S99TE08-0206 9

4. Design and Selection of Ballscrews

4.1 Fundamental Concepts for Selection & Installation
(1). A ballscrew must be thoroughly cleaned in white
spirit and oil to protect against corrosion.
Trichlorethylene is an acceptable degreasing
agent, ensuring the ball track free from dirt and
damage (paraffin is not satisfactory). Great care
must be taken to ensure that the ball track is not
struck by a sharp edged component or tool, and
metallic debris does not enter the ball nut (Fig.
4.1).

Fig 4.1
Carefully clean
and protect.
Fig 4.2(b) The effect on service life of a radial
(2) Select a suitable grade ballscrew for the applica-
load caused by misalignment.
tion (ref. Table 4.5). Install with corresponding
mounting disciplines. That is, precision ground (3) To achieve the
ballscrews for CNC machine tools demand accu- ballscrews’ maxi-
rate alignment and precision bearing arrange- mum life, recom-
ment, where the rolled ballscrews for less preci- mends the use of
sion applications, such as packaging machinery, antifriction bearing
require less precise support bearing arrange- oils.Oil with graphite
ment. and M0S2 additives
It is especially important to eliminate misalign- must not be used.
ment between the bearing housing center and the The oil should be Fig 4.2 Oil lubrication
ballnut center, which would result in unbalanced maintained over the method
loads (Fig. 4.2). Unbalanced loads include radial balls and the ball-
loads and moment loads (Fig. 4.2a). These can tracks.
cause malfunction and reduce service life
(Fig.4.2b). (4) Oil mist bath or drip feeds are acceptable.
However, direct application to the ball nut is
recommended (Fig. 4.3).

Fig 4.2(a) Unbalance load caused by misalignment

of the support bearings and nut brackets,
inaccurate alignment of the guide surface,
inaccurate angle or alignment of the nut
mounting surface Fig 4.3 Carefully protect the nut
10 S99TE08-0206
(5) Select a suitable support bearing arrangement for the screw spindle. Angular contact ball bearings
(angle=60˚) are recommended for CNC machinery. Because of higher axial load capacity and ability to pro-
vide a clearance-free or preloaded assembly (Fig. 4.4).

Fig 4.4 Different arrangement of ballscrew support bearings

(6) A dog stopper should be installed at the end to

prevent the nut from over-travelling which results
in damage to ballscrew assembly (Fig 4.5).

Fig 4.7 Special arrangement for the end journal of an

internal recirculation screw.

(9) After heat treating the ballscrew spindle, both

ends of the balltracks adjacent to the journal
Fig 4.5 A dog stopper to prevent the nut from over
have about 2 to 3 leads left soft, for the purpose
travelling
of machining. These regions are shown in (Fig.
(7) In environments contaminated by dust or metallic 4.8) with the mark “ ”on HIWIN drawings.
debris, ballscrews should be protected using tele- Please contact engineers if special require-
scopic or bellow-type covers. The service life of a ments are needed in these regions.
ballscrew will be reduced to about one-tenth nor-
mal condition if debris or chips enter the nut. The
bellow type covers may need to have a threaded
hole in the flange to fix the cover. Please contact
engineers when special modifications are need-
ed (Fig 4.6).

Fig 4.8 The heat treatment range of the ballscrew

spindle
(10) Excessive preload increases the friction torque
and generates heat which reduces the life
expectancy. But insufficient preload reduces
stiffness and increases the possibility of lost
Fig 4.6 Ballscrew protection by telescopic or bellow
motion. recommends that the maximum preload
type covers
used for CNC machine tools should not exceed
(8) If you select an internal recirculation type or an 8% of the basic dynamic load C (106 revs).
endcap recirculation type ballscrew, one end of
the ball thread must be cut through to the end (11) For an internal recirculation nut, when the nut
surface. The adjacent diameter on the end jour- needs to be disassembled from/assembled to
nal must be 0.5 ~ 1.0 mm less than the root the screw spindle, a tube with an outer dia. 0.2
diameter of the balltracks (Fig 4.7). to 0.4 mm less than the root diameter (ref. M39)
of the balltracks should be used to release/con-
nect the nut to from/to the screw spindle via one
end of the screw spindle shown in Fig. 4.9.
 S99TE08-0206 11

Fig 4.9 The method of separating the nut from the Fig 4.10 Chamfer for seating the face of bearing
screw spindle end.

(12)As shown in Fig 4.10, the support bearing must have a chamfer to allow it to seat properly and maintain prop-
er alignment. suggests the DIN 509 chamfer as the standard construction for this design (Fig. 4.11).

Fig 4.11 Suggested chamfer dimension per DIN 509 for the "A" dimension in Fig 4.10
12 S99TE08-0206
4.2 Ballscrews Selection Procedure
The selection procedure for ballscrews is shown in (Table 4.1) From the known design operation condition,
(A) select the appropriate parameter of ballscrew, (B) follow the selection procedure step by step via the refer-
ence formula, and (C) find the best ballscrew parameters which can be met for the design requirements.

Step Design operation condition (A) Ballscrew parameter (B) Reference formula(C)

Step 1 Positioning accuracy Lead accuracy Table 5.2

(1) Max. speed of DC motor (Nmax) Ballscrew lead
Step 2 ≥ Vmax
(2) Rapid feed rate (Vmax) Nmax
Total length=
thread length + journal end length
Step 3 Total travel distance Total thread length
Thread length =
stroke + nut length + 100 mm
(unused thread)
(1) Load condition(%) Mean axial load M7 ~ M10
Step 4
(2) Speed condition(%) Mean speed
Step 5 Mean axial force (≤ 1/5 C is the best) Preload M1
(1) Service life expectancy Basic dynamic load M13 ~ M14
Step 6 (2) Mean axial load
(3) Mean speed
(1) Basic dynamic load Screw diameter and nut type M31 ~ M 33 and dimension table
(2) Ballscrew lead (select some range)
Step 7
(3) Critical speed
(4) Speed limited by Dm-N value
(1) Ballscrew diameter Stiffness M34 ~ M40
(2) Nut type (check the best one via lost motion
Step 8
(3) Preload value)
(4) Dynamic load
(1) Surrounding temperature Thermal displacement and target M41 and 4.6 temperature
Step 9
(2) Ballscrew length value of cumulative lead (T) rising effect
(1) Stiffness of screw spindle Pretension force M45
Step 10
(2) Thermal displacement
(1) Max. table speed Motor drive torque and M19 ~ M28
Step 11 (2) Max. rising time motor specification
(3) Ballscrew specification

