figure 1.1. schematics of the heart showing the path of
TRANSCRIPT
Left atrium
Right atrium
Mitral valve
Left ventricle
Tricuspid valve
Right ventricle
From the lungs
To the body
To the lungs
From the body
From the body
Right ventricle
Left ventricle
Left atrium
Right atrium
Pulmonaryartery
Aortic valve
Pulmonary valve
Figure 1.1. Schematics of the heart showing the path of blood flow.
Figure 1.2. Left heart pumping cycle (2 - 540).
Systole Diastole
Aorta
Leftventricle
Left atrium
120
80
Pressure (mm Hg)
40
0
Left ventriclevolume
Figure 1.3. The twitch and tetanus mechanical responses of a muscle (12 -557).
Twitch
Unfused tetanus
Tetanus
0 1 2
Forc
e
Time (seconds)
Figure 1.4. Shear Induced Damage of Erythrocytes (64 - 500).
Crit
ical
She
ar S
tres
s (N
/m2 )
Lethal Damage
1000
100
10
1
0.001 0.01 0.1 1 10 100 1000 10000 100000
Exposure Time (seconds)
Figure 1.5. Shear Induced Damage of Platelets (the studies that provided the data for this graph were conducted at room temperature, nominally 24OC) (64 -500).
Lethal Damage
Crit
ical
She
ar S
tres
s (N
/m2 )
1000
100
10
1
0.001 0.01 0.1 1 10 100 1000
Exposure Time (seconds)
Figure 1.6. Relationship between viscosity and rate of shear at different hematocrits, Hn. For human red blood cell suspensions in plasma at 25OC. Hn=volume of erythrocytes multiplied by 0.96 put in (74 - 494).
0.2 0.5 1 5 10 50 100 500 1000
70
60
50
30
20
10
0
40
Visc
osity
(Cen
tipoi
se)
Shear Rate (1/second)
Plane A
Diaphragm
Figure 1.7. Schematic of a sac-type blood pump. The direction of fluid flow is indicated with arrows.
Diaphragm ring
Figure 1.8. Schematic of the Penn State University sac-type blood pump.
Pusher Plate
Figure 1.9. Schematic of the Penn State University electric blood pump.
Diaphragm
Figure 1.10. Schematic a pneumatic blood pump used for tracer residence time studies.
Figure 1.11. Schematic of a transparent sac-type blood pump, in which fluid flow was studied by tracer particle visualization.
Concentric sinus Triple sinus Modified triple sinus
Figure 1.12. Three cross-section schematics of blood pump outflow housing designs.
Figure 1.13. Two axisymmetric, valveless, skeletal muscle ventricles.
Inlet Outlet
central rigid support of adjustable length
Latissimus dorsi cone with muscle acting longitudinally
Firgure . Propsed method of harnessing skeletal muscle to form the SMV.
Latissimus tendon attached to rib
Central support attached to rib
Latissimus dorsimuscle cone
Inlet conduit joined to left atrium
Outlet conduit joined to descending aorta
Firgure . Proposed placement of SMV.
Crit
ical
She
ar S
tres
s (N
/m2 ) Lethal Damage
1000
100
10
1
0.001 0.01 0.1 1 10 100 1000 10000 100000
Exposure Time (seconds)
Figure 2.10. Combination of figures 1.3 and 1.4 showing the levels of shear stress and exposure time that are likely to damage blood elements.
Lower half of bearing clamp
Superpump piston rod
Cylindrical casing
Supporting rod
Hole tapped in the top of the supporting rod for attachment of blood pump proto-types
Linear bearing
Push rods
Acrylic disc
Force transducer
Figure 3.1. The mechanical system for harnessing a servo-controlled linear actuator to pump proto-type ventricles, as a complete solid model and with a section cut-away.
PressureTransducer
Adjustable resistance
Compliance
Reservoir
DownstreamFlowProbe
Figure 3.2. Schematic of the mock, closed, circulatory loop showing a ventricle in its uncompressed and compressed states at positions A and B, respectively.
