Mechanical seals • Mechanical seals for pumps • Pusher seals

M2N
l8

b l5 l6

1.4 1.3 1.2 1.1 2 3

H11

H8
min

h6
d3

d7
d6
d1
db
20°
Features d8
• For plain shafts l3 / l 21 1) l9 5
• Single seal
l 1N l 7 +0,5
• Unbalanced
• Conical spring rotating
• Dependent on direction of rotation
Item Part no. to DIN 24250 Description EagleBurgmann © M2N
Product variants
Advantages 1.1 472 Seal face M2
• Economical seal solution 1.2 412.1 O-Ring Rotating unit M2 with seat G4
• No damage of the shaft by set screws 1.3 474 Thrust ring or G16 (shorter installation length)
• Short installation length possible (G16) 1.4 478 Righthand spring Seal face: Carbon graphite resin impregnated (B)
1.4 479 Lefthand spring Seat G4: Silicon carbide (Q1),
Operating range (see note on page 1) 2 475 Seat (G9) Special cast CrMo steel (S)
Shaft diameter: d1 = 6 … 38 mm (0.25" … 1.5") 3 412.2 O-Ring Seat G16: Silicon carbide (Q1, Q2),
Pressure: p1 = 10 bar (145 PSI) Special cast CrMo steel (S), Aluminium oxide (V)
Temperature: t = –20 °C … +140 °C (–4 °F … +355 °F)
Sliding velocity: vg = 15 m/s (50 ft/s) M2N4
Axial movement: ±1.0 mm Rotating unit M2 with seat G6
Seal face: Carbon graphite resin impregnated (B)
Materials Seat G6: Silicon carbide (Q1),
Seal face: Carbon graphite resin impregnated (B) Special cast CrMo steel (S)
Seat G9: Silicon carbide (Q1, Q2),
Special cast CrMo steel (S), Aluminium oxide (V)
Seat alternatives
Standards and approvals
l6 l 16 l6
• EN 12756 l5 l5
l 15
20°

30°

20°

Recommended applications
• Water and waste water technology
• Building services industry l 28 l 19
l 14
+0,1

H11
H7

H8

l 10 l 12 l 18
d 6 H11

d 7 H8

d 11

d 12

d7
d6

G6 (EN 12756) G4
G4 © EagleBurgmann G16G16 (EN 12756)
© EagleBurgmann
G6 (M32N4) © EagleBurgmann

6
 Dimensions in mm
d1 d3 d6 d7 d8 d11 d12 db l1N l 31) l5 l6 l7 l8 l10 l12 l14 l15 l16 l18 l19 l211) l28 b R
6 15 – – – 11.8 16 8 – – – – – – – 6.5 5.6 1.2 3.8 – – 10.9 – – 1.2
8 18 – – – 15.5 19.2 11 – – – – – – – 8 7 1.2 3.8 – – 15.5 – – 1.2
10 20 17 21 3 15.5 19.2 13 40 17.5 1.5 4 8.5 17.5 7.5 7.5 6.6 1.2 3.8 – – 15.9 6.6 8 1.2
12 22 19 23 3 17.5 21.6 16 40 17.5 1.5 4 8.5 17.5 7.5 8 7 1.2 3.8 – – 16 6.6 8 1.2
14 25 21 25 3 20.5 24.6 18 40 17.5 1.5 4 8.5 17.5 7.5 8 7 1.2 3.8 – – 16 6.6 8 1.2
15 27 – – – 20.5 24.6 19 – – – – – – – 7.5 6.6 1.2 3.8 – – 17.4 – – 1.2
16 27 23 27 3 22 28 21 40 19.5 1.5 4 8.5 17.5 7.5 8.5 7.5 1.5 5 – – 19 6.6 8 1.5
18 30 27 33 3 24 30 23 45 20.5 2 5 9 19.5 8.5 9 8 1.5 5 15 7 20.5 7.5 8 1.5
20 32 29 35 3 29.5 35 26 45 22 2 5 9 19.5 8.5 8.5 7.5 1.5 5 15 7 22 7.5 8 1.5
22 35 31 37 3 29.5 35 28 45 23.5 2 5 9 19.5 8.5 8.5 7.5 1.5 5 15 7 23.5 7.5 8 1.5
24 38 33 39 3 32 38 30 50 25 2 5 9 19.5 8.5 8.5 7.5 1.5 5 15 7 25 7.5 8 1.5
25 40 34 40 3 32 38 31 50 26.5 2 5 9 19.5 8.5 8.5 7.5 1.5 5 15 7 26.5 7.5 8 1.5
26 41 – – – 34 40 32 – – – – – – – 9 8 1.5 5 – – 26.5 – – 1.5
28 43 37 43 3 36 42 35 50 26.5 2 5 9 19.5 8.5 10 9 1.5 5 15 7 26.5 7.5 8 1.5
30 47 – – – 39.2 45 37 – – – – – – – 11.5 10.5 1.5 5 15 7 25 – – 1.5
32 48 – – – 42.2 48 39 – – – – – – – 13 10.5 1.5 5 15 7 28.5 – – 1.5
35 53 – – – 46.2 52 43 – – – – – – – 13.5 11 1.5 5 15 7 28.5 – – 1.5
38 56 – – – 49.2 55 47 – – – – – – – 13 10.3 1.5 5 16 8 32 – – 1.5
1) l valid for M2N, l21 valid for M2
3