Table 4.1 Ballscrew selection procedure

4.3 Accuracy Grade of HIIWIN Ballscrews

Precision ground ballscrews are used in applications requiring high positioning accuracy and repeatability,
smooth movement and long service life.Ordinary rolled ballscrews are used for application grade less accurate
but still requiring high efficiency and long service life.Precision grade rolled ballscrews have an accuracy between
that of the ordinary grade rolled ballscrews and the higher grade precision ground ballscrews. They can be used
to replace certain precision ground ballscrews with the same grade in many applications.
HIWIN makes precision grade rolled ballscrew up to C6 grade. Geometric tolerances are different from those
of precision ground screws (See Chapter 6).Since the outside diameter of the screw spindle is not ground, the
 S99TE08-0206 13
set-up procedure for assembling precision rolled ballscrews into the machine is different from that of ground
ones.Chapter 7 contains the entire description of rolled ballscrews.

(1) Accuracy grade

There are numerous applications for ballscrews from high precision grade ballscrews, used in precision
measurement and aerospace equipment, to transport grade ballscrews used in packaging equipment.The
quality and accuracy classifications are described as follows: lead deviation, surface roughness, geometrical
tolerance, backlash, drag torque variation, heat generation and noise level.
HIWIN precision ground ballscrews are classified to 7 classes.In general, HIWIN precision grade
ballscrews are defined by the so called “e300” value see Fig 4.12 and rolled grade ballscrews are defined dif-
ferently as shown in Chapter 7.
Fig. 4.12 is the lead measuring chart according to the accuracy grade of the ballscrews. The same chart
by the DIN system is illustrated in Fig. 4.13.From this diagram, the accuracy grade can be determined by
selecting the suitable tolerance in Table 4.2. Fig. 4.14 shows HIWIN’s measurement result according to the
DIN standard. Table 4.2 shows the accuracy grade of precision grade ballscrews in HIWIN’s specification.The
relative international standard is shown in Table 4.3.
The positioning accuracy of machine tools is selected by ± E value with the e300 variation. The recom-
mended accuracy grade for machine applications is shown in Table 4.5. This is the reference chart for select-
ing the suitable ballscrews in different application fields.

Accuracy Grade 0 1 2 3 4 5 6
e2π 3 4 4 6 8 8 12
e 300 3.5 5 6 8 12 18 23
Item
Thread length
+
_E e +
_E e +
_E e +
_E e +
_E e +
_E e +
_E e
above below
- 315 4 3.5 6 5 6 6 12 8 12 12 23 18 23 23
315 400 5 3.5 7 5 7 6 13 10 13 12 25 20 25 25
400 500 6 4 8 5 8 7 15 10 15 13 27 20 27 26
500 630 6 4 9 6 9 7 16 12 16 14 30 23 30 29
630 800 7 5 10 7 10 8 18 13 18 16 35 25 :0.001mm
Unit 35 31
800 1000 8 6 11 8 11 9 21 15 21 17 40 27 40 35
1000 1250 9 6 13 9 13 10 24 16 24 19 46 30 46 39
1250 1600 11 7 15 10 15 11 29 18 29 22 54 35 54 44
1600 2000 18 11 18 13 35 21 35 25 65 40 65 51
2000 2500 22 13 22 15 41 24 41 29 77 46 77 59
2500 3150 26 15 26 17 50 29 50 34 93 54 93 69
3150 4000 30 18 32 21 60 35 62 41 115 65 115 82
4000 5000 72 41 76 49 140 77 140 99
5000 6300 90 50 100 60 170 93 170 119
6300 8000 110 60 125 75 210 115 210 130
8000 10000 260 140 260 145
10000 12000 320 170 320 180

Table 4.2 HIWIN accuracy grade of precision ballscrew

(2) Axial play (Backlash)

If zero axial play ballscrews (no backlash) are needed, preload should be added and the preload drag
torque is specified for testing purpose. The standard axial play of HIWIN ballscrews is shown in Table 4.4.For
CNC machine tools, lost motion can occur in zero-backlash ballscrews through incorrect stiffness. Please
consult our engineers when determining stiffness and backlash requirements.
14 S99TE08-0206
:0.001mm
Grade 0 1 2 3 4 5 6 7 10
ISO,DIN 6 12 23 52 210
e300 JIS 3.5 5 8 18 50 210
3.5 5 6 8 12 18 23 50 210
Table 4.3 International standard of accuracy grade for ballscrews

:0.001mm

Grade 0 1 2 3 4 5 6
Axial Play 5 5 5 10 15 20 25
Table 4.4 Standard combination of grade and axial play

Tp Target point of accumulated lead.

This value is determined by customers’
different application requirements.
Ep Total reference lead deviation.
Maximum deviation for accumulated ref-
erence lead line over the full length.
e2πp Single lead variation.
Ea Real accumulated reference lead mea-
sured by laser system.
ep Total relative lead deviation.
Maximum deviation of the real accumu-
lated lead from the real accumulated ref-
erence lead in the corresponding range.
e300p Lead deviation over path of 300mm.
The above deviation in random 300 mm
within thread length.

Fig 4.12 HIWIN lead measuring curve of precision ballscrew

eoa (Ea ) : Average lead deviation over useful

path Lu.
A straight line representing the tendency
of the cumulative actual lead.
This is obtained by the least square method
and measured by the laser system. The value
is added by path compensation over the use-
ful path and the mean travel deviation.
c (Tp ) : Path compensation over useful path Lu.
Selection parameter.
This value is determined by customer
and maker as it depends on different applica-
tion requirements.
ep (Ep ) : Mean travel deviation.
Vup (e p ) : Lead variation over useful path Lu.
V300p (e 300p ) : Lead variation over path of 300mm.
V2π p (e 2π p ) : Lead variation over 1 rotation.

