Journal of Physics: Conference Series

PAPER • OPEN ACCESS

Cylinder head gasket design optimization based on high definition

metrology
To cite this article: Yaxiang Yin et al 2021 J. Phys.: Conf. Ser. 1983 012116

View the article online for updates and enhancements.

This content was downloaded from IP address 181.215.8.89 on 13/08/2021 at 04:05

MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

Cylinder head gasket design optimization based on high

definition metrology

Yaxiang Yin1, 2, Kun Wang1, 2, Yiping Shao3, Shichang Du1, 2, 4, Lifeng Xi1, 2 and
Tangbin Xia1, 2
1
State Key Lab of Mechanical System and Vibration, Shanghai Jiao Tong University,
Shanghai, 200240, China
2
School of Mechanical Engineering, Shanghai Jiao Tong University, No. 800
Dongchuan Road, Shanghai 200240, China
3
College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou,
China
4
E-mail: [email protected]

Abstract. Traditional cylinder head gasket design is based on the assumption that the mating
surfaces’ form is ideal plane. However, high definition metrology (HDM) shows that the top
surface of cylinder blocks and the bottom surface of cylinder heads are far from ideal plane and
presents a macro form error. To compensate the macro form error of the mating surfaces, a
gasket design optimization method is proposed in this paper. It mainly consists of two modules.
The first module is surface alignment and gap calculation. Measured surface topography is
aligned by control point registration and the control points are found by region labelling
algorithm. The second module is gasket design optimization based on surface gap. A case
study shows the details of the proposed gasket design optimization method.

1. Introduction
Cylinder head gasket is a crucial element for sealing in automobile engine assembly. The multi-layer
steel (MLS) gasket is the most popular gasket scheme for its reliable sealing performance and rich
design possibilities [1]. The most important function of the gasket is to prevent the gas in combustion
chamber from leakage. The first combustion seal on the gasket is called stopper, and it is a zone at the
bore opening which is thicker than the rest of the gasket. The secondary combustion seal is the full
bead or embossments. Figure 1 shows the concept of stopper and full bead. Since the surface
topography of the mating cylinder head and block is rough in micro view, the pressure distribution on
the stopper zone is not uniformly [2]. To guarantee the sealing of combustion chamber, it is required
that the weakest pressure spot in stopper zone is greater than the maximum operation pressure of the
engine. So, in practice the torque of the cylinder head bolts is very high. However, high loading by
bolts caused the bore distortion and reduced the life of the gasket. Thus, it is desired to achieve
uniformly distributed pressure on the stopper zone.
One important reason of the non-uniformity of pressure distribution on the stopper zone is the
rough topography of the mating surfaces. The contact pressure will be high if the surface gap on that
spot is small and the contact pressure will be low if the surface gap on that spot is big. If the surface
gap on the stopper zone is known in advance, the stopper topography could be designed to compensate
the surface gap so as to achieve a relative uniform distributed pressure. However, surface gap is not

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

measureable directly. To solve this problem, an indirect surface gap inferring method is proposed in
this paper with the aid of high definition metrology (HDM).

Figure 1. Structure of a single layer of the MLS gasket.

In recent years, precision measurement has developed rapidly. One of the noncontact measurement
instrument named high definition metrology (HDM) is used to measure engineering surfaces with
large size. It can generate a surface height map of millions of data points for 3D inspection of a surface
[3]. The HDM instrument and an example of the engine block surface measured by HDM are shown in
Figure 2. The 3D surface topography examined by HDM presents a new platform, several researches
based on HDM such as 3D surface topography filtering [4-6], classification [7, 8], forecasting [9, 10]
and leakage monitoring [11-13] have been explored. With the help of HDM, a surface topography
alignment method is proposed to calculate the surface gap distribution.

Figure 2. Measurement by HDM.

The remainder of this paper is organized as follows: a detailed surface topography alignment
method is presented in Section 2. In Section 3, a case study demonstrates the effectiveness of the
surface alignment method and shows the gasket design optimization result. The last section draws the
conclusions and discusses the future research.

2. Methodology
HDM is employed to measure the engineering surfaces and generate millions of points. Since the
original HDM data is in a coordinates form and not exactly equidistant sampled. To extract the feature
points and apply image registration algorithm to align two measured surfaces, it is necessary to
convert the measurement data into a matrix form or an image. The preprocessing method developed in
Reference [14] is used to transform HDM coordinates [Xi, Yi, Zi] into a gray pixel I(m, n). The
transformed gray image contains all the height and spatial information of the original HDM data. A
binary image B(m, n) can also be generated to judge the boundaries and holes of the surfaces. Based
on the matrix form of the surface topography, the workflow to align two engineering surfaces is shown
in Figure 3.

