Combustion and Flow Measurement in PistonCylinder Assemblies Harold Schock

1
Combustion and Flow Measurement in
Piston-Cylinder AssembliesHarold Schock
Michigan State University
Mechanical Engineering
ME444 Fall 2007
2
In-Cylinder Process Quantification

• State measurement (average /local) temperature,
  pressure
• Air-flow quantification
• Swirl meter
• Tumble meter
• Laser diagnostics
• Species quantification
• Laser induced fluorescence
• Sampling valves (helenoid mass spectrometer)
• Flame ionization detector
• Chem luminescence detector

3
SAE 2000-01-2798The 3-D In-Cylinder Charge
Motion of aFour-Valve SI Engine under Stroke,
Speed and Load Variation

• Hans G. Hascher, Mark Novak Tom Stuecken and
  Harold J. Schock
• Engine Research Laboratory
• Michigan State University, East Lansing, MI
• James Novak
• Powertrain Operations
• Ford Motor Company, Dearborn, MI

International Fall Fuels and Lubricants
Baltimore, Maryland October 16-19, 2000
4
Outline

• Engine and measurement system
• Mean velocity vector fields
• Turbulent kinetic energy
• Isosurfaces
• Average
• Planar distribution
• Tumble and swirl ratios
• Fixed vs. moving origin
• Planar
• Conclusions

5
Engine and Measurement System
6
Model and Make Ford 4-Valve 4.6L Ford
4-Valve 5.4LBore and Stroke 90.2 mm / 90.0
mm 90.2 mm / 105.4 mmConnecting Rod Length
150.7 mm 170.0 mmValve Activation
DOHC DOHCIntake Valve Diameter 37.0
mm 37.0 mmExhaust Valve Diameter 30.0
mm 30.0 mmMaximum Valve Lift 10.02 mm at
120 CAD 10.02 mm at 120 CADZero CAD Intake
TDC Intake TDCIntake Valve Opening 6 CAD
Before TDC 6 CAD Before TDCIntake Valve
Closure 250 CAD After BDC 250 CAD After
BDCExhaust Valve Opening 126 CAD After TDC 126
CAD After TDCExhaust Valve Closure 16 CAD After
TDC 16 CAD After TDCCompression Ratio 9.85
1 9.40 1Piston Top Flat Flat
Prototype Engine Specifications
7
Schematic of the LDV Setup for 3-D Measurements
8
Coordinate System inside the Cylinder Volume
with Ten Measured Planes
9
Port Geometry with Pentroof CombustionChamber at
BDC for the 4.6L Setup
10
Calculations Grid for 127 Measurement Locations
in One Horizontal Slice
11
Mean Velocity Vector Fields
12

• Mean Velocity Comparison in the Two Center Planes
  
• for the 4.6L Setup WOT, between 600 rpm
• and 1500 rpm at 130 CAD

13
Mean Velocity Comparison in the Two Center Planes
for the 5.4L Setup WOT, between 600 rpm and
1500 rpm at 130 CAD
14
Mean Velocity Comparison in the Center Planefor
the 4.6L Setup Part Throttle Conditions, between
600 rpm and 1500 rpm at 130 CAD
15
Mean Velocity Comparison in the Two Center
Planes for the 4.6L Setup WOT, between 600 rpm
and 1500 rpm at 243 CAD
16
Mean Velocity Comparison in the Two Center
Planes for the 5.4L Setup WOT, between 600 rpm
and 1500 rpm at 243 CAD
17
Mean Velocity Comparison in the Center Plane

for the 4.6L Setup Part Throttle Conditions,

between 600 rpm and 1500 rpm at 243 CAD
18
Mean Velocity Distribution at z 45mm in the
Main Tumble Plane for the 5.4L Setup, at 600
rpm and 1500 rpm, and a Crank Angle of 130
Degrees
19
Turbulent Kinetic Energy

• Isosurfaces
• Average
• Planar Distributions

20
Turbulent Kinetic Energy
21
Isosurfaces of TKE at 75 CAD for 4.6L, WOT at 600
rpm (left), IS-Level 18.0 m2/s2 and for 4.6L,
Part Throttle at 1500 rpm (right), IS-Level
90.0 m2/s2 (scaled to 600 rpm 14.4 m2/s2). J/kg
m2/s2
22
Isosurfaces of TKE 5.4L, WOT at 600 rpm at 298
CAD (left), IS-Level 3.8m2/s2 and at 313 CAD
(right), IS-Level 3.0m2/s2. J/kg m2/s2
23
Specific TKE for Nine Measured Planes for the
4.6L, Part Throttle Setup at 1500 rpm (TKE Axis
Scaled to Lower Engine Speed)
24
z
z
x
TDC
y
x
y
BDC
Cylinder Coordinate System
with Moving Origin about the Instantaneous
Center of the Volume
Cylinder Coordinate System
with Fixed Origin at the Top Dead Center
25
Angular momentum per unit mass can be written as
Hence, the angular momentum around the principal
axis is
Using the moment of inertia around the principal
axis,
The tumble ratio per specific CA can be defined
26
x-Tumble Ratios around the Moving Origin for All
Six Engine Setups
27
x-Tumble Ratios around the Fixed Origin for all
Six Engine Setups
28
x-Tumble Ratios around the Fixed Origin for the
4.6L WOT Setup at 600 rpm
29
Conclusions