7
Mechanical seals • Mechanical seals for pumps • Pusher seals

M3N
l8
b l5 l6

1.4 1.3 1.2 1.1 2 3

min

H11
h6

H8
d3

db
d1

d6

d7
20°
Features d8
3)
• For plain shafts l 3 / l 21 l9 5
• Single seal l 1N l 7 +0,5
• Unbalanced
• Rotating conical spring
• Dependent on direction of rotation
Item Part no. to DIN 24250 DescriptionEagleBurgmann © M3N
Advantages 1.1 472 Seal face
• Universal application opportunities 1.2 412.1 O-Ring
• Insensitive to low solids content 1.3 474 Thrust ring
• No damage of the shaft by set screws 1.4 478 Righthand spring
• Large choice of materials 1.4 479 Lefthand spring
• Short installation lengths possible (G16) 2 475 Seat (G9)
• Variants with shrink-fitted seal face available 3 412.2 O-Ring
Operating range (see note on page 1) Product variants
Shaft diameter: d1 = 6 … 80 mm (0.24" … 3.15") M3 M37G
Pressure: p1 = 10 bar (145 PSI) Items and description as M3N. Items and descriptions as for type M3N, but with shrink
Temperature: t = –20 °C … +140 °C (–4 °F … +355 °F) Seal face: Special cast CrMo steel (S) fitted silicon carbide seal face to the seal face carrier
Sliding velocity: vg = 15 m/s (50 ft/s) Seat G13: Carbon graphite antimony impregnated (A), (Item no. 1.1)
Axial movement: ±1.0 mm Carbon graphite resin impregnated (B) Shaft diameter: d1 = 16 … 80 mm (0.63" … 3.15")
Temperature: t = –20 °C … +180 °C (–4 °F … +355 °F)
Materials M32 Sliding velocity: vg = 10 m/s (33 ft/s)
Seal face: CrMo special cast steel (S) Items and descriptions as for type M3N, but with carbon
Seat G9: Carbon graphite antimony impregnated (A), graphite seal face shrink-fitted into the seal face carrier Seal face: Silicon carbide (Q12, Q22),
Carbon graphite resin impregnated (B) (Item no. 1.1). Tungsten carbide (U22)
Seat G4: Silicon carbide (Q1, Q2)
Standards and approvals Seal face: Carbon graphite antimony impregnated (A), Seat G13: Carbon graphite antimony impregnated (A),
• EN 12756 Carbon graphite resin impregnated (B) Carbon graphite resin impregnated (B)
Seat G4: Special cast CrMo steel (S), Seat G6 (M37GN4): Silicon carbide (Q1, Q2)
Recommended applications Silicon carbide (Q1, Q2) Seat G9 (M37GN): Carbon graphite antimony impregnated
• Chemical industry Seat G6 (M32N4): Special cast CrMo steel (S), (A), Carbon graphite resin impregnated (B),
• Pulp and paper industry Silicon carbide (Q1, Q2) Silicon carbide (Q1, Q2)
• Water and waste water technology Seat G9 (M32N): Silicon carbide (Q1, Q2)
• Building services industry Seat G6 also available in A, B = G30 (longer installation
• Food and beverage industry Seat G6 also available in A, B = G30 (longer installation length than G6)
• Sugar industry length than G6)
• Low solids content media
• Water and sewage water pumps
• Submersible pumps
• Chemical standard pumps
• Eccentric screw pumps
• Cooling water pumps l22 l 23
• Basic sterile applications
d 13

d 13

M32 M37G

M32 © EagleBurgmann
M37G © EagleBurgmann

8
 Seat alternatives
l6 l6 l 16 l 16 l6
l5 l5 l 15 l 15 l5

20°

30°
30°

R
R

20°
20°

l 28 l 14 l 13 l 19

+0,1

H7

H11
+0,1
H11

H8
l 10 l 18
H8

l 12 l 11

d 6 H11

d 12 H7
d 7 H8

d 12
d 11

d 11

d7
d6
d7
d6

G9 (EN 12756)
G9 © EagleBurgmann
G6 (EN 12756) G4
G4 © EagleBurgmann
G13 G16G16 (EN 12756)
© EagleBurgmann
G6 (M32N4) © EagleBurgmann G13 (M3) © EagleBurgmann