Figure 3. Workflow to align two engineering surfaces.

2
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

2.1. Region labelling and feature points extraction

Labelling the feature regions (connected components) of the binary image, such as bolt holes and
cylinder bores, can help to extract assembly feature points. Typical feature points include the centroids
of circles or the endpoints of edges.
The basic steps to label the connected components of a binary image are as follows:
Step 1: Search for the next unlabeled position, p.
Step 2: Label all the positions in the connected component containing p by a flood-fill algorithm.
Step 3: Repeat steps 1 and 2 until all the positions are given a label number.
The labelling procedures and results are demonstrated by a simple example shown in Figure 4.

Figure 4. Region labelling procedures.

Then a coordinates set of the labelled region can be defined as Ci  ( m, n) L ( m, n )  i , where
L(m, n) represent the label of B (m, n) .
Once all the separated regions are found, the centroids can be calculated as Equation (1),
  ( m, n ) 
Pi    ,(m, n)  Ci
 Ci  (1)
where Ci represents the number of elements in set Ci and  means a rounding operation. Pairs of
feature points could be extracted from the two mating surfaces according to the assembly requirements.
The coordinates of feature points of B1 and B2 are denoted as [P1 ,P2  Pm ] and [Q1 ,Q2 ,Qm ]
respectively, where m is the number of feature points. Two example surfaces are generated, the
labelling result and feature points are shown in Figure 5.

Figure 5. Simulated surfaces and their feature points labelling.

3
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

2.2. Control points registration

Two mating surfaces can be measured and transformed to two gray images I1 and I2. They could be
registered to the same coordinate system by an affine transformation. Setting I1 as the reference image,
the registration error function is defined as Equation (2).
m
f ( A,b)   AQi  b  Pi
2
(2)
i 1
Solving the optimization problem min f ( A,b) could get the affine transformation matrix
 A 0
T   . Denote the pixel coordinates of I2 as (w, z), and the pixel coordinates of I2 after
 b 1
registration as (wr, zr), then (wr, zr) can be calculated as Equation (3).
[ wr , z r ,1]  [ w, z,1]T (3)
r
After registration, the coordinates of feature points on I and I1 become the same and the two
2
images are in a common coordinate system. Figure 6 shows an example of mating surfaces whose
feature points are aligned.

Figure 6. Example of aligned two surfaces.

2.3. Surface gap calculation

For aligned surface topography z1 and z2, the surface gap could be determined by Equation (4).
d  max( z1  z2 )  ( z1  z2 ) (4)
According to Equation (4), gap image G can be defined as Equation (5).
G  max( I1  I 2r )  ( I1  I 2r ) (5)

3. Engineering case study

3.1. Surface gap between the cylinder block and head

The engine block and head in this case are from an automobile company. The material of the engine
block and head are Cast Iron FC250 and aluminum alloy respectively. The required flatness of the two
mating surfaces are both 0.05mm. The assembled engine and two mating surfaces are shown in Figure
7. The two surfaces were measured by HDM equipment ShaPix3D® 3000 series, with a vertical
resolution of 0.05 µm and lateral resolution of 150 µm respectively. Its field of view is 280 mm × 280
mm, and the maximum number of sampling points is 4 million in each view. For workpieces whose
sizes exceed the field of view, the HDM could stitch multiple point clouds to generate the full view of
the surfaces. After pre-processing of the measurement data, the converted gray images which represent
the height map of the surfaces are shown in Figure 8.
By labelling algorithm introduced in Section 2.1, the feature points (centroids of the bores and
holes) are obtained. The labelling results are shown in Figure 9. And the assembly requirements of
these points are listed in Table 1.

4
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

Figure 7. The assembled engine and the two mating surfaces.

Figure 8. The height map of the mating surfaces.

Figure 9. Labelling results of the mating surfaces.

Table 1. Assembly relationship between surface features of cylinder head and block.
surfaces Label numbers
head 2 28 15 17 18 16 13 11 14 12 38 37 40 39
block 3 30 16 19 18 17 12 13 15 11 37 39 40 41
Using the control points registration algorithm proposed in Section 2.2, the feature points are
aligned into the same coordinate system. The magenta regions in Figure 10(a) are on the surface of the
block but not on the surface of the head. And the green regions indicate the opposite cases. Surface
gap is calculated according to the definition in Section 2.3 and shown in Figure 10(b).

5
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

Figure 10. Alignment of the two mating surfaces and the surface gap.

