr g(z)dz
   
∆Fdu,ij µkv τ vc
 ∆Fdv,ij  =  kv  ∆hij dz i l
(28) 0
τ vc
∆Fdw,ij 0

where n is the spindle speed, vc is the cutting speed at 2.3 Elastic force
different axial positions.
In the quasi-static equation, KX is the elastic force,
vc = [πd0 + nP t tan κr ]n (29) where K is the stiffness matrix of the tool. For a two-
degree-of-freedom system, it can be expressed as:
Substituting the dynamic uncut chip thickness into E-
quation 28, and then based on the coordinate trans-
 
k k
K = xx xy (34)
formation, the elementary cutting damping force in the kyx kyy
Mathematical model for tapping simulation to predict Radial Pitch Diameter Difference of threads 7

Due to the small bending deformation of the tool,

the vibration of the tool can be considered to be re- 
X(t)

stricted to a plane perpendicular to the axis of the un-  X(t − τ ) 
 
deflected tool. Therefore, the relationship between the 

.. 

.
elastic force and the axis offset can be expressed as:
X(t − mτ + τ )
 
B(t − τ ) 0 ··· 0 A(t − τ ) (40)
 
  
k k x(t)  I 0 ··· 0  X(t − τ )
KX = xx xy (35)  ··· 
kyx kyy y(t)  .. .. .. ..   X(t − 2τ ) 
 . .   
= . . × . 
   .. 
 .. .. .. .. 
 . . . .  X(t − mτ )
It is assumed that the tap is a cantilever beam whose
0 ··· ··· I 0
force is concentrated on the tip of the tool in this analy-
sis. According to literature [27],the stiffness matrix can this is
be written as:
  qt = Qt−τ qt−τ (41)
EI 30
K= 3 (36) where the matrix Qt−τ is the state-transition matrix at
L 03
time t−τ ,I is a second-order identity matrix. According
to the literature [34], Equation 41 can be written as
where EI is the bending stiffness of the tool material,E
follows:
is the elastic modulus of material of the tap,and L is
the total length of the tap. Qt−τ qt−τ = µt−τ qt−τ (42)

where
2.4 Solutions of quasi-static equations µ = eλτ (43)

Now substituting Equations 19 and 32 into Equation where λ is a characteristic exponent, whose imaginary
4, the quasi-static equilibrium equation is rewritten as and real parts represent the oscillation frequency and
follows according to the displacement of the tap axis: growth or decay rate respectively.
Equation 42 is a characteristic equation with m eigen-
KX = Kc [X − X(t − τ0 )] + Kv [X − X(t − τ )] (37) values. The eigenvalues and eigenvectors satisfying the
matrix Qt−τ are µt−τ and qt−τ respectively, which can
be determined by solving the characteristic equation.
The current value of X is expressed as a function
Therefore, the instantaneous position vector Xi t−τ of
of the previous value, and the friction time delay τ is
m modes at time t − τ can be extracted from the m
approximately a fraction of the cutting time delay τ0 ,
eigenvectors respectively. The displacement vector Xi
that is, τ = τ0 /m. Therefore, the successive tool axis
of the i th mode at time t can be written as:
position equation can be expressed as:
X i = µit−τ Xt−τ
i
, (i = 1, 2, ..., m) (44)
X = A(t − τ )X(t − mτ ) + B(t − τ )X(t − τ ) (38)
At time t, the displacement vector X of the tool ax-
is can be obtained by superimposing the displacement
where
vectors of m modes. The pitch diameter of the thread
will also change with the axis offset. The pitch diameter
A(t) = [K − Kc (t) − Kv (t)]−1 Kc (t)
(39) at time t can be expressed as:
B(t) = [K − Kc (t) − Kv (t)]−1 Kv (t)
D2 (t) = ∆D(t) + D − 0.6495P (45)
Equation 38 is a difference equation with variable
where D2 (t) is the pitch diameter of the thread at time
coefficients. In order to obtain the solution of the e-
t, and D is the nominal diameter of the thread which
quation, based on the freezing coefficient method[33],
is equal to the nominal diameter d of the tap,the offset
the successive cutting process is discretized into several
∆D(t) at time t can be expressed as:
steps with time τ , each of which is approximated by
a constant coefficient difference equation. For the con- ∆D(t) = x(t) cos φ(t) + y(t) sin φ(t) (46)
venience of solving, Equation 38 is written as a large
matrix form: where φ(t) is the angle of rotation in time t.
8 Jie Ren1 , Xianguo Yan1