Fig 4.13 DIN lead measuring curve of precision ballscrew

 S99TE08-0206 15

eoa (Ea ) :
Lead deviation over useful thread
length relative to the nominal deviation.
(This measurement is made according to
DIN standard 69051-3-1).
C (T)-ep (Ep) ≤ eoa(Ea) ≤ C(T)+ep(Ep)

Vua (e a) :
Total relative lead variation over use-
ful thread length.
(This measurement is made according
to DIN standard 69051-3-2).
Vua(ea) ≤ Vup(ep)

V300a (e 300a ) :
Relative lead variation in random
300mm length within thread length.
(This measurement is made according to
DIN standard 69051-3-3).
V300a(e 300a ) ≤ V300p(e 300p)

V2π a (e 2 π a ) :
Single lead variation over 2π .
(This measurement is made according to
DIN standard 69051-3-4).
V2π a (e 2π a ) ≤ V2π p (e 2π p )

Fig 4.14 Lead accuracy measuring chart from dynamic laser measurement equipment according to DIN 69051 stan-
dard
16 S99TE08-0206

A Accuracy grade
Application grade X
I
S 0 1 2 3 4 5 7 PR1 PR2 PR3

Lathes X
Z
X
Milling machines
Y
Boring machines
Z
X
Machine Center Y
Z
CNC Machinery Tools

X
Jig borers Y
Z
X
Drilling machines
Y
Z
X
Grinders
Y
X
EDM Y
Z
X
Y
Wire cut EDM
U
V
X
Laser Cutting Machine Y
Z
X
Punching Press
Y
Single Purpose Machines
Wood working Machine
Industrial Robot (Precision)
General Machinery

Industrial Robot (General)

Coordinate Measuring Machine
Non-CNC Machine
Transport Equipment
X-Y Table
Linear Actuator
Aircraft Landing Gear
Airfoil Control
Gate Valve
Power steering
Glass Grinder
Surface Grinder
Induction Hardening Machine
Electromachine
Table 4.5 Recommended accuracy grade for machine applications
 S99TE08-0206 17
(3) Geometrical tolerance
It is crucial to select the ballscrew of the correct grade to meet machinery requirements. Table 4.6 and
Fig 4.15 are helpful for you to determine the tolerance factors, which are based on certain required accu-
racy grades.

Fig 4.15 Geometrical tolerance of HIWIN precision ground ballscrew

18 S99TE08-0206

T1: True running deviation of external diameter rela-

,
tive to AA (This measurement is made according
to DIN 69051 and JIS B1192)

Nominal
T1p [ µ m ]
Diameter L5
do [mm] for tolerance class

above up to 0 1 2 3 4 5 6 7
6 12 80
12 25 160
25 50 315 20 20 20 23 25 28 32 40
50 100 630
100 200 1250

T1max[ µ m ]
Lt / do ( for Lt ≥ 4L5)
for tolerance class
above up to 0 1 2 3 4 5 6 7
40 40 40 40 45 50 60 64 80
40 60 60 60 60 70 75 85 96 120
60 80 100 100 100 115 125 140 160 200
80 100 160 160 160 180 200 220 256 320

,
T2: Run out deviation of bearing relative to AA
(This measurement is made according to DIN
69051 and JIS B1192)

Nominal refer- T2p [ µ m ]

Diameter ence ( for L1≤L r)
do [mm] for tolerance class
length
above up to Lr 0 1 2 3 4 5 6 7
6 20 80 6 8 10 11 12 16 20 40
20 50 125 8 10 12 14 16 20 25 50
50 125 200 10 12 16 18 20 26 32 63
125 200 315 - - - 20 25 32 40 80
L1
if L1>L r, then t2a ≤T2p ___
Lr

,
T3: Coaxial deviation relative to AA
(This measurement is made according to DIN
69051 and JIS B1192)

Nominal refer- T3p [ µ m ]

Diameter ence ( for L2 ≥ Lr)
do [mm] for tolerance class
length
above up to Lr 0 1 2 3 4 5 6 7
6 20 80 4 5 5 6 6 7 8 12
20 50 125 5 6 6 7 8 9 10 16
50 125 200 6 7 8 9 10 11 12 20
125 200 315 - - - 10 12 14 16 25
L2
if L2 >L r, then t3a ≤T3p ___
Lr
 S99TE08-0206 19

T4 : Run-out deviation of bearing end shoulder rela-

,
tive to AA (This measurement is made accord-
ing to DIN 69051 and JIS B1192)

Nominal
T4p [ µ m ]
Diameter
do [mm] for tolerance class

above up to 0 1 2 3 4 5 6 7
6 63 3 3 3 4 4 5 5 6
63 125 3 4 4 5 5 6 6 8
125 200 - - - 6 6 8 8 10

T5 : Face running deviation of locating face (only

,
for nut) relative to BB (This measurement is
made according to DIN 69051 and JIS B1192)
Nut Flange T5p [ µ m ]
Diameter
D [mm] for tolerance class
above up to 0 1 2 3 4 5 6 7
- 20 5 6 7 8 9 10 12 14
20 32 5 6 7 8 9 10 12 14
32 50 6 7 8 8 10 11 15 18
50 80 7 8 9 10 12 13 16 18
80 125 7 9 10 12 14 15 18 20
125 160 8 10 11 13 15 17 19 20
160 200 - 11 12 14 16 18 22 25
200 250 - 12 14 15 18 20 25 30

T6 : Run-out deviation of external diameter (only

,
for nut) relative to BB (This measurement is
made according to DIN 69051 and JIS B1192)
Nut Flange T6p [ µ m ]
Diameter
D [mm] for tolerance class

above up to 0 1 2 3 4 5 6 7
- 20 5 6 7 9 10 12 16 20
20 32 6 7 8 10 11 12 16 20
32 50 7 8 10 12 14 15 20 25
50 80 8 10 12 15 17 19 25 30
80 125 9 12 16 20 24 22 25 40
125 160 10 13 17 22 25 28 32 40
160 200 - 16 20 22 25 28 32 40
200 250 - 17 20 22 25 28 32 40