Dimensions in mm
d1 d3 d6 d7 d8 d111) d121) d13 db l1N l 33) l5 l6 l7 l8 l9 l10 l11 l12 l13 l14 l15 l16 l18 l19 l213) l22 l23 l28 b2) R
6 14 – – – 11.5 16 16 8 – – – – – – – – 9 6.5 7.1 5.6 1.2 3.8 – – 10.5 11.9 – – – 1.2
8 18 – – – 15.5 19.2 18 11 – – – – – – – – 9 8 7.1 7 1.2 3.8 – – 15.5 16.9 – – – 1.2
10* 19 17 21 3 15.5 19.2 20 13 40 15.5 1.5 4 8.5 17.5 10 7.5 9 7.5 7.1 6.6 1.2 3.8 – – 15.5 16.9 – 6.6 (8) 1.2
12* 21 19 23 3 17.5 21.6 22 16 40 16 1.5 4 8.5 17.5 10 7.5 10 6.5 7.6 5.6 1.2 3.8 – – 15.5 17.4 – 6.6 (8) 1.2
14* 23 21 25 3 20.5 24.6 24 18 40 16.5 1.5 4 8.5 17.5 10 7.5 10 6.5 7.6 5.6 1.2 3.8 – – 15.5 17.4 16.5 6.6 (8) 1.2
15 24 – – – 20.5 24.6 25 19 – – – – – – – – 11 7.5 8.6 6.6 1.2 3.8 – – 15.5 17.4 – – – 1.2
16* 26 23 27 3 22 28 26 21 40 18 1.5 4 8.5 17.5 10 7.5 11.5 8.5 9 7.5 1.5 5 – – 17.5 19.5 16.5 6.6 (8) 1.5
18* 29 27 33 3 24 30 31 23 45 19.5 2 5 9 19.5 11.5 8.5 12.5 9 10 8 1.5 5 15 7 18.5 20.5 18 7.5 (8) 1.5
20* 31 29 35 3 29.5 35 34 26 45 22 2 5 9 19.5 11.5 8.5 12.5 8.5 9.5 7.5 1.5 5 15 7 20 22 19 7.5 (8) 1.5
22* 33 31 37 3 29.5 35 36 28 45 21.5 2 5 9 19.5 11.5 8.5 12.5 8.5 9.5 7.5 1.5 5 15 7 21.5 23.5 20.5 7.5 (8) 1.5
24* 35 33 39 3 32 38 38 30 50 23.5 2 5 9 19.5 11.5 8.5 12.5 8.5 9.5 7.5 1.5 5 15 7 23 25 22 7.5 (8) 1.5
25* 36 34 40 3 32 38 39 31 50 26.5 2 5 9 19.5 11.5 8.5 12.5 8.5 9.5 7.5 1.5 5 15 7 24.5 26.5 23.5 7.5 (8) 1.5
26 37 – – – 34 40 40 32 – – – – 9 – – – 13 9 10 8 1.5 5 – – 24.5 26.5 23.5 – – 1.5
28* 40 37 43 3 36 42 42 35 50 26.5 2 5 9 19.5 11.5 8.5 14 10 11 9 1.5 5 15 7 24.5 26.5 24.5 7.5 (8) 1.5
30* 43 39 45 3 39.2 45 44 37 50 26.5 2 5 9 19.5 11.5 8.5 14 11.5 11 10.5 1.5 5 15 7 24.5 25 24.5 7.5 (8) 1.5
32* 46 42 48 3 42.2 48 46 39 55 28.5 2 5 9 19.5 11.5 8.5 14 11.5 11 10.5 1.5 5 15 7 28 28.5 28 7.5 (8) 1.5
33* 47 42 48 3 – – 47 40 55 28.5 2 5 9 19.5 11.5 8.5 – 12 – – – – 15 7 – – – 7.5 (8) 1.5
35* 49 44 50 3 46.2 52 49 43 55 28.5 2 5 9 19.5 11.5 8.5 14.5 12 11.5 11 1.5 5 15 7 28 28.5 28 7.5 (8) 1.5
