Paul C. Miles

1
The effect of swirl and injection phasing on flow
structures and mixingin an HSDI diesel engine

• Paul C. Miles
• Sandia National Laboratories
• Leif Hildingsson, Anders Hultqvist
• Lund Institute of Technology

Co-sponsors
Swedish Energy Agency
US Dept. of Energy Office of Freedom Car
Vehicle Technology
THIESEL 2006 13-15 September 2006, Valencia
2
Background

• Research towards low NOx and soot has led to a
  number of combustion concepts.
• Some of these concepts are
• PREDIC Premixed lean diesel combustion
• PCCI Premixed charge compression ignition
• UNIBUS Uniform bulky combustion
• PCI Premixed compression ignited combustion
• Toyota, AVL injection closer to TDC, much EGR
• MK Modulated Kinetics

3
Motivation

• In MK and the Toyota/AVL concepts, in-cylinder
  mixing processes are important, due to
• high EGR levels
• shorter time available compared to PREDIC and
  similar concepts

Injection and combustion phasing are different
for these concepts, this will influence bulk-flow
structures which have great impact on transport,
mixing and oxidation of POF and soot.
Perform planar measurements, PIV, to provide
information. How do these flow structures change
with injection and combustion phasing and swirl
ratio?
4
Engine facility Volvo D5

• Engine
• 4 valves
• Bore 81 mm stroke 93.2 mmDisplacement 480cc
• Compression ratio (non-mod engine) 17.31

• Fuel injection
• Common rail, central injector
• 5 hole, 140
• Injection pressure 750 bar

• Optical piston / Bowl geometry
• Squish height 2.2 mm
• Top ring land 40 x 0.5 mm
• Optical engine compression ratio 13.751

5
Operating conditions
Toyota/AVL-like O2 12 (56 EGR) ?
1.15 SOI - 12.5 ATDC CA50 5 ATDC
MK-like O2 15 (40 EGR) ? 1.50 SOI -
4.0 ATDC CA50 18 ATDC
Speed 1200 RPM Load 4.0 bar IMEP Fuel 70
n-heptane 30 iso-octane Swirl two levels, 2.0
2.6
6
Set-up

• 1280 x 1024 sensor, 12 bit resolution
• 50 mJ/pulse of 532 nm laser energy
• 1/e2 sheet thickness 0.25 mm
• 532 nm bandpass filter mechanical shutter to
  block soot luminosity
• 2.0 2.5 µm SiO2 t15 µsIntroduced with a
  modified TSI 3400 bronze-bead fluidized bed

7
PIV processing

• Multi-pass, finishing with 32 x 32, 50 overlap
• Distortion correction via 2-d cubic polynomial
• We correct for distortion and analyze the bowl
  and squish region separately...
• then combine
• Bowl curvature provides unexpected
• advantages
• A clear image of the dot target is obtained to
  within 1 mm of the piston top
• The flow is most accessible in the near lip
  region

8
Systematic error due to viewing angle

• The typical normal (swirl) velocity is large (
  2 Sp ), hence this error is of order Sp / 3 (1.2
  m/s)

Also, randomly distributed false vectors will
influence velocity magnitude
9
Heat releases

• MK-like
• O2 15 (40 EGR)
• SOI - 4.0 ATDC
• CA50 18 ATDC

Toyota/AVL-like conventional timing O2 12
(56 EGR) SOI - 12.5 ATDC CA50 5 ATDC
10
Fuel jet setting up vortex
Late injection SOI -4

• Conventional injection timing
• SOI -12.5

Fuel jet sets up a standing vortex, clearly
visible before main heat release Peak velocities
around 4 Sp
11
Premixed combustion conv. timing

• Strong vortex structure
• still present

Development of reverse-squish flow Vortex near
pip develops
12
Mixing-controlled combustion conv. timing

• Vortex formed at centreline moves
• outwards
• imparts looping motion on reverse-squish flow
• fluid entering squish region comes from uppermost
  regions of the cylinder

13
Reversal of bowl flow direction conv. timing

• The initial clockwise motion has evolved
• into a counter-clockwise movement

14
Late mixing controlled burn conv. timing

• Vortex formed in centre still present, now at
  bowl mouth
• Second vortex formed at centreline
• Upward looping reverse-squish motion still present

15
Late-cycle burn conv. timing

• Overall dominating flow is downwards
• After 45 CAD the looping structure dissolves to
  great extent
• Vortex formed at cylinder edge
• Probably interaction between downward flow and
  outflow from top ring-land crevice

16
Premixed combustion, late timing

• Some flow directed into squish volume apparent at
  10 CAD
• Reverse-squish motion stronger than for
  conventional injection timing
• Transport into squish volume mainly from mid-part
  of bowl for conv. timing mostly from outer
  parts

17
Mixing-controlled combustion late timing

• Mushroom-shaped structure formed
• transports fluid both inwards and outwards
• Evolves into looping motion at rim, forms full
  vortex

18
Toroidal vortex, in moving reference frame

• Late timing

Conventional timing
Toroidal vortex visualized by subtracting
piston-motion induced velocities Late injection
timing produces a more energetic vortex
19
POF trapped in vortex
Conventional timing

• Partially oxidized fuel trapped within this vortex

20
Late-cycle burn late timing

• No flow structures that might enhance mixing in
  central region
• After 45 CAD toroidal vortex dissipates, looping
  motion persist through 70 CAD
• No clear vortex formed at cylinder periphery

21
Influence of swirl

• Late timing
• Higher swirl

Late timing Lower swirl
Main difference is the fluid motion entering the
squish region. At higher swirl, this motion has a
dominant vertical component and evolves into a
toroidal vortex by 30 CAD. At lower swirl, fluid
trajectory is much flatter and the formation of
the toroidal vortex is delayed until about 35-40
CAD.
22
Summary and conclusions I

• Employing PIV to measure vertical-plane flow
  structures in a fired diesel engine has been
  demonstrated
• Geometric distortion can be corrected
• Fuel injection event sets up a clockwise-rotating
  vortex in the bowl region. Peak velocities in
  this flow are up to about 4 times the mean piston
  speed.
• The trajectory and timing of the reverse squish
  flow is found to be an important factor
  influencing the bulk flow motions which likely
  influence the subsequent mixing of soot and
  partially-oxidized fuel with unused oxygen
• Peak reverse squish flow is found to coincide
  closely with the timing of the peak heat release

23
Summary and conclusions II

• For the higher swirl cases, the location of the
  peak maximum in the reverse squish flow are
  different for the two operating conditions
• Conventional timing has maximum close to bowl rim
• Later timing has maximum closer to bowl
  mid-radius
• Vortex appears near bowl rim just after start of
  mixing-controlled combustion
• In late injection case it forms an easily
  recognized toroidal vortex
• For conventional timing the vortex is not so
  easily seen
• The rapid inward motion at the vortex seen with
  late injection may transport soot and POF inwards
  at a higher rate than with conventional injection
  timing
• The more energetic vortical motion may trap POF

24
Summary and conclusions III

• Flow structures measured near the cylinder
  centreline suggest that combustion products will
  have very little radial motion, thus will
  probably not mix well with available O2
• Looping structure at bowl rim is flatter at lower
  swirl and the formation of toroidal vortex is
  delayed
• probably due to changes in the location of heat
  release
• Beyond 45 CAD there are no notable differences in
  the flow structures with different swirl ratios
  for either the conventional or the late injection
  case.