90
Pitch diameter

Maximum Pitch diameter

5 2 threads Minimum pitch diameter

0 180 0

270

90
Pitch diameter

4 threads
Maximum Pitch diameter
5
Minimum pitch diameter

0 180 0

270

90
Pitch diameter

Maximum Pitch diameter

5 8 threads Minimum pitch diameter

0 180 0

270

Fig. 5 Change of the pitch diameter of the first full thread and the movement trajectory of the outermost cutting edge of the
tap in the global coordinate system during the entire cutting process when the chamfer length is 2 threads, 4 threads and 8
threads

Table 1 Geometric parameters of straight flute taps Table 2 Parameters of each group test
Nominal Diameter Tap Length Rake Angle Flank Angle 1 2 3 4 5 6
Flute Number
(mm) (mm) (deg) (deg)
Spindle Speed
700 1050 1400 1750 700 700
10 80 8 4 3 (rev/min)
Chamfer Length
4 4 4 4 2 8
(threads)

RPDD of the thread can be expressed as the differ-

ence between the maximum and minimum pitch diam-
eters:
RP DD = D2max − D2min (47) ject to regenerative cutting and cutting damping forces,
where D2max and D2min are the maximum and mini- is found at every time step in the simulation, and then
mum pitch diameters of the thread, respectively. the dimensional of RPDD is determined by the Equa-
tion [Link] tool used in the simulation is a standard
straight flute high-speed steel tap, and its parameter-
3 Simulation s are shown in Table [Link] workpiece is an AISI1045
plate with pre-drilled holes. In the cutting simulation,
Time domain simulation is performed based on the tool 6 groups of tests whose parameters are listed in Table
motion equation of Equation 38, which is also quasi- 2 were performed to study the influence of the chamfer
static. The static equilibrium position of the tool, sub- length and the spindle speed on RPDD.
Mathematical model for tapping simulation to predict Radial Pitch Diameter Difference of threads 9

90
700 rev/min Pitch diameter

Maximum Pitch diameter

5
Minimum pitch diameter

0 180 0

270

90 Pitch diameter
1050 rev/min Maximum Pitch diameter
5
Minimum pitch diameter

0 180 0

270

90 Pitch diameter
1400 rev/min Maximum Pitch diameter
5
Minimum pitch diameter

0 180 0

270

90
1750 rev/min Pitch diameter

Maximum Pitch diameter

5
Minimum pitch diameter

0 180 0

270

Fig. 6 Change of the pitch diameter of the first full thread and the movement trajectory of the outermost cutting edge of
the tap in the global coordinate system during the entire cutting process when the spindle speed is 700rev/min, 1050rev/min,
1400rev/min and 1750rev/min

4 Results the real part will decrease and eventually remain con-
stant, and the tool begins to vibrate periodically. The
From the simulation results, it is found that the imag- change of chamfer length and spindle speed will affect
inary part of the characteristic exponent of each mode the characteristic exponent. As the chamfer length in-
remains almost unchanged over time, while the real part creases, the increase in the real part of the characteristic
starts from a negative value and gradually becomes pos- exponent will increase, while the increase in the spindle
itive as time passes and the tap proceeds, resulting in speed will reduce the rate of change of the real part of
an increase in the vibration amplitude. With the fur- the characteristic exponent.
ther engagement of the cutting edge and the workpiece,
10 Jie Ren1 , Xianguo Yan1