UpstreamFlowProbe
A
B
Flat compression plate
Direction of fluid flow
Valves
Bubbling of polyurethane
Figure 3.3. A polyurethane ventricle made by covering a wax mould with polyurethane, dissolved in DMAC.
(a)
Figure 3.4. (a) Silicone rubber ventricle manufactured by rotating silicone rubber as it cures inside a mould (b) in an oven.(b)
Air bubbles in upper ventricle surface
Molten polymer injection
Figure 3.6. Section taken through an injection mould intended for manufacturing hemispherical pumping chambers.
Figure 3.6. Teflon dipping bowl with a screw top lid for storingpolyurethane dissolved in DMAC. The large width of the bowl allows artificial ventricle sized geometries to be dipped in thesolution.
(a)
Figure 3.7. (a) A polyurethane ventricle made by gluing two half ventricle forms together with a solution of DMAC and polyurethane. (b) The male resin former and the female mould it was cast from.
(b)
Figure 3.8. The smooth female mould, in two sections, made from acetal copolymer, and a wax mould cast from it.
Section of female mould for upper ventricle surface
Bolts
Wax form cast from the female mould
Threaded rod imbedded in wax form
Section of mould for ventricle base and quarter toroidal surface
Inlet and outlet holes
(a)
Figure 3.9. (a) The base of a polyurethane ventricle produced by coating the inside walls of a female mould with polyurethane solution. (b) A view of the ventricle showing the tear along the seam.
(b)
Tear along seam.
Extension piece
Thin section f the two part female mould
Figure 3.10. Extension for the smooth female mould inlet and outlet holes.
Orientation A Orientation B
Wax male form
Threaded rod with one end embedded in wax mould
Polyurethane drips formed at the inlet and outlet holes
(a)
Figure 3.11. (a) Two orientations in which dipped wax moulds are dried. (b) Supporting method for melting wax out of a polyurethane ventricle.
Polyurethane ventricle
(b)
Glass beaker
Supporting block of acetal copolymer
Melted wax.
Figure 3.12. A polyurethane ventricle made by dipping a high melting point wax mould into polyurethane solution. The holes in the base are trimmed by hand.
Plan view of ventricle base
14
15
16
171819
20
21
22
23
24
25
26
Plan view of ventricle upper surface
1
2
3
4
5
6
7
89
10
11
12 13
Figure 3.13. Thickness measurement positions made on each manufactured polyurethane ventricle. The measurements are numerically labeled, by order of measurement.
0
100
200
300
400
500
600
700
800
900
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Measurement Position
Thic
knes
s (m
icro
ns)
Mean-SDMeanMean+SD
Upper ventricle surface
Base
Quarter toroidal join between the base and upper ventricle surface
Figure 3.14. Mean thickness measurements of six 4 dip polyurethane ventricles.
(a)350
Figure 3.15. Box plots of the upper surface thickness measurements for the six 4 dip ventricles. The range is plotted together with the 25th and 75th percentiles. The filled circle represents the mean in (a) and the median in (b).
1 2 3 4 5 6
Ventricle number
100
150
200
250
300V
entri
cle
thic
knes
s (µ
m)
1 2 3 4 5 6
Ventricle number
100
150
200
250
300
350
Ven
tricl
e th
ickn
ess
(µm
)
(b)
B1P1 B1P2
B2P1 B2P2
Figure 3.16. The four geometrical variations of proto-type blood pump made from combinations of B1 (base with straight pipes) and B2 (base with angled pipes) with P1 (flat compression plate) and P2 (domed compression plate).
Figure 4.1. Signal from the force transducer and meter, on an oscilloscope, after delivering an impulse of short duration to the compression plate.
Outlet Pipe
Inlet Pipe
Region ofinterest
Region for assessing background counts
8502 counts
0 counts
Figure 4.2. Totalised image of counts from within an artificial ventricle. The region of interest defines the area within which counts are summed for the calculation of ejection fraction. The relationship between grey scale and gamma ray counts is linear.
12000
10000
8000
Cou
nts
6000
4000
2000
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Time (seconds)
Figure 4.3. Gamma ray counts during one cycle inside a region of interest, for a 19.6mm stroke length at 40bpm.