3.2. Gasket stopper height optimization

To optimize the stopper topography around the cylinder bores, the sealing gap around the bores are
extracted as Figure 11 shows. The sealing gap distribution around the four bores is plotted in Figure 12.
Some statistics about the four sealing gaps are calculated and listed in Table 2.

Figure 11. Stopper location around four bores. Figure 12. Four sealing gap distribution.

Table 2. Statistics about the four sealing gap topography.

Statistics of sealing gap Bore1 Bore2 Bore3 Bore4
Mean height(mm) 0.0535 0.0615 0.0646 0.0556
Range of height(mm) 0.0231 0.0204 0.0150 0.0285
Standard deviation of height(mm) 0.0041 0.0032 0.0027 0.0064
From Figure 12 and Table 2, some optimization advices could be drawn for the gasket design. The
average height of the bore 2 and bore 3 are 0.01mm higher than bore1 and bore4, so it is better to
design the stopper around bore2 and bore3 higher than bore1 and bore4 to achieve better load balance.
Another observation is that the height range of sealing gap around bore1 and bore4 are higher than
bore2 and bore3. So it is better to design the stopper topography varies in accordance with the gap
height distribution rather than a constant height of all the stoppers, especially for stoppers around
bore1 and bore4.

6
MEIE 2021 IOP Publishing
Journal of Physics: Conference Series 1983 (2021) 012116 doi:10.1088/1742-6596/1983/1/012116

4. Conclusions
This paper presents a novel surface alignment method to calculate the surface gap distribution based
on HDM. The alignment algorithm mainly consists of three steps: region labelling to find the feature
points, control points registration to align these feature points, and surface gap calculation. The case of
interface between cylinder block and head shows the effectiveness of the surface alignment method,
and some gasket optimization advices are given to improve the sealing performance of this interface.

Acknowledgements
Financial supports from National Natural Science Foundation of China (Grant No. 51535007,
51775343, 51875359) and Ministry of Education-China Mobile Research Foundation (CMHQ-JS-
201900003) are acknowledged.

References
[1] Capretta R and Ohigashi H 1995 Design methodology for automotive multi-layer steel cylinder
head gaskets SAE Technical Paper 950322
[2] Kestly M, Unseld G, Weiss A and Ludwig J 2003 MLS cylinder head gasket wave stopper SAE
Technical Paper 2003-01-0474
[3] Huang Z, Shih A J and Ni J 2006 Laser interferometry hologram registration for three-
dimensional precision measurements Journal of Manufacturing Science and Engineering
128(4) pp. 887-896
[4] Du S C, Liu C P and Huang D L 2015 A shearlet-based separation method of 3D engineering
surface using high definition metrology Precision Engineering 40 pp. 55-73
[5] Wang M, Shao Y P, Du S C and Xi L F 2015 A diffusion filter for discontinuous surface
measured by high definition metrology International Journal of Precision Engineering and
Manufacturing 16(10) pp. 2057-2062
[6] Shao Y, Wang K, Du S and Xi L 2018 High definition metrology enabled three dimensional
discontinuous surface filtering by extended tetrolet transform Journal of Manufacturing
Systems 49 pp. 75-92
[7] Du S C, Liu C P and Xi L F 2015 A selective multiclass support vector machine ensemble
classifier for engineering surface classification using high definition metrology Journal of
Manufacturing Science and Engineering 137 011003-1-15
[8] Du S C, Huang D L and Wang H 2015 An adaptive support vector machine-based workpiece
surface classification system using high-definition metrology IEEE Transactions on
Instrumentation and Measurement 64(10) pp. 2590-2604
[9] Wang M, Du S C and Xi L F 2015 Predicting machined surface topography based on high
definition metrology 15th IFAC Symposium on information control problems in
manufacturing INCOM pp. 1013-1017
[10] Shao Y, Du S and Xi L 2017 3D machined surface topography forecasting with space-time
multioutput support vector regression using high definition metrology 37th Computers and
Information in Enginnering Conference Cleveland Ohio, USA, p. V001T002A069
[11] Shao Y, Yin Y, Du S, Xia T and Xi L 2018 Leakage monitoring in static sealing interface based
on three dimensional surface topography indicator Journal of Manufacturing Science and
Engineering 140 p. 101003
[12] Shao Y, Yin Y, Du S and Xi L 2019 A surface connectivity based approach for leakage channel
prediction in static sealing interface Journal of Tribology pp. 1-32
[13] Ren J, Park C and Wang H 2018 Stochastic modeling and diagnosis of leak areas for surface
assembly Journal of Manufacturing Science and Engineering 140(4)
[14] Meng W, Xi L and Du S 2014 3D surface form error evaluation using high definition metrology
Precision Engineering 38(1) pp. 230-236