1

GT Power Simulation of the Influence of Exhaust Manifold Design on Sound Quality

Ivan Arbuckle

Emcon Technologies

GT Power User Conference – Detroit 2009

2

� Objective

• Study the exhaust manifold design factors that influence noise− Asymmetry

− Restriction

− Cylinder-Cylinder Interference

• Characterize the Engine Noise Source Characteristics − Ps and Zs from the MultiLoad Template

3

EVO EVC

Why is there noise in the Exhaust?

EVO EVC

EVO EVC

•The noise is caused by the unsteady mass flow rate from the engine. This leads to oscillations in the exhaust pipe which are emitted at the outlet as tail pipe noise.

•If the flow from the engine was a steady (non oscillating) flow there would be no order noise.

4

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

0.5 EO Cylinder 1

Why is 0.5 OE suppressed in a 2 cylinder engine ?

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

0 180 360 540 720

0.5 EO SummationTotal Cancellation-∞ dB re 1 Cylinder

+ =

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

All orders are ‘shifted’ 360°CA for the second cylinder since it fires 1 revolution later.

0.5 EO Cylinder 2

5

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

0 180 360 540 720

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

1.0 EO Cylinder 1

1.0 EO Cylinder 2

1.0 EO Summation(Doubled, or +6dB)

Why is 1 EO not suppressed in a 2 cylinder engine ?

6

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

0.5 EO Cylinder 1

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

0.5 EO Cylinder 2Phase Shift 45°

What if firing is not 360 °°°° separated for a twin cylinder ?

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

0 180 360 540 720

Total 0.5 EOOnly Partial Cancellation

-6dB re 1 Cylinder

7

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

0 180 360 540 720

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

-1.5

-1

-0.5

0

0.5

1

1.5

0 180 360 540 720

0.5 EO Cylinder 1

0.5 EO Cylinder 2Lower Amplitude

Total 0.5 EOOnly Partial Cancellation

-6dB re 1 Cylinder

An amplitude shift can be caused by;

•incomplete combustion in 1 cylinder

•inconsistent cylinder-to-cylinder Veff

•or imbalanced restriction in exhaust manifolds.

What else can cause incomplete 0.5 EO Cancellation ?

8

Singles & Twins

45°V Twin Common

Crank Throw

(405°-315°)Single Cylinder

(4 Stroke)

Horizontally Opposed Twin (360°Firing)

9

Singles & Twins: Source Level (Ps)Engine Source Strength, Ps

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re

20µP

a

Single

360° H-O Twin

45° V Twin

0.5 EO Engine Source Strength, Ps

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re

20µP

a

Single

360° H-O Twin

45° V Twin

1 EO

Engine Source Strength, Ps

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re

20µP

a

Single

360° H-O Twin

45° V Twin

1.5 EO Engine Source Strength, Ps

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re

20µP

a

Single

360° H-O Twin

45° V Twin

2 EO

10

4 Cylinder (Firing 1-2-4-3)

4-2-1 Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re

20µP

a

0.5 EO 1 EO1.5 EO 2 EO

4-2-1 with equal length primary &

secondary

I4 with Log Manifold

Log Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

0.5 EO 1 EO1.5 EO 2 EO

1

2

3

4

Boxer 4

12

34

Boxer 4

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

0.5 EO 1 EO1.5 EO 2 EO

11

V6(Firing 1-4-2-5-3-6 = R-L-R-L-R-L)

1

2

3

4

5

6

6-2-1 Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

0.5 EO 1 EO1.5 EO 3 EO Log Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

0.5 EO 1 EO1.5 EO 3 EO

1

2

3

4

5

6

90°V6Firing 90°-150°-90°-150°-90°-150°

Odd Fire V66-2-1

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpmP

s, d

B(li

n) re

20µ

Pa

0.5 EO 1 EO1.5 EO 3 EO

6-2-1 SymmetricalAsymmetric

Log

12

Cylinder 2

Cylinder 1

Cylinder Interference in Twin Cylinder Engines

13

Cross Plane V8(Firing 1-3-7-2-6-5-4-8 = R-R-L-R-L-L-R-L)