150
-100
-50Flow
Rat
e (m 0
0.3 0.5
100
l/sec
)
50
0.1 0.7 0.9 1.1 1.3 1.5
Time (seconds)
Net flow between the major pumping phases.
-150
-200
Figure 4.4 Net fluid flow rate through a ventricle. (Flow out of the ventricle is positive.)
0
20
40
60
80
100
120V
entri
cle
Vol
ume
(ml)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Time (seconds)
Figure 4.5. Ventricle volume during one cycle derived from flow measurements up and down stream of a ventricle (19.6mm stroke length at 40bpm).
0
10
20
30
40
50
60
70
80
90
100
Per
cent
age
Cou
nts
2nd Pass1st Pass of tracer through ventricle
4 20 36 52 68 84 100 116 132 148 164 180 196
Time (seconds)
Figure 4.6. Clearance curve derived from gamma ray counts, within a region of interest, after the introduction of a Tc-99m bolus upstream of a ventricle (20.9mm stroke length at 30bpm).
Inlet Outlet
Region of interest
Figure 4.7. Captured image of dye inside the polyurethane ventricle at maximum ventricle volume, during a pumping cycle.
0
20
40
60
80
100
120
140
160
0 4
Ave
rage
Inte
nsity
(on
a 0-
255
grey
sca
le)
Figure 4.8. Average intensity, within a region of interest, after the introduction of an upstream dye bolus (20.9mm stroke length at 30bpm) plotted against time.
8 12 16 20 24 28 32 36 40 44 48
Time (seconds)
5
4.5
4
0
0.5
1
1.5
2
2.5
0
log(
Ave
rage
Inte
nsity
) 3.5
3
0.5 1 1.5
Dye Tracer (ml)
Figure 4.9. Logarithm of the average intensity plotted against incremental changes of, uniformly mixed, dye within the ventricle.
Per
cent
age
Dye
100
80
60
40
20
0
0 10 20 30 40 50
Time (seconds)Figure 4.10. Comparison of clearance curves produced by the pixel method , that takes account of ventricle geometry, and the average intensity method , (20.9mm stroke length at 30bpm).
Figure 4.11. Displacement of the linear actuator during one pumping cycle of the polyurethane ventricle, 18.3mm stroke length at 40bpm, at a flow rate of 1.98 litres/minute . Positive displacement values are measured when the compressionplate moves away from the ventricle, i.e. during filling. The velocity of the compression plate is derived from the displacement data. The force data is captured from a force transducer placed in series with the linear actuator piston.
-40
-20
0
20
40
60
80
0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5
Time (seconds)
Forc
e (N
ewto
ns),
Vel
ocity
(mm
/sec
),
Dis
plac
emen
t (m
m).
Force (N)Velocity (mm/sec)
Displacement (mm)
2.50
2.00
Pow
er (W
atts
)
1.50
1.00
0.50
0.000.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5
Time (seconds)
Figure 4.12. Graph of power necessary to actuate the compression plate while pumping at a flow rate of 1.98 litres/minute (18.3mm stroke length at 40bpm). This graph is in phase with the plot in figure 4.11. The power curve from 0 to 0.75 seconds corresponds to compression of the ventricle during ejection. 0.75 to 1.5 seconds corresponds to ventricle filling.
Tech
netiu
m E
ject
ion
Frac
tion
(%)
Figure 4.13. Comparison of ejection fractions measured with Tc-99m and EM flow meters at different flow rates in silicone rubber (r = 0.94, gradient = 1.02) and polyurethane ventricles (r = 0.99, gradient = 0.94).
Silicone rubber ventricle
Polyurethane Ventricle
Silicone Rubber ventricle fitted linePolyurethane Ventricle Fitted Line
40
35
30
25
20
15
10
5
00 5 10 15 20 25 30 35 40
Flow Meter Ejection Fraction (%)
Rad
ionu
clid
e C
oeffi
cien
ts
-0.2
-0.18
-0.16
-0.14
-0.12
-0.1
-0.08
-0.06
-0.04
-0.02
0
-0.2
Data
Fitted line
-0.18 -0.16 -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 0
Dye Coefficients
Figure 4.14. Comparison of exponential curves, y= aebx, fitted to the Tc-99m and dye clearance curves (fitted line: r = 0.997, gradient = 0.91).