In order to analyze the influence of the chamfer

length and the spindle speed on RPDD, the eigenval-
ues and eigenvectors of all modes of each group of tests
are obtained by simulation. The vibrations of first four
modes which are selected from all modes according to
the rate of decay or growth of each mode because of
their importance are considered here, and RPDD ob-
tained is a combination of these four modes.
Figure 5 shows the change of the pitch diameter
of the first full thread and the movement trajectory
of the outermost cutting edge of the tap in the glob-
al coordinate system during the entire cutting process
when the chamfer length is 2 threads, 4 threads and 8
threads. From the comparison of the three figures, it
is found that the change of the chamfer length has an
effect on [Link] can be seen in Figure 7a, the di- Fig. 7 Dimensional of RPDD under different spindle speed
mensional of RPDD shows a trend of first increasing and chamfer length
and then decreasing with the increase of the chamfer
length and the smallest RPDD is obtained when the
PDD is predicted from the characteristic solutions of
chamfer length is 2 threads. The possible reason is that
the first four modes.
fewer cutting edges are involved in cutting, causing the
This model was used to simulate the tapping of
vibration amplitude to increase briefly and then become
AISI1045 plates using standard straight flute high-speed
a steady periodic vibration.
steel taps, and the dimensional of RPDD under differ-
Figure 6 shows the change of the pitch diameter of
ent chamfer lengths and spindle speeds was obtained.
the first full thread and the movement trajectory of the
By comparison, it is found that the chamfer length and
outermost cutting edge of the tap in the global coor-
spindle speed will affect RPDD, and RPDD is the s-
dinate system during the entire cutting process when
mallest when the chamfer length is 2 threads and the
the spindle speed is 700 rev/min, 1050 rev/min, 1400
spindle speed is 1400 rev/min. Since the research on R-
rev/min and 1750 rev/min. From the comparison of the
PDD is still blank so far, the development of this model
three figures, it is found that the change of the spindle
not only provides a cheap and effective method for the
speed has an effect on [Link] can be seen in Fig-
study of RPDD, but also lays a foundation for further
ure 7b, the dimensional of RPDD shows a trend of first
experimental research.
decreasing and then increasing with the increase of the
spindle speed and the smallest RPDD is obtained when This model provides a method for the research of
the chamfer length is 1400 rev/min. The possible reason RPDD and the research on the stability of tapping in
is that more cutting edges quickly participate in cutting the literature [23] also confirms the conclusion of this
when the spindle speed increases, resulting in a rapid article, but it needs to be verified by experiments in
increase in damping and a decrease in amplitude. But order to better illustrate its effectiveness. In addition,
at the same time, the increase of the spindle speed will in the actual tapping process, the distributed force acts
also lead to the decrease of the damping change rate, on the entire engagement length of the tool, but this
so a higher speed will restrain the amplitude reduction. model applies concentrated force on the tip of the tool,
which will have a certain impact on the accuracy of the
model. These issues will be studied in the future.
5 Summary
Funding information The research is financially sup-
A quasi-static tapping model, including delayed cutting ported by the National Natural Science Foundation of
force and cutting damping force was established. A sim- China (No. 51275333)
ilar cutting force model and cutting damping force were
developed based on the work of previous researchers.
The force equilibrium equation is expressed as a discrete-
time matrix equation describing the successive state of Conflict of interest
the tool. According to the eigenvalues and eigenvectors
of the state transition matrix, the characteristic solu- The authors have no relevant financial or non-financial
tion of the equilibrium equation is found, and then R- interests to disclose.
Mathematical model for tapping simulation to predict Radial Pitch Diameter Difference of threads 11