T7 : Deviation of parallelism (only for nut) relative to

,
BB (This measurement is made according to
DIN 69051 and JIS B1192 )
Mounting T7p [ µ m ] / 100mm
basic lenght
(mm) Lr for tolerance class
above up to 0 1 2 3 4 5 6 7
- 50 5 6 7 8 9 10 14 17
50 100 7 8 9 10 12 13 15 17
100 200 - 10 11 13 15 17 24 30

Table 4.6 Tolerance table and measurement method for HIWIN precision ballscrews
20 S99TE08-0206
4.4 Preload Methods
The specially designed Gothic ball track can make the ball contact angle around 45˚. The axial force Fa,
which comes from an outside drive force or inside preload force, causes two kinds of backlash. One is the
normal backlash, Sa, caused by the manufacturing clearance between ball track and ball. The other is the
deflection backlash, , caused by the normal force Fn which is perpendicular to the contact point.
The clearance backlash can be eliminated by the use of an preload internal force P. This preload can be
obtained via a double nut, an offset pitch single nut, or by adjusting the ball size for preloaded single nuts
(Fig. 5.7~ Fig. 5.8).
The deflection backlash is caused by the preload internal force and the external loading force and is
related to that of the effect of lost motion.

Fig 4.16 Gothic form profile and preloading relation

(1) Double nut preloading

Preload is obtained by inserting a spacer between the 2 nuts (Fig. 4.17). “Tension preload” is created
by inserting an oversize spacer and effectively pushing the nuts apart. “Compression preload” is created
by inserting an undersize spacer and correspondingly pulling nuts together. Tension preload is primarily
used for precision ballscrews. However, compression preload type ballscrews are also available upon
your request. If pretension is necessary to increase stiffness, please contact us for the amount of preten-
sion to be used in the ballscrew journal ends. (0.02mm to 0.03mm per meter is recommended, but the T
value should be selected according to the compensation purpose).

Fig 4.17 Preload by spacer

 S99TE08-0206 21
(2) Single nut preloading
There are two ways of preloading a single nut. One is called “the oversized-ball preloading method”.
The method is to insert balls slightly larger than the ball groove space (oversized balls) to allow balls to
contact at four points (Fig. 4.18).
The other way is called “The offset pitch preloading method” as shown in Fig. 4.19.
The nut is ground to have a δ value offset on the center pitch. This method is used to replace the tra-
ditional double nut preloading method and has the benefit of a compact single nut with high stiffness via
small preload force. However, it should not be used in heavy duty preloading. The best preload force is
below 5% of dynamic load (C).

Fig 4.18 Preload by ball size Fig 4.19 Offset type preloading

(3) Preload calculation

Kp is between 0.1 and 0.3.
Fbm
p= M1 η1,η2 are the mechanical efficiencies of the
2.8
ballscrew.
P :preload force(kgf) (1) for common transmission (to convert rotary
Fbm:Mean operating load(kgf) motion to linear motion)
(Ref. M8~10)

M3

(2) for reverse transmission (to convert linear

rotary motion to rotary motion)

M4

M5

Fig. 4.20 : Preload drag torque measuring ethod

(according to JIS B1192) β = tan −1 µ M6

α : lead angle (degrees)

M2 Dm: pitch circle diameter of screw shaft (mm)
Preload drag torque(Fig. 5.9) : lead (mm)
Td : preload drag torque(kgf -mm) ß : friction angle (0.17 ˚~0.57 ˚)
P : preload (kgf ) µ : friction coefficient (0.003~0.01)
: lead (mm)
Kp: preload torque coefficient
Kp= 1 -η2
η1
22 S99TE08-0206
(4) Uniformity of preload drag torque
(1) Measuring method (2) Measuring conditions
Preload creates drag torque between the nut 1. Without wiper.
and screw. It is measured by rotating the screw 2. The rotating speed, 100rpm.
spindle at constant speed while restraining the 3. The dynamic viscosity of lubricant, 61.2
nut with a special fixture as shown in Fig. 4.20. ~74.8 cSt (mm/s) 40 , that is , ISO VG
The load cell reading force Fp is used to calcu- 68 or JIS K2001.
late the preload drag torque of the ballscrew. 4. The return tube up.
HIWIN has developed a computerized drag
torque measuring machine which can accurately (3) The measurement result is illustrated by the
monitor the drag torque during screw rotation. standard drag torque chart. Its nomenclature
Therefore, the drag torque can be adjusted to is shown in Fig. 4.21.
meet customer requirements (Fig. 2.5). The
measurement standard for preload drag torque is (4) The allowable preload drag torque variation
shown in Fig. 4.21 and Table 4.7. as a function of accuracy grade is shown in
Table 4.7.

Unit: ± %

(1) Useful stroke length of thread (mm)

Basic 4000 mm maximum over 4000 mm
Dragtorque Slender ratio < 40 Slender ratio < 60
(kgf - cm) Accuracy grade Accuracy grade Accuracy grade
Above Up To 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
2 4 30 35 40 40 45 55 60 - 40 40 50 50 60 60 70 - - - - - - - - -
4 6 25 30 35 35 40 40 50 - 35 35 40 40 45 45 60 - - - - - - - - -
6 10 20 25 30 30 35 35 40 40 30 30 35 35 40 40 45 45 - - - 40 43 45 50 50
10 25 15 20 25 25 30 30 35 35 25 25 30 30 35 35 40 40 - - - 35 38 40 45 45
25 63 10 15 20 20 25 25 30 30 20 20 25 25 30 30 35 35 - - - 30 33 35 40 40
63 100 - 15 15 15 20 20 25 30 - - 20 20 25 25 30 35 - - - 25 23 30 35 35
Table 4.7 : Variation range for preload drag torque. (According to JIS B1192)
Note : 1. Slender ratio=Thread length of spindle/ Nominal spindle O.D.(mm)
2. Refer to the designing section of the manual to determine the basic preload drag torque.
3. Table 4.10 shows the conversion table for Nm.
4. For more information, please contact our engineering department.