-0.05

0

0.05

0.1

0.15

0.2

0.25

0 90 180 270 360 450 540 630 720

Crank Angle °

Exh

Val

ve M

ass

Flo

w

(kg/

sec)

Cylinder 1 Cylinder 2 Cylinder 3 Cylinder 4 Cylinder 5Cylinder 6 Cylinder 7 Cylinder 8 Summation

Symmetrical 8-1 Manifold

Summation

Uneven bank-to-bank firing on a cross-plane crank V8 causes variation from cylinder to cylinder. A 8-1 manifold can overcome this

(as can a flat plane crank).

8-2-X-2

-0.05

0

0.05

0.1

0.15

0.2

0.25

0 90 180 270 360 450 540 630 720

Crank Angle °

Exh

Val

ve M

ass

Flo

w

(kg/

sec)

Cylinder 1 Cylinder 2 Cylinder 3 Cylinder 4 Cylinder 5Cylinder 6 Cylinder 7 Cylinder 8 Summation

14

V8 (Firing 1-3-7-2-6-5-4-8)

Dual V8 – No Bank-to-Bank

mixing.

H-Pipe V8 with equal length

primary

X-Pipe V8 with equal length

primary

5

6

7

8

1

2

3

4

8 to 2 Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO

H-Pipe

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO

X-Pipe

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO

15

V8 (Firing 1-3-7-2-6-5-4-8)

8-into-1 V8 with equal

length primary

Y-Pipe V8 with equal length

primary

Y-Pipe V8 with equal primary, Unequal Secondary

8 to 1 Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO

Asymetric Y-Pipe

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO8-2-1 Y Manifold

100

110

120

130

140

150

160

170

180

190

200

1000 2000 3000 4000 5000 6000

rpm

Ps,

dB

(lin)

re 2

0µP

a

2.5 EO 1.5 EO2 EO 4 EO

16

H-Pipe ‘Tuning’ on a V8 Application

60

70

80

90

100

110

120

1000 2000 3000 4000 5000 6000 7000

rpm

Tai

l Pip

e N

oise

- d

B(A

)

None

1.75"

2.5"

X

1.5EO

60

70

80

90

100

110

120

1000 2000 3000 4000 5000 6000 7000

rpm

Tai

l Pip

e N

oise

- d

B(A

)

2.5EO

60

70

80

90

100

110

120

1000 2000 3000 4000 5000 6000 7000

rpm

Tai

l Pip

e N

oise

- d

B(A

)

4EO

60

70

80

90

100

110

120

1000 2000 3000 4000 5000 6000 7000

rpm

Tai

l Pip

e N

oise

- d

B(A

)

8EO

Varying the H-pipe diameter can have a large affect on half order levels.

This is used as another tuning parameter to achieve the best sound quality.

Small

Large

17

modeFrontier Optimization V6 →→→→V8 : Work Flow

1

4

2 3

Latin Hypercube

MOGA

18

modeFrontier Optimization V6 →→→→V8 : Targets

100

110

120

130

140

150

160

170

180

190

0 50 100 150 200 250 300

Frequency (Hz)

Sou

rce

Pre

ssur

e Le

vel (

dB)

at 2

000

rpm

6-2-1 Baseline V8 with H-pipe

Ps at 2000 rpm

19

Baseline

V6 (6-2-1)

V8

V6 Log

modeFrontier Optimization V6 →→→→V8 : Results

Optimized V6

20

Conclusions

Multi-Cylinder Engines have the potential to significantly suppress half orders

But there are many design features which can re-introduce half order content;

Engine Factors�Odd-Fire Engine�Even Fire engine with Odd-Fire on each bank

Manifold Factors�Length Differences - introduces phase errors�Restriction Differences - introduces amplitude errors�Inadequate Bank-to-Bank Mixing