Availability of data and material References

The data sets supporting the results of this article are 1. Fetullazade E, Akyildiz HK, Saritas S (2010) Effects of
the machining conditions on the strain hardening and the
included within the article and its additional files. residual stresses at the roots of screw threads. Mater Des
31:2025-2031.
2. Dong Y, Hess DP (1999) The effect of thread dimensional
conformance on vibration-induced loosening. J Vib Acoust
Authors’ contributions 121:209–213.
3. Korin I, J. Perez Ipina (2011) Experimental evaluation of
fatigue life and fatigue crack growth in a tension bolt–nut
Jie Ren developed a mathematical model to predict the threaded connection. Int J Fatigue 33:166-175.
radial diameter difference of threads, analyzed the sim- 4. Pesin MV (2012) Improving the reliability of threaded pipe
ulation results, and was a major contributor in writing joints. Russ Eng Res 32:210–212.
5. Min J (2015) Comprehensive measurement system for
the manuscript. Xianguo Yan provided guidance for the screw thread parameter based on machine vision. In: 2015
writing of manuscript. All the authors read and ap- International Conference on Computer and Computational
proved the final manuscript. Sciences (ICCCS), pp 266-270
6. Araujo AC, Mello GM, Cardoso FG (2015) Thread milling
as a manufacturing process for API threaded connection:
Geometrical and cutting force analysis. J Manuf Process
18:75-83.
Ethics approval 7. F.M Leon and N.G Pai and D.P Hess (2001) The effec-
t of thread dimensional conformance on yield and tensile
All analyses in this paper are based on previously pub- strength. Eng Fail Anal 8:49-56.
lished research and this paper does not involve animal 8. Nassar SA, Munn BS, Yang X (2009) Effect of Non-
Conforming Thread Root Radius on the Fatigue Perfor-
and human testing, so this item is not applicable to this mance of Preloaded Threaded Fasteners. In: ASME 2008
paper. Pressure Vessels and Piping Conference, pp 803-809.
9. Carvalho AOD, Branda eo LC, Panzera TH, Lauro CH
(2012) Analysis of form threads using fluteless taps in
cast magnesium alloy (AM60). J Mater Process Technol
212:1753-1760.
Consent to participate
10. Igor, Cézar, Pereira, et al (2016) Analysis of tapping pro-
cess in three types of cast iron. Int J Adv Manuf Technol
All analyses in this paper are based on previously pub- 82:1041-1048.
lished research and this paper does not involve animal 11. Malkov OV, Malkova LD (2019) Improving thread accu-
racy in machining components for rocket and space tech-
and human testing, so this item is not applicable to this
nologies. AIP Conf Proc 2171:200006.
paper. 12. Khoshdarregi M, Altintas Y (2018) Dynamics of multi-
point thread turning—part I: General formulation. J Manuf
Sci Eng 140: 061003.
13. Araujo AC, Fromentin G (2018) Investigation of tool de-
Consent for publication flection during milling of thread in Cr-Co dental implant.
Int J Adv Manuf Technol 99, 531-541.
14. Malkov OV, Karelsky AC (2019) Rising the work unifor-
The Author confirms: mity of thread milling cutters in machining parts of rocket
that the work described has not been published before and space technology. AIP Conf Proc 2171:200005.
(except in the form of an abstract or as part of a pub- 15. Wan M, Altintas Y (2014) Mechanics and dynamics of
thread milling process. Int J Mach Tools Manuf 87:16-26.
lished lecture, review or thesis); 16. Khoshdarregi MR, Altintas Y (2015) Generalized mod-
that it is not under consideration for publication else- eling of chip geometry and cutting forces in multi-point
where; thread turning. Int J Mach Tools Manuf 98:21-32.
that its publication has been approved by all co-authors 17. Wan M, Ma YC, Feng J, Zhang WH (2017) Mechanics
of tapping process with emphasis on measurement of feed
,if any; error and estimation of its induced indentation forces. Int
that its publication has been approved (tacitly or ex- J Mach Tools Manuf 114:8-20.
plicitly) by the responsible authorities at the institution 18. Fromentin G, Poulachon G, Moisan A, et al (2005) Pre-
cision and surface integrity of threads obtained by form
where the work is carried out.
tapping. CIRP Ann - Manuf Technol 54:519-522.
The author agrees to publication in the Journal indi- 19. Piska M, Sliwkova P (2015) A study of cutting and form-
cated below and also to publication of the article in En- ing threads with coated HSS taps. J Mach Eng 15:65-74.
glish by Springer in Springer’s corresponding English- 20. Chuan-Chi Hsu, Syh-Shiuh Yeh, Jien-I Lee (2016) Effec-
t analysis and optimal combination of cutting conditions
language journal. The copyright of the English article is on the cutting torque of tapping processes using Taguchi
transferred to Springer effective if and when the article methods. In: 2016 IEEE International Conference on Au-
is accepted for publication. tomation Science and Engineering (CASE), pp 1215-1218.
12 Jie Ren1 , Xianguo Yan1