(a) : basic drag torque.

(b) : Variation of basic preload drag
torque.
(c) : Actual torque.
(d) : Mean actual preload drag
torque.
(e) : Variation value of actual pre-
load drag torque.
(f) : Starting actual torque.
Lu : Useful travelling distance of nut

Fig 4.21 Nomenclature of drag torque measurement

 S99TE08-0206 23
4.5 Calculation Formulas
Service life
The average number of rpm, nav

t1 t t
nav = n1 × + n2 × 2 + n3 × 3 + ..... M7
100 100 100

nav = average speed (rpm)

n : speed (rpm)

=% of time at speed n1 etc.

The average operating load Fbm

(1) With variable load and constant speed

M8

Fbm = average operating load (kgf)

fp : operation condition factor
fp = 1.1 ~ 1.2 when running without impact
1.3 ~ 1.8 when running in the normal condition
2.0 ~ 3.0 when running with heavy impact and vibration

(2)With variable load and variable speed

M9

(3)With linear variable load and constant speed

M10

Fig 4.22 Equivalent speed

24 S99TE08-0206
Example 4.5 - 1
A HIWIN ballscrew is subjected to the following operating conditions. Calculate the average running
speed and operating load.

Operating Condition :
For smooth running without impact fp = 1.1

Condition Axial load(kgf) Revolution(rpm) Loading time ratio (%)

(Fb) (n) (t)
1 100 1000 45
2 400 50 35
3 800 100 20
Calculation

45 35 20
nav = 1000 × + 50 × + 100 × = 487.5rpm (ref.M7)
100 100 100

The resultant axial force, Fa

For a single nut without preload

Fa = Fbm M11

For a single nut with preload P

Fa ≤ Fbm + P M12

Expected service life for applications

Table 4.8 shows the recommended service life for general applications by service distance.
In the right of Table 4.8 is the formula for service life in hours.
Shock load, vibration, temperature, lubrication, position deviation, etc. must be taken into account also.

For single nut

Service life represented in revolutions :

M13
M14

L : Service life in running revolution (revolutions)

L : Service life in running revolution (revolutions)
C : dynamic load rating (kgf) (106 revs)
P : Preload force (kgf)

For symmetrical preload double nut arrangement

(b) conversion from revolutions to hours :
(a) Service life represented in revolutions :
L
Lh = M15
nav × 60

Lh : Service life in hours (hours)

Fbm (2) = Fbm (1) − Fbm
nav: average speed (rpm, Ref. M7)
 S99TE08-0206 25
(c) Conversion from travel distance to hours :

M16

Running life calculation (in hours)

Lh : Running life (in hours)

Ld : Running life (in distance, Km)
: Ballscrew lead (mm per rev)
nav :Average running speed (rpm)

Machine Type Service Life in Distance,

Ld (km)
Machine Tools 250
General Machinery 100~250
Control Mechanisms 350
Measuring Equipment 210
Aircraft Equipment 30

Table 4.8 Typical design service life for general application

(The above service life is calculated by the dynamic load rating for 90% reliability.

(d) the modified service life for different reliability factors is calculated by

Lm = L × fr M17

Lhm = Lh × fr M18

with the reliability factor fr ( Table 4.9)

Reliability
fr
%
90 1
95 0.62
96 0.53
97 0.44
98 0.33
99 0.21
Table 4.9 Reliability factor for service life.
26 S99TE08-0206
Example 4.5 - 2
By the example 5.4-1, if the design service life of the ballscrew is 3500 hours, lead = 10mm, single
nut with zero backlash, find the nominal diameter of the HIWIN ballscrew.
Calculation

(Assume zero backlash when Fbm = 318.5 kgf)

(Ref formula M1)

(revolutions)

C’≤C rating

So, from the dimensions table of HIWIN ballscrews, select FSV type nut with spindle nominal diame-
ters equals 32mm and C1 circuits which can satisfy this application.

Example 4.5 - 3
If the ballscrew nominal diameter = 50mm, lead = 8mm, and service life L = 7x106 revolutions, find
the permissible load on the screw spindle.
Calculation
From the dimensions table of HIWIN ballscrew, the FSV type ballscrew with nominal diameter
= 50 mm, lead = 8 mm and B3 type return tube has the dynamic load rating C = 5674.

Drive torque and drive power for the motor

(Friction force + operation force)

Gear 2

Motor
Ballscrew

Gear 1

Fig 4.23 Load operation by ballscrew

 S99TE08-0206 27
Fig. 4.23 shows the terms for a feed system (c) Motor drive torque
operated by ballscrew. The formula for motor drive For normal operation :
torque is given below : N1
(a) Common transmission (to convert rotary TM = (Ta + Tb + Td ) × M21
N2
motion to linear motion)
TM = Motor drive torque (kgf-mm)
M19 Tb = Friction torque of supporting bearing (kgf-
mm)
Ta = Drive torque for common transmission (kgf- Td = Preload drag torque (kgf-mm, Ref. M2)
mm) N1 = Number of teeth for driver gear
Fb = Axial load (kgf) N2 = number of teeth for driven gear
Fb = Fbm+µxW(for horizontal motion) For acceleration operation :
= Lead (mm)
M22
η1 = Mechanical efficiency (0.85 ~ 0.95, Ref.
M3) : Motor drive torque during acceleration
W = Table wight + Work piece weight (kgf) (kgf-mm)
µ= Friction coefficient of table guide way (0.005 J : System inertia (kgf-mm-sec2)
~ 0.02) a : Angular acceleration (rad/sec2)

(b) Reverse transmission (to convert linear M23

motion to rotary motion)
Ndif = rpmstage2 -rpmstage1
M20 ta : acceleration rising time. (sec)

η2 = Mechanical efficiency (0.75 ~ 0.85, Ref.