• The measured velocities and flow patterns are
  mildly sensitive to the bore/stroke ratio
  studied, flow velocities scale with piston speed
  changes, and flow patterns diverge significantly
  under the changes in the throttle conditions
  investigated.
• Normalization of total turbulent kinetic energy
  to specific turbulent kinetic energy can obscure
  important aspects of the flow, such as the
  increase in total TKE late in compression for the
  5.4L setup, WOT, 1500 rpm.

30
Conclusions

• A similar sickle-shaped, TKE-isosurface
  structure appeared in almost all engine setups,
  at and after BDC. Exception Part throttle at
  600 rpm.
• The fluctuating part of the measured
  velocities, calculated as TKE, after BDC,
  exhibited similar exponential decay, between
  different four-valve engine setups and intake
  configurations.
• Due to the absence of counter-rotating vortices
  in this study, tumble ratios did detect major
  changes in the mean flow. Three dimensional
  planar tumble ratio plots vs. crank angles
  provide a significant improvement in flow
  description compared to the bulk two-dimensional
  plots.

31
Conclusions

• Although the general shape of a bulk tumble ratio
  curve vs. crank angle may be independent of the
  calculation center, significant differences in
  important regions can occur based on the
  reference chosen.
• Mean velocity collision regions contribute to the
  increase of TKE.
• The 5.4L configuration exhibited significant TKE
  gradients late in compression.

32
SAE 2000-01-2799A Comparison of Modeled
and Measured 3-D In-Cylinder
Charge Motion Throughout the Displacement of a
Four-Valve SI Engine
Hans G. Hascher and Harold J. SchockEngine
Research Laboratory

Michigan State University, East Lansing,
Michigan Oshin Avanessian and James Novak
Powertrain Operations

Ford
Motor Company, Dearborn, Michigan
International Fall Fuels and Lubricants
Baltimore, Maryland October 16-19, 2000
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
Modeled and Measured Mean Velocity Distribution
within the Main Tumble Plane for the 4.6L, WOT
Setup at 75 CAD during Intake Stroke
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
Conclusions

• The modeled and measured mean flow patterns match
  well in amplitude and direction for the
  investigated engine setup.
• Differences in the mean flows occurred mostly in
  the outer regions of the cylinder, away from the
  main tumble plane. The flow differences in the
  off-center planes indicate the need for
  additional modeling effort. In multi-cylinder
  engines, the modeling should include evaluation
  of wave dynamics, associated with the intake and
  exhaust and other cyclic phenomena that can
  influence in-cylinder flows.
• Prediction of tumble showed reasonable agreement
  with the measured results. Investigation for
  other geometries would allow one to determine if
  tumble ratio vs. crank angle is a useful
  descriptor of engine flows.

47
Conclusions, continued

• The prediction of turbulent kinetic energy
  matches well after BDC, but exhibits significant
  differences during the intake stroke, both in
  amplitude and location. Improved turbulence
  models and higher grid resolution clearly need to
  be implemented if one is to resolve the
  significant flow details.
• For engines with a similar geometry to the one
  studied, simulated flows using current RANS
  models to represent internal flows can predict
  general flow features. However, they are unlikely
  to be useful for prediction of heat transfer,
  ignition events, flame propagation rates or
  stratified charge engine performance.

48
(No Transcript)
49
Delayed Undelayed Images
Undelayed Image (crankangle
90 degrees)
Delayed Image
(crankangle90 degrees, delay70µs)
50
Test Rig for Flow Control Experiments
51
Open Port
52
Tumble Plate
53
Swirl Plate
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
CONFIGURATION A
63
CONFIGURATION B
64
Summary and Conclusions

• MTV technology provides a revolutionary new tool
  for quantifying engine combustion chamber flows
• Analysis of approximately 400 cycles is necessary
  to determine averages which can be used for
  comparative evaluations (for this engine
  configuration)
• Cycle-to-cycle variability requires that a
  statistical description of combustion chamber
  flows be evaluated
• CMCVs have a profound influence on the
  flowfield, even to late compression

65
Summary and Conclusions

• Nature of the CMCV influence depends on the
  geometry of the CMCV as well as the induction
  system each new geometry is likely to produce
  different flows
• The valve timing difference studied also had an
  influence on the flowfield to late compression
• Circulation PDFs comparisons showed considerable
  differences for the cases studied
• Other FM PDFs descriptions should be examined
  and the results compared to combustion events