B1P1
B1P2
B2P1
B2P2
0
0.5
1
1.5
2
2.5
3
0
Flow
Rat
e (li
tres/
min
ute)
5 10 15 20 25 30 35
Stroke length of linear actuator (mm)
Figure 5.1. Calibration of the four designs for flow rate against superpump stroke length. The r values for the least squares linear fitted lines are all above 0.999.
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
Vol
ume
of D
ye (m
l) = ± 1 standard deviation.
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Time (seconds)
Figure 5.2. The mean and standard deviation of ten clearance curves, for the same pumping chamber, at the same flow and pumping rate.
Figure 5.3. Process for converting a video frame of dye, inside a pumping chamber, to a colour map of dye concentration along each line of site through the pumping chamber.
Path length corrected image (grey scale corresponds, directly, to average concentration of dye along each line of site).Raw video image.
Background subtracted from raw video image. Colour scale applied to path length corrected image.
Figure 5.4. Path lengths as a grey scale image, with a linear scale from white (grey value of 255) to black (grey value of 0) corresponding to a scale of 0mm to 56mm path length.
Figure 5.5. A typical calibration image after the background effects have been subtracted. 0.9 ml of dye, uniformly concentrated, present in a ventricle.
0.6
0.5
Figure 5.6. Resolution curve showing the relationship between the amount of dye along the path length at a pixel (K) and the measured grey scale intensity of the pixel (Lp).
0.4
0.2
0.3
K(m
l/cm
2 )
0.1
05 55 105 155 205 255
Lp (grey scale)
Raw video image.
Figure 5.7. Enhancement of a raw video image. The inflow jet is visualized with dye.
Enhanced image.
-200
-150
-100
-50
0
0.5 1 1.5 2
Time (seconds)
Flow
Rat
e (m
l/sec
ond)
B1P1
B1P2
B2P1
B2P2
200
150
100
50
-250
Figure 5.8. Resultant flow curves for the four designs at a flow rate of 2.1 litres/minute. Outflow is posivitve and inflow negative.
200
0.5 1 1.5 2
Time (seconds)
Resultant Flow Rate (ml/second)
Force (Newtons)
B
150
100
50
0
-50
-100
-150
-200
A-250
Figure 5.9. Comparison of force and flow curves for B2P2, flow rate of 2.0 litres/minute at 30bpm.
3
B1P1
B1P2
B2P1
B2P2
2.5
2
Pow
er (W
atts
)
1.5
1
0.5
0
0 0.5 1 1.5 2
Time (seconds)
Figure 5.10. Power curves for the four designs at a flow rate of 2.0 litres/minute, pumping at 30bpm. The curves from 0 to 1 seconds represent power expended in pumping fluid by compressing an artificial ventricle, while from 1 to 2 seconds represents power imparted to the compression plate by in-flowing fluid and back pressure.
10
20
30
40
50
Per
cent
age
Dye B1P1
B1P2
B2P1
B2P2
100%
100
90
80
70
60
00 10 20 30 40 50 60
Time(seconds)
Figure 5.11. Clearance curves for the 4 geometries at a flow rate of 1.0 litres/minute and a 30 bpm pumping rate. The characteristic quick filling of the pumping chambers is evident, followed by a decreasing exponential emptying.
Per
cent
age
Dye
B1P1
B1P2
B2P1
B2P2
100%
100
10
1
0.10 5 10 15 20 25 30
Time (seconds)
Figure 5.12. Clearance curves for the 4 geometries at a flow rate of 1.0 litres/minute and a 30 bpm pumping rate.
Per
cent
age
Dye
B1P1B1P2B2P1B2P2100%
100
10
1
0.10 2 4 6 8 10 12 14 16 18
Time (seconds)
Figure 5.13. Clearance curves for the 4 geometries at a flow rate of 1.55 litres/minute and a 30 bpm pumping rate.