M4)
Tc = Torque for reverse transmission (kgf-mm)

Where

= Motor inertia + Equivalent gear inertia + Ballscrew inertia + Load inertia

(Fig.4.23) M24

Ws : Ballscrew weight (kgf)

DN : Ballscrew nominal diameter (mm)
g : Gravity coefficient (9800 mm/sec2 )
JM : Inertia of motor (kgf-mm-sec2)
JG1 : Inertia of driver gear (kgf-mm-sec2)
JG2 : Inertia of driver gear (kgf-mm-sec2)
28 S99TE08-0206
Total operating torque : (d) Drive power

Tp max × Nmax
M25 Pd = M27
974
TMa : Total operating torque (kgf-mm) Pd : Maximum drive power (watt) safety
The inertia of a disc is calculated as following : Tpmax : Maximum drive torque
For disc with concentric O.D. (safety factor x TMa ,kgf-mm)
1 Nmax : Maximum rotation speed (rpm)
J= πρd R 4 L M26
2g
(e) Check the acceleration time
J : Disc inertia (kgf • mm • sec2 )
ρd : Disc specific weight (7.8x10-6 kgf/mm3 ) for J 2πNmax
ta = × ⋅f M28
steel TM1 − TL 60
R: Disc radius (mm) ta = Acceleration rising time
L : Disc length (mm) J= Total inertia moment
g : Gravity coefficient (9800 mm/sec2 ) TM1 = 2 x TMr
TMr = Motor rated torque
TL= Drive torque at rated feed
f = Safety factor = 1.5

Table 4.10 : Shows the conversion relationship of different measurement units for the motor torque or
preload drag torque.

kgf -cm kgf mm Nm kpm (kgf -m) OZ-in ft- bf

9.8×10
−2
10 7.23301×10
−2 −2
1 10 13.8874
9.8×10
−3
1.0×10 7.23301×10
−3 −3
0.1 1 1.38874
1.019716×10 1.41612×10
2 2
10.19716 1 0.1019716 0.737562
10 10 1.38874×10
2 3 3
9.80665 1 7.23301
7.20077×10 7.06155×10 7.20077×10 5.20833×10
−2 −3 −4 3
0.720077 1
13.82548 1.382548×10 2
1.35582 0.1382548 1.92×10
2
1
Table 4.10 Conversion table for motor torque.

Example 4.5 - 4
Consider the machining process driven by the motor and ballscrew as Fig. 4.24.
Table weight W1 = 200 kgf
Work weight W2 = 100 kgf
Friction coefficient of slider µ = 0.02

Operating condition : Smooth running without impact.

Axial feed force Revolution Loading time ratio(%)
(kgf) (rpm) (t)
100 500 20
300 100 50
500 50 30
 S99TE08-0206 29
Acceleration speed :100 rad/sec2
Motor Condition :Motor diameter : 50 mm, Motor length : 200 mm,
Gear condition :Driver gear diameter G1 : 80 mm, Thickness : 20 mm, Teeth : 30
Driven gear diameter G2 : 240 mm, Thickness : 20 mm, Teeth : 90

Ballscrew condition :
Nominal diameter : 50 mm, Pitch : 10 mm
Length : 1200 mm, Weight : 18 kgf
No backlash when axial feed force = 300 kgf
Bearing torque Tb = 10 kgf-mm
Mechanical efficiency η1 = 0.80

W2
F

G2

W1

motor G1

Fig 4.24 Milling process in the machine

Calculation
(1) Motor drive torque in normal rating condition :

(Ref. M7)

(Ref. M9)

(axial feed force = 300 kgf ,Ref. M1)

(Ref. M19)

(Ref. M2)

(Ref. M21)

(2) Motor torque in acceleration operation :

(I) Inertia of motor

M
30 S99TE08-0206
(II) Inertia of gear

(III) Inertia of ballscrew

(IV) Inertia of load

(V) Total inertia

(3) Total motor torque

(4) Drive power

(safety factor = 2)

(5) Selection motor

Select the DC motor rated torque : TMr>1.5TM ,
and maximum motor torque : TMax>1.5Tpmax
Thus the DC servo motor with following specification can be chosen.
Rated output : 950 W
Rated torque : 30 kgf-cm (300 kgf • mm)
Rated rotational speed : 2000 rpm
Maximum torque : 65 kgf • cm (650 kgf • mm)
Moment of inertia of motor : 0.20 kgf • mm • sec2

(6) Check the acceleration time

 S99TE08-0206 31
Buckling load

M29

Fp = 0.5 Fk M30
Fk = Permissible load (kgf) fixed - fixed Nf = 1.0
Fp = Maximum permissible load (kgf) fixed - supported Nf = 0.5
dr : Root diameter of screw shaft (mm) supported - supported Nf = 0.25
Lt : distance between support bearing (mm) Fixed - free Nf = 0.0625
Nf : Factor for different mounting types *1kgf = 9.8N;1daN=10N
The buckling load diagram for different spindle diameter and support method is shown in Fig 4.25.

Critical speed
M f dr
Nc = 2.71 × 108 × M31
L2t

N p = 0.8 Nc M32

Nc = critical speed (rpm) fixed - fixed Mf = 1

Np = Maximum permissible load (rpm) fixed - supported Mf = 0.692
dr : Root diameter of screw shaft (mm) supported - supported Mf = 0.446
Lt : distance between support bearing (mm) Fixed - free Mf = 0.147
Mf : Factor for different mounting types
The critical speed for different spindle and support method is shown in (Fig 4.26).

Fig 4.25 Shows the buckling load for different Fig 4.26 shows the critical speed for different
screw spindle diameter and length screw spindle diameter and length
32 S99TE08-0206
Dm-N value for ballscrew surface speed
Dm-N value has a strong influence over ballscrew noise, working temperature and service life of return sys-
tem.
For HIWIN ballscrew,
M33

Dm : Pitch circle diameter (mm)

N : Maximum speed (rpm)
Ballscrew structure enhancement designed by HIWIN when Dm-N value ranges from 70,000 to 150,000 . If
Dm-N value above 150,000 , please consult our company.