38* 53 49 56 4 49.2 55 54 45 55 33.5 2 6 9 22 14 10 14.5 11.3 11.5 10.3 1.5 5 16 8 31 32.2 31 9 7.5 1.5
40* 56 51 58 4 52.2 58 56 49 55 36 2 6 9 22 14 10 14.5 11.8 11.5 10.8 1.5 5 16 8 34 34.7 34 9 (8) 1.5
42 59 – – – 53.3 62 58 52 – – – – 9 – – – 17 13.2 14.3 12 2 6 – – 35 37.3 35 – – 2.5
43* 59 54 61 4 – – 59 52 60 38.5 2 6 9 22 14 10 – 13.2 – – 2 – 16 8 – – – 9 7.5 2.5
45* 61 56 63 4 55.3 64 61 55 60 39.5 2 6 9 22 14 10 17 12.8 14.3 11.6 2 6 16 8 36.5 39.2 36.5 9 (8) 2.5
48* 64 59 66 4 59.7 68.4 64 58 60 46 2 6 9 22 14 10 17 12.8 14.3 11.6 2 6 16 8 42 44.7 42 9 (8) 2.5
50* 66 62 70 4 60.8 69.3 66 61 60 45 2.5 6 9 23 15 10.5 17 12.8 14.3 11.6 2 6 17 9.5 43 45.7 43 9.5 (8) 2.5
53* 69 65 73 4 – – 69 64 70 47 2.5 6 9 23 15 12 – 13.5 – – – – 17 9.5 – – – 11 8 2.5
55* 71 67 75 4 66.5 75.4 71 66 70 49 2.5 6 9 23 15 12 18 14.5 15.3 13.3 2 6 17 9.5 47 49 47 11 (8) 2.5
58* 76 70 78 4 69.5 78.4 78 69 70 55 2.5 6 9 23 15 12 18 14.5 15.3 13.3 2 6 18 10.5 50 52 50 11 (8) 2.5
60* 78 72 80 4 71.5 80.4 79 71 70 55 2.5 6 9 23 15 12 18 14.5 15.3 13.3 2 6 18 10.5 51 53 51 11 (8) 2.5
63* 83 75 83 4 – – 83 74 70 55 2.5 6 9 23 15 12 – 14.2 – – – – 18 10.5 – – – 11 (8) 2.5
65* 84 77 85 4 76.5 85.4 85 77 80 55 2.5 6 9 23 15 12 18 14.2 15.3 13 2 6 18 10.5 52 54.3 52 11 (8) 2.5
68* 88 81 90 4 82.7 91.5 88 80 80 55 2.5 7 9 26 18 12.5 19 14.9 16 13.7 2 6 18.5 11 53 55.3 52.7 11.3 (8) 2.5
70* 90 83 92 4 83 92 90 83 80 57 2.5 7 9 26 18 12.5 18 14.2 15.3 13 2 6 19 11.5 54 56.3 54 11.3 (10) 2.5
75* 98 88 97 4 90.2 99 98 88 80 62 2.5 7 9 26 18 12.5 18 15.2 15.3 14 2 6 19 11.5 55 56.3 54 11.3 (10) 2.5
80* 100 95 105 4 95.2 104 103 93 90 61.8 3 7 9 26.2 18.2 13 19 16.2 16.3 15 2 6 19 11.5 58 59.3 58 12 10 2.5
1) Fitting
dimensions d11 and d12 only apply to type M37G with d1 >16 mm
2) Dimensions in brackets lie either above or below l1N
3) l valid for M3...N, l valid for M3
3 21
* in accordance with EN 12756