Per
cent
age
Dye
B1P1
B1P2
B2P1
B2P2
100%
100
10
1
0.10 2 4 6 8 10 12 14
Time (seconds)
Figure 5.14. Clearance curves, on a log scale, for the 4 geometries at a flow rate of 2.1 litres/minute and a 30 bpm pumping rate.
Figure 5.15. Clearance curves for the two P1 geometries at a flow rate of 2.9 litres/minute and a 30 bpm pumping rate.
B1P1
B2P1
100%
Per
cent
age
Dye
100
10
1
0.1
0 2 4 6 8 10
Time (seconds)
0.037 0 0.046 0
B1P1 B1P2
0.03 0 0.03 0
B2P1 B2P2
Figure 5.16a. Linear scales of dye concentration (ml of dye / cm3) against colour for the quantitative images for each pumping chamber geometry.
Figure 5.16b. Images of dye concentration, at the instant of maximum pumping chamber volume, over consecutive cycles..
B1P1 B1P2
B2P1 B2P2Cycle 1
B1P1 B1P2
B2P1 B2P2
Cycle 2
B1P1 B1P2
B2P1 B2P2
Cycle 3
B1P1 B1P2
B2P1 B2P2
Cycle 4
B1P1 B1P2
B2P1 B2P2
T0
B1P1 B1P2
B2P1 B2P2
T1
B1P1 B1P2
B2P1 B2P2
T2
B1P1 B1P2
B2P1 B2P2
T3
Figure 5.17. Consecutive video images (40 msec exposure) of a dye bolus entering each pumping chamber geometry.
Figure 5.18a. Conveyor belt (Mode a) and 100% mixing (Mode b) clearance curves at 50% ejection fraction.
010
2030
405060
7080
90100
Per
cent
age
Dye
Mode (a)Mode (b)
0 1 2 3 4 5 6 7
Cycle
15.7%
26.1%3
6.1%
48.3%
56.2%
65.4%
Figure 5.18b. Schematic of the artificial ventricle showing the regions within which apparent ejection fraction is calculated. The %standard deviation, from the mean, of each region’s apparent ejection fraction is shown, below the region number (n=5).
Figure 5.18c. Apparent ejection fractions for each design in six regions, at 2.1 litres/minute. The regions are defined and labeled in the schematic of figure 13b.
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
Region
1
Region
2
Region
3
Region
4
Region
5
Region
6
App
aren
t Eje
ctio
n Fr
actio
ns
B2P2
B2P1
B1P2
B1P1
Viewing tank
Blood pump
Figure 6.1. Three dimensional image of a proto-type blood pump suspended inside a viewing tank. Laser light is incident perpendicular to the orientation of the camera.
(a) Inlet Outlet
(b) Inlet Outlet
Figure 6.2. Fluid flow inside B2P1 recorded with single shot camera on 400ASA film at an exposure of 50msec and widest aperture (f=4.5), laser at maximum power output (5W). (a) flow during filling and (b) during ejection.
Figure 6.3. Fluid flow inside B2P1 visualized with polystyrene (above) and fluorescent (below) tracer particles, at a flow rate of 2.1 l/min at 30bpm. The fluorescent particles were made in the laboratory. The image grey scales have been reversed, so particle streaks are black rather than white.
S1S2 S3
Inlet
Outlet
Figure 6.4. The schematics of two views of the ventricles show the positions of the sections (S1, 2.. 8) as viewed in figures 6.6, 6.7, 6.8 and 6.9. (a) The 3 two-dimensional sections in which fluid flow is studied parallel to the long axis of the ventricle, selected with laser light sheets S1, S2 and S3. (b) The 5 two-dimensional sections in which fluid flow is studied, selected with laser light sheets S4, S5, S6, S7 and S8.