Stiffness
Stiffness is an indication of the rigidity of a machine. The stiffness of the ballscrew is determined by nut-spin-
dle rigidity via axial load, balltrack contact rigidity and screw spindle rigidity. When assembling the ballscrew in
the machine, the stiffness of support bearing, mounting condition of nut with machine table etc. also should be
considered. Fig 4.27 shows the relation of total stiffness of the machine feed system.
From testing, the stiffness of nut-spindle relation and ball and balltrack relation can be combined into the stiff-
ness of nut, Kn, and listed in dimension table of different nut type. The stiffness of the ballscrew is shown as :

M34

The stiffness of the screw spindle is shown as :

M35

M36

The stiffness chart is shown in Fig 4.28

K : Total stiffness of ballscrew (kgf/µm)

M37

Db : Diameter of ball (mm)

Ks : Screw spindle stiffness (kgf/µm)
Kn : Nut stiffness (kgf/µm)
The stiffness of the nut is tested using an axial force equal to the highest possible preload of 10% dynamic
load (C) and is shown in the dimension table of each nut. When the preload is less than this value, the stiffness of
the nut is calculated by extrapolation method as :

M38

kn : Stiffness of nut
K : Stiffness in the dimension table
P : Preload
C : dynamic load on dimension table (106 rev)
Since the offest pitch type preloading method is single nut instead of double nut, it has a good stiffness with
a small preload force. The preload of the offset type nut is calculated by 5% of the dynamic load by formula :

M39
 S99TE08-0206 33
Single nut with backlash is calculated when the external axial force is equal to 0.28 C, thus :

M40

The axial stiffness of the whole feed system includes the stiffness of support bearings and nut mount-
ing table. The designer should consider the total stiffness carefully.

Kt

Ks
Ktot K
Knb
Kn
Knr
Kb

Ktot : Total stiffness of machine feed system.

Kt: Table mounting stiffness
Kb: Support bearing stiffness.
K: Ballscrew stiffness.
Ks: Ballscrew spindle stiffness.
Kn: Ballscrew nut stiffness
Knb: Ball and balltrack stiffness.
Knr: Nut-spindle stiffness by radial load

Fig 4.27 Stiffness distribution for ballscrew feed Fig 4.28 Stiffness chart for ballscrew spindle
system

Thermal expansion

M41

L : Thermal expansion of screw spindle (mm)

T : ( ) Temperature rise at screw spindle
Ls : Total length of screw spindle (mm)
The T value should be chosen to compensate for the temperature rise of the ballscrew.
HIWIN recommends a T value of -0.02 ~ -0.03 per meter for CNC machine tools.

Basic dynamic axial load rating C (theoretical)

The dynamic load is the load at which 90% of the ballscrews will achieve the service life of 1 x 106 rev
(C). The reliability factor can be adjusted by Table 4.9. The dynamic load is shown on the dimension table of
each nut type.

Basic static axial load rating Co (theoretical)

The static load is the load which will cause the balltrack to have a plastic deformation exceeding 0.0001x
ball diameter. To calculate the maximum static load of a ballscrew, the static safety factor Sf of the application
condition should be considered.

M42
Sf : Static factor = 2.5 max
Co : Static load from the dimension table of the nut type.
Fa(max) : Maximum static axial load.
34 S99TE08-0206
Example 4.5 - 5
Ballscrew specification :
1 R40-10B2-FSW-1000-1200-0.012 Lead = 10 mm
Pitch circle diameter Dm = 41.4 mm Turns = 2.5x2
Ball diameter : 6.35 mm Lead angle α= 4.4˚
Root diameter dr = 34.91 mm Friction angle β= 0.286˚
Column load : fixed - supported Preload P = 250 kgf
Critical speed : fixed - supported Mean axial force Fb = 700 kgf
Stiffness of bearing Kb = 105 kgf/ µm Nf = 0.5 ; Lt = 1000 mm ; Mf = 0.692
Calculation
1. Buckling load Fp

2. Critical speed Np

3. Mechanical efficiency η (theoretical)

(I) Common transmission

(II)Reverse transmission

4. Stiffness K

5. Lost motion during axial force Fb=700kgf

(each way) Total lost motion δ =2x0.056=0.112 mm

If the preload increases to 2x250=500 kgf then Kn=58 kgf/µm and K=15.1 kgf/µm. Total stiffness Kt=13.2
kgf/µm and total lost motion d=0.106 mm. The difference is only 6µm (5% change). comparing with 250 kgf,
preloaded nut, but the temperature rise caused by 500kgf preload is heavy. The spindle stiffness is some-
times more important than the nut stiffness. The best way to increase the stiffness of the system is not in the
heavy preloading of the ballscrew nut. If the support method changes to fixed-fixed, then Ks=82 kgf/µm and Kt
becomes 23 kgf/µm. The total lost motion δ=0.061 mm. The difference is 51µm (45%).
 S99TE08-0206 35
Material specification
Table 4.11 shows the general material used for HIWIN ballscrew. HIWIN also makes ballscrew from stain-
less steel. Please contact us if you have special requirements.

Manufacturing range
The maximum length to which a ballscrew can be manufactured depends on spindle diameter and accu-
racy grade (Table 4.12). Since high accuracy ballscrews require a high degree of straightness to the screw
spindle, the higher the slender ratio (length/diameter), the more difficult to manufacture and the less the spin-
dle stiffness.
HIWIN recommends the maximum lengths shown in Table 4.12.
If a longer length is required, please contact with HIWIN engineer.

Steel specification
Item
BSI DIN AISI JIS
EN43C 1.1213 1055 S55C
Spindle 1.7225 4140 SCM440H
EN19C 1.7228 4150 SCM445H
EN34 3310 SNCM 220(21)
Nut 1.6523 SCM420H
EN36 8620 SCM415H
Ball EN31 1.3505 52100 SUJ2
Table 4.11 Material Specifications
Unit:mm
Total O.D.
length 8 10 12 16 20 25 28 32 36 40 45 50 55 63 70 80 100
GRADE
C0 170 300 400 600 700 1000 1000 1200 1300 1500 1600 1800 2000 2000 2000 2000 2000