9
Mechanical seals • Mechanical seals for pumps • Pusher seals

M7N
l3 l8

f l9
1) d >100 mm: 30°
1
2) d >100 mm: +0.1
1
mx 3) d >100 mm: H7
1

1.6 1.5 1.4 1.1 2 3

(>100 = h8)

2)

3)
d6 H11

d7 H8
d1 h6
d3
1.2 1.3

1)
20
° d8
Features
l5 l6
• For plain shafts
• Single seal I1k 5 I7 +0,5
• Unbalanced
• Super-Sinus-spring or multiple springs rotating
• Independent of direction of rotation
• Pumping screw for media with higher viscosity (M7..F) Materials Item Part no. to DIN 24250 Description
M7N © EagleBurgmann
• Variant with PTFE secondary seals for high chemical Seal face: Special cast CrMo steel (S), Silicon carbide 1.1 472 Seal face
resistance (M78N) (Q1, Q2), Aluminium Oxide (V) 1.2 485 Drive collar
Seat G9: Carbon graphite antimony impregnated (A), 1.3 474 Thrust ring
Advantages Carbon graphite resin impregnated (B), Silicon carbide 1.4 412.1 O-Ring
• Universal application opportunities (Q1*, Q2*) 1.5 477 Spring
• Efficient stock keeping Seat G4: Silicon carbide (Q1*, Q2*) 1.6 904 Set screw
• due to easily interchangeable faces Seat G6: Silicon carbide (Q1*, Q2*) 2 475 Seat (G9)
• Extended selection of materials Seat G13: Carbon graphite antimony impregnated (A), 3 412.2 O-Ring
• Insensitive to low solids contents Carbon graphite resin impregnated (B)
• Flexibility in torque transmissions Recommended applications
• Self cleaning effect Secondary seals: EPDM (E), NBR (P), FKM (V), FFKM (K) • Process industry
• Short installation length possible (G16) Springs: CrNiMo steel (G) • Chemical industry
Metal parts: CrNiMo steel (G), Duplex (G1) • Pulp and paper industry
Operating range (see note on page 1) • Water and waste water technology
Shaft diameter: d1 = 14 … 100 mm (0.55" … 3.94") * Cannot be combined with seal face in S • Shipbuilding
Pressure: p1 = 25 bar (363 PSI) • Food and beverage industry
Temperature: t = –50 °C … +220 °C Standards and approvals • Lube oils
(–58 °F … +428 °F) • EN 12756 • Low solids content media
Sliding velocity: vg = 20 m/s (66 ft/s) • Water / sewage water pumps
Axial movement: Torque transmissions • Chemical standard pumps
d1 = … 25 mm: ±1 mm • Vertical screw pumps
d1 = 28 … 63 mm: ±1.5 mm • Gear wheel feed pumps
d1 = from 65 mm: ±2 mm • Multistage pumps (drive side)
• Circulation of printing colors with viscosity
• 500 … 15,000 mm2/s
d1 Torque
>100transmission
mm (3,94") Drive key
Torque transmission by (M7S2 / M74S2)
Set screw with cone point © EagleBurgmann
4 set screws with cone Torque transmission
point. Offset: 90°
Drive key © EagleBurgmann
Seat alternatives
l6 l6 l 16 l 16 l6
l5 l5 l 15 l 15 l5
20°

30°

30°
R

20°
20°

l 28 l 14 l 13 l 19
+0,1

H11
H11

H7
d 11 +0,1

H8

l 10 l 18
H8

l 12 l 11
d 6 H11

d 12 H7
d 7 H8

d 12
d 11
d6

d7

d7
d6

G9 (EN 12756) G6 (EN 12756) G4 G13 G16

G16 (EN 12756, but l1k
© EagleBurgmann
G9 (M7N) © EagleBurgmann G4 © EagleBurgmann
G6 (M32N4) © EagleBurgmann G13 (M3) © EagleBurgmann is shorter than specified)

10
MATEC Web of Conferences 232, 04051 (2018) https://doi.org/10.1051/matecconf/201823204051
EITCE 2018

EPS Sensor Torsion Bar Design and Structure Analysis

Shu-juan Fang1,a and Qing-tao Lv2

1
Hulunbuir University Mining College, Inner Mongolia Hulunbuir, China
2
Hella Shanghai Electronics Co.,Ltd.,Shanghai,China

Abstract. Torsion bar design is very important on Electrical power steering(EPS) development. The mathematical
model and structure for torsion bar are developed, and the torsion bar outer spline is cancelled based on the
automotive parameters. This kind of structure leads to cost down and time saving on torsion bar manufacturing
process. The 3D model and finite element on sub-EPS(torsion bar, input shaft, output shaft) are developed, and then
do the contact nonlinear analysis based on interference between torsion bar and output shaft. The EPS structure and
test reliability according to function test and durability test are verified and the test results are passed.

1 Introduction realize more than 30Newton on steering assist. In addition,

in order to make the steering system being forced
Since the first Electrical Power Steering (EPS) was
uniformly, the P-EPS motor and deceleration mechanism
developed in the 1980s, the assembly rate of EPS was
are assembled separately, which is called as double pinion
increased by 8%-10% annually. Nowadays, EPS have
type EPS(DP-EPS) [3]. Details are shown in Figure 1.
been installed in most of the automotives all over the
world, and around 80% automotives have EPS in United
States, Japan and other developed countries [1, 2]. EPS is
very important in energy saving, emission reduction in
technological progress of the entire automotive industry.
In parallel, the requirements for the key components of
the electrical power steering system are also increasing.