(a)
S1 S2S3
S4 S5 S6 S7 S8
(b)
Inlet Outlet
S4 S5 S6 S7 S8
Inlet Outlet
streamline
Figure 6.5. Key to flow pattern symbols used for figures 6.6, 6.7, 6.8 and 6.9.
small vortices (eddies)
fluid tracer particles either stationary or passing through the plane of the light sheet
slow vortex
streamlines representing slower velocity particles
fluid tracer particles either virtually stationary or passing through the plane of the light sheet, but exhibiting some slow movement.
small slow vortices
Figure 6.6. Fluid flow in B1P1 (a) at frame 10, (b) at frame 16, (c) at frame 24 and (d) at frame 40.
S1
S2
S3
S4
S5
S6
S7
S8
Inlet Outlet
S1
S2
S3
S4
S5
S6
S7
S8
Inlet Outlet
Inlet Outlet
Figure 6.6a.
Figure 6.6b.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.6c.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.6d.
Figure 6.7. Fluid flow in B1P2 (a) at frame 10, (b) at frame 16, (c) at frame 24 and (d) at frame 40.
Figure 6.7a.
Inlet Outlet
Inlet Outlet
S1
S2
S3
S4
S5
S6
S7
S8
Inlet Outlet
Figure 6.7b.
S1
S2
S3
S4
S5
S6
S7
S8
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.7c.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.7d.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.8. Fluid flow in B2P1 (a) at frame 10, (b) at frame 16, (c) at frame 24 and (d) at frame 40.
Inlet Outlet
S1
S2
S3
S4
S5
S6
S7
S8
Inlet Outlet
Inlet Outlet
Figure 6.8a.
Figure 6.8b.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.8c.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.8d.
Figure 6.9b.
S1
S2
S3
S4
S5
S6
S7
S8
Inlet Outlet
Inlet Outlet
Figure 6.9a.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.9. Fluid flow in B2P2 (a) at frame 10, (b) at frame 16, (c) at frame 24 and (d) at frame 40.
Inlet Outlet
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.9c.
S1
S2
S3
S4
S5
S6
S7
S8
Figure 6.9d.
CCD camera
Compliance chamber
Reservoir
Pump
Direction of flow
Line generator
Viewing tank
Light sheet
Figure 7.1. Rendered solid model of the experimental flow circuit for producing developed laminar flow in a cylindrical pipe, and capturing illuminated fluid tracer particles on video.
(b)
(a)
Figure 7.2. The steady, developed, velocity profile in a pipe calculated by fitting a quadratic function to velocity measurements made by measuring (a) the distance between dots and (b) the length of streaks.
0
Vel
ocity
(mm
/sec
)
60
20
40
-20 -10 0 10 20
Radial position (mm)
-20
0
20
-20
Vel
ocity
(mm
/sec
)
60
40
-10 0 10 20
Radial position (mm)
Data points.Quadratic fitted to data points.Velocity profile calculated from equation 7.1.
Area 2Area 1
Area 3
Figure 7.3. Images of particles in design B1P1 in section S1. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 1 Area 2
Area 3 Area 4
Figure 7.4. Images of particles in design B1P1 in section S5. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 2Area 1
Area 3 Area 4
Area 6Area 5
Figure 7.5. Images of particles in design B1P2 in section S1. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 2Area 1
Area 4Area 3
Area 5
Figure 7.6. Images of particles in design B1P2 in section S5. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 1 Area 1
Area 1
Figure 7.7. Images of particles in design B2P1 in section S1. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 2Area 1
Area 3 Area 4
Figure 7.8. Images of particles in design B2P2 in section S1. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Area 1 Area 2
Area 4Area 3
Figure 7.9. Images of particles in design B2P2 in section S5. The light sheet has been pulsed to show particle paths as dots. The black rectangles are the areas in which fluid shear stresses have beenmeasured more accurately by zooming the camera in on them.
Uniform inlet velocity profile 40mm
upper solid boundary
Figure 7.10. Schematic of the numerical fluid model of a submerged jet impacting a plate.
30 mmExit
lower solid boundary
50mm
Axis of rotational symmetry
Figure 8.1. A potentially improved inlet/outlet pipe configuration for the proposed skeletal muscle powered VAD. The light grey outlet pipe is behind the inlet pipe.
Figure 8.2. Curved tube type pumping chamber.
Figure 8.3. A pumping chamber based upon the shape of the natural ventricle in a human heart.