21. Bratan S, Novikov P, Roshchupkin S (2016) Application

of Combined Taps for Increasing the Shaping Accuracy of
the Internal Threads in Aluminium Alloys. Procedia Eng
150:802-808.
22. Ahmadi K, Altintas Y (2013) Stability of lateral, torsion-
al and axial vibrations in drilling. Int J Mach Tools Manuf
68:63-74.
23. Ma YC, Wan M, Yang Y, Zhang WH (2019) Dynamics
of tapping process. Int J Mach Tools Manuf 140:34-47.
24. Bayly PV, Young KA, Calvert SG, Halley JE (2001)
Analysis of tool oscillation and hole roundness error in a
quasi-static model of reaming. J Manuf Sci Eng 123:387-
396.
25. Deng CS, Chin JH (2004) Roundness errors in BTA
drilling and a model of waviness and lobing caused by res-
onant forced vibrations of its long drill shaft. J Manuf Sci
Eng 126:524-534.
26. Whitehead BT, Bayly PV, Calvert SG (2001) The Effect
of Process Damping On Stability and Hole Form in Drilling.
SAE International, Warrendale, PA.
27. S Towfighian, K Behdinan, M Papini, et al (2007) Fi-
nite element modeling of low speed reaming vibrations with
reamer geometry modifications. J Intell Manuf 18:647-661.
28. Chen NM, Smith AJR (2011) Modelling of straight-flute
machine tapping. Proc Inst Mech Eng Part B J Eng Manuf
225:1552-1567.
29. Kaymakci M, Kilic ZM, Altintas Y (2012) Unified cut-
ting force model for turning, boring, drilling and milling
operations. Int J Mach Tools Manuf 54-55:34-45.
30. Ahmadi K, Savilov A (2015) Modeling the mechanics and
dynamics of arbitrary edge drills. Int J Mach Tools Manuf
89:208-220.
31. Jiménez A, Arizmendi M, Cumbicus WE (2018) Mod-
el for the prediction of low-frequency lateral vibrations in
drilling process with pilot hole. Int J Adv Manuf Technol
96:1971-1990.
32. Eynian M, Altintas Y (2010) Analytical Chatter Stabil-
ity of Milling With Rotating Cutter Dynamics at Process
Damping Speeds. J Manuf Sci Eng 132:021012.
33. Wang C, Guan W, Wang JY, et al (2018) Adaptive op-
erational modal identification for slow linear time-varying
structures based on frozen-in coefficient method and lim-
ited memory recursive principal component analysis. Mech
Syst Signal Process 100:899-925.
34. Insperger T, Stépán G (2011) Semi-Discretization for
Time-Delay Systems. Springer New York, New York.
Figures

Figure 1

Dynamic model of tapping process by considering the transverse vibrations

Figure 2

Position change of the tap axis under transverse vibration

Figure 3

Axial position of each cutting edge on the straight flute tap

Figure 4

Tool path and cutting damping force geometry

Figure 5

Change of the pitch diameter of the first full thread and the movement trajectory of the outermost cutting
edge of the tap in the global coordinate system during the entire cutting process when the chamfer length
is 2 threads, 4 threads and 8 threads
Figure 6

Change of the pitch diameter of the first full thread and the movement trajectory of the outermost cutting
edge of the tap in the global coordinate system during the entire cutting process when the spindle speed
is 700rev/min, 1050rev/min, 1400rev/min and 1750rev/min
Figure 7

Dimensional of RPDD under different spindle speed and chamfer length