2 Classification of EPS system Figure 1. Classification of EPS

EPS is classified as following based on different assembly

locations in the vehicles: column type EPS (C-EPS),
pinion type EPS (P-EPS) and rack type EPS (R-EPS).
C-EPS is mainly installed under the steering wheel, which
is suitable for Class A0 and Class A vehicles. Normally,
EPS can realize more than 10 Newton on steering assist.
P-EPS is mainly installed in the front cabin near the rack
position, which is suitable for Class B or some of Class C
vehicles. Normally, P-EPS can realize more than 20
Newton on steering assist. R-EPS is mainly installed near
the engine and parallel with rack, which is suitable for C
or C above vehicles or trucks. Normally, R-EPS can
3 Principle of electric power steering
system
EPS mainly includes in the following 4 parts: Electrical
Control Unit (ECU), motor, sensor and deceleration
mechanism [4, 5]. Among all the parts, the torsion bar is a
very important part of the deceleration mechanism which
converts the torque of the human hand into the
mechanical angle, and then EPS sensor send out signal
after testing the angle. When the automotive is started up,
EPS ECU does self-test firstly, including power supply
voltage signal test, engine signal test, speed signal test,

a Corresponding author: 32577140@qq.com

© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(http://creativecommons.org/licenses/by/4.0/).
MATEC Web of Conferences 232, 04051 (2018) https://doi.org/10.1051/matecconf/201823204051
EITCE 2018

torque signal test, etc. After all the tests successfully, then Then,
ECU drives the motor to realize the steering assist T T×16
τ = W = π×d3 (5)
according to the assist curve. p

Strength condition
4 Torsion bar design τmax ≤ [τ] (6)
[τ] is the maximum allowable shear stress for the
In order to be assembled easily, the torsion bar in the
materials [7].
electric power steering system is usually designed as a
The target vehicle type is A0. The torque applied to
slender rod structure by arc in the middle, and an outer
the torsion bar by turning the steering wheel manually is
spline at one end and a pinhole near the other end. The
normally below 10Nm. If the safety factor is 1.5, then
spline side of the torsion bar is pressed into the output
Tmax = 10Nm ×1.5 = 15Nm. The torsion bar material is
shaft by a hydraulic press, and then a torque sensor is
spring steel 60Si2MnA, G=79000MPa, [τ] = 590MPa. In
installed. After the torque midpoint of the EPS system is
the formula (6), d=5.01mm can be obtained. Then take
adjusted, the input shaft is installed, and the pin is
d2=5.5mm.
punched into the pinhole of the input shaft and the torsion
Generally, the mechanical limit angle of the torsion
bar.
bar of EPS deceleration mechanism is ±4 degrees, so the
However, in order to save the processing cost and
working torsion angle of the torsion bar is set to ±6
simplify the processing steps, a torsion bar without spline
degrees, which can meet the measurement range of most
is designed. Torque is transmitted by the interference fit
torque sensors. If put the φ = 6° and d2=5.5mm into
between the torsion bar and the output shaft. The radial
formula (3), the working length can be calculated La =
contact pressure is generated by the interference, and
22.3mm , then choose as 22mm. End diameter is
torque and axial force are transmitted by the friction force
recommended as d1 = d2 × (1.2~1.3) = (7.2~7.8)mm,
generated by the interference.
then d1=7.8mm is taken.
The torsion bar is regarded as a rod with the same
The transition part from the end of the torsion bar to
section diameter to simplify the calculation. The torsion
the diameter of the working length is called as the
deformation of the rod is studied by using the principle of
transition section. The size of the transition section is
material mechanics, and the stiffness formula of the rod is
gradually changing so as to minimize the stress
obtained as follows [6].
concentration.
TL 180°
φ = GI × (1) L2 −L1 d1 −d2
ρ π The transition section is Lg = ≈ =
2 2tan15°
φ is the torsion angle (degree), T is the torsion
3.4mm.
moment on the section of the rod, L is the length of the
L2
g d1 −d2
rod, G is the shear modulus of the material, Iρ is the The transition section radius R = + =
d1 −d2 4
polar inertia moment, and GIρ is the torsion stiffness. For
6.87mm, take as R=6.9mm.
solid circular rods, the Iρ moments are as follows.
In addition, the transition section is generally divided
πd4 into effective length and non effective length. The
Iρ = 32
(2)
effective length is as following
Therefore, the torque angle φ is as follows.
Lg d2 d 2 d 3
TL T×L×32 180° Le = [ + ( 2 ) + ( 2 ) ] = 2.06mm，take as Le =
φ = GI = × 3 d1 d1 d1
ρ G×π×d4 π
2mm
(3)
Based on the above analysis, the length of the middle
When the rod is subjected to torque T, the shear stress
segment can be obtained L1 = La − 2 × Le = 18mm.
is calculated as follows.
Then set the middle parts length as Ld = 20mm
T
τ= (4) So, the total length is as following
Wp
L = L1 + 2 × Lg + 2 × Ld = 64.8mm