C1 170 400 500 720 950 1300 1500 1800 1800 2300 2500 3100 3500 4000 4000 4000 4000

C2 170 500 630 900 1300 1700 1800 2200 2200 2900 3200 4900 5000 5200 5500 6300 6300

C3 200 500 630 1000 1400 1800 2000 2500 3200 3500 4000 4500 5000 6000 7100 10000 10000
C4 250 500 630 1000 1400 1800 2000 2500 3200 3500 4000 4500 5000 6000 7100 10000 10000
C5 250 500 630 2400 1700 2400 2500 3000 3200 3800 4000 5000 5500 6900 7100 10000 10000
C6 1200 1200 1200 1500 1800 2500 3000 3000 4000 4000 4000 5600 5600 6900 7100 10000 10000
C7 1200 1200 1200 3000 3000 4000 4000 4500 4500 5600 5600 5600 5600 6900 7100 10000 10000

Table 4.12 General manufacturing range of HIWIN screw spindle vs. diameter and accuracy grade
Please consult with HIWIN in this area.
36 S99TE08-0206
Heat treatment
HIWIN’s homogenous heat treatment technique gives the ballscrew maximum life capability. Table
4.13 shows the hardness value of hardness in each component of HIWIN ballscrews. The surface hard-
ness of the ballscrew affects both dynamic and static load value. The dynamic and static values shown in
the dimension table are the values for a surface hardness equal to HRC 60. If the surface hardness is low-
er than this value, the following formula will give you the calibration result.

M43

M44

Where fH and fHO are the hardness factor.

C’o: Calibrated static load
Co : Static load
C’ : Calibrated dynamic load
C : Dynamic load

Item Treat Method Hardness (HRC)

Spindle Induction Hardening 58-62
Carburizing
Nut 58-62
or Induction Hardening
Ball 62-66

Table 4.13 Hardness of each component of HIWIN ballscrew

4.6 Temperature Rise Effect on Ballscrews

The temperature rise of ballscrew during the Preloading the nut will increase the friction
working period will influence the accuracy of the torque of the screw, making it more sensitive to an
machine feed system, especially in a machine increase in temperature during working period.
designed for high speed and high accuracy. HIWIN recommends using a preload of 8% of
The following factors have the effect of raising the dynamic load for medium and heavy preload, 6%
the temperature in a ballscrew. ~ 8% for medium preload, 4% ~ 6% for light and
(1) Preload medium and below 4% for light preload.
(2) Lubrication The heaviest preload should not exceed 10% of
(3) Pretension the dynamic load for best service life and a low tem-
Fig 4.29 shows the relation of working speed, perature rise effect.
preload and temperature rise. Fig 4.30 shows the
relation of nut temperature rise to preload friction (2) Lubrication effect
torque. From Fig 4.29,Fig 4.30 and example 4.5-5, The selection of lubricant will directly influence
doubling the preload of the nut will increase the tem- the temperature rise of the ballscrew.
perature about 5 degrees, but the stiffness increase HIWIN ballscrews require appropriate lubrica-
only by about 5% (few µm). tion either by greasing or oiling. Antifriction bearing
oil is recommended for ballscrew oil lubrication.
(1) Preload effect Lithium soap based grease is recommended for
To avoid any lost motion in the machine feed ballscrew greasing. The basic oil viscosity require-
system, increasing the rigidity of the ballscrew nut is ment depends on the speed, working temperature
important. However, to increase the rigidity of the and load condition of the application.
ballscrew nut, it is necessary to preload the nut to a (Fig 4.31) shows the relation of oil viscostity,
certain level. working speed and rise in temperature.
 S99TE08-0206 37

Fig 4.31 The influence of oil viscosity on the fric-

tion torque
: 1500 rpm with 200 kgf preload
When the working speed is higher and the work-
: 1500 rpm with 100 kgf preload
: 500 rpm with 200 kgf preload ing load is lower, a low viscosity oil is better. When
: 500 rpm with 100 kgf preload the working speed is lower and the working load is
heavy, a high viscosity oil is preferred.
Fig 4.29 The relation of working speed, preload Generally speaking, oil with a viscosity of 32 ~
and temperature rise 68 cSt at 40 (ISO VG 32-68) is recommended for
high speed lubrication (DIN 51519) and viscosity
above 90 cSt at 40 (ISO VG 90) is recommended
for low speed lubrication.
In high speed and heavy load applications the
use of a forced coolant is necessary to lessen the
temperature. The forced lubrication of coolant can
be done by a hollow ballscrew.
Fig 4.32 shows the comparsion of a ballscrew
applied with coolant and without coolant.

Fig 4.30 The relation of nut temperature rise to

preload friction torque

Fig 4.32 Ballscrew temperature rise with the coolant

and without the coolant
38 S99TE08-0206
Fig 4.33 shows a typical application for hollow mends using pretension when the temperature rise
ballscrew in machine tools. The inspection and is below 5 . Also, if the diameter of the screw spin-
replenishing of the ballscrew lubricant is listed in dle is greater than 50mm, it is not suitable for preten-
Table 4.14. sion.A large spindle diameter requires a high preten-
sion force, causing burn down of the supporting
(3) Pretension effect bearing.
When the temperature rises in the ballscrew, the HIWIN recommends a T compensation value of
effect of thermal stress will elongate the screw spin- about 3˚, (about -0.02 ~ 0.03 for each 1000 mm
dle. It can make the spindle length unstable. screw spindle).
The elongating relationship can be calculated Since different applications require different T
according to M41. This elongation can be compen- values, please contact HIWIN engineer.
sated via the pretension force. For the purpose of The pretension force is calculated as :
pretension, there is a negative T value indicated in
Pf = Ks × ∆L M45
the design drawing to compensate the pretension
value. Ks : Stiffness of screw spindle (kgf/µm)
Since a large pretension force will cause the Pf : Pretension force (kgf)
burn down of the supporting bearing, HIWIN recom-
L : Pretension value (µm)

Lubrication Method Inspection & Replenishment Guide

• Check the oil level and clean the contamination once a week.
Oil
• When contamination happens, replacing the oil is recommended.

• Inspect for contamination of chips every 2 or 3 months.

Grease • If contamination happens, remove old grease and replace with new grease.
• Replace grease once a year.

Table 4.14 : Inspection and replenishment of Lubricant

Fig 4.33 High accuracy machine tools with hollow ballscrew lubrication.