2
MATEC Web of Conferences 232, 04051 (2018)
EITCE 2018
5 Structure analysis of torsion bar
In general, mechanical design is based on experience, but
for the complex structure with strict safety requirements,
the stress and strain analysis is needed [8,9]. In addition,
it is difficult to directly measure the stress and strain on
the two contact surfaces of the interference fit between
the torsion bar and the output shaft, and there is
alternating load at this kind of contact surface. Therefore,
it is necessary to obtain the appropriate interference fit
parameters by theoretical analysis.
5.1 Establish a geometric model
Because of the complexity of the model, the 3D model
(input shaft, torsion bar, output shaft) is established by 3D
software, and then it is transferred to the intermediate
format file and imported into the finite element analysis
software. The torsion bar and the input shaft are fixed by
pins. When doing the finite element analysis, the nodes of
the outer diameter of the torsion bar and the inner
diameter of the input shaft can be bind together, so the
finite element model of the pin is no needed. That means,
the 3D model of the pin need not to be imported. It saves
the workload of dividing grid and saves computing time.
Table 1. Analysis results based on different meshing
Mesh method Sweep(Hex Smart)
Density/mm 5 6 7 8
Displacement/mm 0.218 0.215 0.204 0.203 0.213
(1)The displacement of the hexahedral element
SOLID45 decreases gradually from 5 to 8. This is because
the analysis results deviated when the mesh density
decreasing. When the mesh density is 5mm, the
displacement is the largest 0.218mm, but the difference is
only 0.003mm comparing with density 6.
(2)When Hexahedral element SOLID45 mesh density
is changed from 8 to 10, the displacement is increasing,
because the analysis results are more and more deviated
from the true value with the mesh density decreasing.
(3) The displacement gradient is much larger than that
of hexahedron element when the mesh density of
tetrahedron element is changed from 6 to 8. It can not
converge when the mesh density is 8. It shows that
different mesh density has great influence on the result of
tetrahedral element analysis.
3
https://doi.org/10.1051/matecconf/201823204051
Output shaft
Torsion bar pin Input shaft
Figure 2. EPS sub-assembly (input shaft, output shaft, torsion
bar, pin)
5.2 Mesh
The traditional linear analysis method divides the contact
surfaces into meshes with the same nodes and assumes
that two contact objects transmit contact force through the
nodes. This method needs to know the exact contact
location in advance so as to set boundary conditions.
However, this linear analysis method can not simulate the
real friction state [10-12]. To reflect the real status of the
torsion bar accurately, the strain analysis results of the
torsion bar under 15Nm torque are discussed with
different mesh densities and element types. Table 1 shows
the maximum strain in the Y direction when using the
hexahedral element SOLID45 and the tetrahedral element
Tet with different mesh densities.
Free(Tes smart)
9 10 6 7 8
0.249 0.166 0.113 No convergence
(4)When the mesh density is same, the displacement
of the hexahedron element and the tetrahedron element is
much different, and the difference is more than 0.05 mm.
This shows that the hexahedral mesh is more suitable for
the important analysis place.
In summary, the hexahedron element can get a result
closer to the true value than the tetrahedron element. So,
for the contact nonlinear analysis, hexahedron element is
used as far as possible, wedge element is used in the
complex part of the model, and mesh is divided by free
method. In order to reduce the calculation time,
hexahedron element is selected and the mesh density is 6.
Hexahedral element meshed with density 10 is shown in
Figure3. Tetrahedral element meshed with density 6 is
shown in Figure4.
MATEC Web of Conferences 232, 04051 (2018) https://doi.org/10.1051/matecconf/201823204051
EITCE 2018

of steel and high-quality structural steel in material

mechanics).

5.5 Loading

The external torque can not be applied by finite element

analysis software [13].
Figure 3. Hexahedral element density 10
Considering the safety factor, the maximum torque by
hand is set as 15Nm, and 1000N force is applied to the
tangent direction of the outer diameter circle of the input
shaft.

Figure 4. Tetrahedron element density 6 5.6 Analysis results

The analysis results show that maximum stress is 5751N,

5.3 Material properties the maximum tangential displacement is 0.22mm and the
strain analysis cloud chart are shown in Figure 5. But this
The torsion bar is made of spring steel, and the input shaft
is the applied point of the external load, which can not
and the output shaft are made of 40Cr. The material
completely reflect the displacement of the torsion bar.
properties are in table2.
The edge nodes of the middle section of the torsion bar
Table 2. Material properties are node255 and Y displacement is 0.255mm. The

Part material Material Elastic Possion’s

diameter of the middle section is 6mm, which shows that
name density/ modulus/ ratio τ 0.255
kg/m³ GPa the angle of rotation isθ = = = 0.085. If change to
γ 3
Torsion 60Si2MnA 7850 213 0.29
bar angle, it will be 4.9deg. The delta angle is 1.1deg
Input 40Cr 7850 206 0.3 (6deg-4.9deg).
shaft
Output 40Cr 7850 206 0.3 The main reason is the finite element analysis is based
shaft on the length（L1 = 18.8mm） of the middle section of
the torsion bar, and the theoretical analysis is based on the
5.4 Boundary conditions length（La = 22mm） of the working section of the

In the nonlinear analysis, only the key parts are needed to torsion bar, which inevitably results in the analysis results

establish the contact surface, and the rest can be fixed by lower than the theoretical value. Then, 1.09 degrees can

element coupling. The specific boundary conditions are be obtained, when put the delta 4.2mm(L2 − L1 ) into the

(1) Combine the outer diameter of the torsion bar and formula (3). It is concluded that the difference between

the inner diameter of the input shaft. The Ground the theoretical analysis and the finite element analysis is

resistance moment is simulated by the full coupling of the only 0.01 degrees and the error is 0.1%.

output shaft.
(2) By using contact nonlinear analysis, the contact
surface attributes are defined for the contact parts of the
torsion bar and the output shaft. The contacted parts of the
torsion bar outer diameter are defined as CONTA173
contact surface, and the contacted parts of the output shaft
inner diameter are defined as TARGE170.
(3) The contact stiffness factor FKN is 0.1.The friction
coefficient of the contact part between the torsion bar and
the output shaft is 0.11 (referring to the press-fit condition

4
MATEC Web of Conferences 232, 04051 (2018) https://doi.org/10.1051/matecconf/201823204051
EITCE 2018

Finite Elements,Journal of System

Design&Dynamics, 3(3):429-440,2009.
4． Yuji Kozaki，GoroHirose, Shozo Sekiya. Electric
Power steering [J]. Motor &Control,1999:449-459
5． Didier ANGLEVIEL，Didier FRACHON，Gerald
MASSON ． Development of a Contactless Hall
effect torque sensor for E1ectric Power
Steering[J],SAE．2006—0 1—0939．
6． Fang Shu-juan,Design and Calculation of
Automotive Electric Power Steering System and
Worm Drive,Mechanical Design, (12):52-55,2015.
7． LV Q T,XUE S, ZHU W Q,el al.. Pull-rod bent
Figure 5. EPS analysis result
focusing mirror subassembly engineering analysis
based on contact nonlinear analysis method[J].
6 Test validations SPIE,2009,7385:73850E_1-73850E_9.
According to the theoretical calculation results, the 8． Lv Qing-tao,Pull-rod bent focusing mirror
torsion bar structure and related parameters suitable for system,Beijing,Graduate University of Chinese
the target vehicle are designed, and the EPS system is Academy of Science,2010.
developed. The system is tested by functional test bench, 9． LEMANQUAND V , LEMANQUARD G. Magnetic
and EPS assistance is obtained. differential torque sensor. IEEE Transactions on
According to the EPS test standard, the durability test Magnetics. 1995 ,31 (6) :3188 - 3190.
is done, and the test results meet the requirements. 10. Lv Qing-tao,Xue Song,Peng Zhong-qi,Zhu
Wan-qian,Fu Yuan,Comparison of linear and contact

7 Conclusions non-linearity method for rod-beam compacting

focusing mirror assembly,Optical and Precision
According to the specific requirements of the target Engineering,18(8):1842-1848,2010 .
vehicle, a new type of torsion bar structure without 11. M. Ben Rabha, M.F. Boujmil, M. Saadoun, B.
external spline is designed. The geometric model of EPS Bessaïs, Eur. Phys. J. Appl. Phys. (to be published)
assembly is developed and the contact non-linear analysis 12. FF Mahmoud,AG EI-Shafei,MA Attia,AAA
is made. The analysis results meet the requirements. The Pahman,Analysis of quasistatic frictional contact
structure is verified by the function and durability test. problems in nonliear viscoelasticity with large
deformations, International Journal of Mechanical
References Science,66(4):109-119,2013.
13. Amassad,Fabre,Sofonea,A quasistatic viscoplastic
1． Ji-Hoon Kim, Jae-Bok Song. Control logic for an
contact problem with normal compliance and
electric power steering system using assist motor [J].
friction,Ima Journal of Applied
Mechatronics (S0957-4158), 2002, 12(3):447-459.
Mathematics,69(5):463-482,2004.
2． Y．TOKUMOTO，M．SHIBA，N．MAEDA，
K．KOTAKA．Development of Next—Generation
Steering Sensor for E1ectric Power
Steering[J] ． Koyo Engineering Journal English
Edition No．1 65E(2004)．
3． F Besson,G Ferraris,M Guingand,JPD
Vaujany,Simulation of the Dynamic Behavior of
Electric Power Steering System Using Dedicated