Artificial Organs

Validation of an Axial Flow Blood Pump Computational Fluid

Dynamics Results Using Particle Image Velocimetry

Journal: Artificial Organs

Manuscript ID: AO-00259-2010.R2

Fo

Manuscript Type: Main Text

Computational Fluid Dynamics, Axial Flow Blood Pump, Particle Image

Keywords:
r
Velocimetry
Re
vi
ew
On
ly
Page 1 of 24 Artificial Organs

1
2
3
4
5
6 Validation of an Axial Flow Blood Pump Computational
7
8 Fluid Dynamics Results Using Particle Image Velocimetry
9
10 Su, Boyang (contact); Chua, Leok Poh; Wang, Xikun

11
12
13
14
15
16
17
18
Fo
19
20
21
22 Boyang Su
r

23
School of Mechanical and Aerospace Engineering
24
Re

25 Nanyang Technological University

26
27 bysu@ntu.edu.sg
28
vi

29
30
31
ew

32
33
Received: November 2, 2010 Formatted: Left

34
35
36 Revised: July 5, 2011
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55 1
56
57
58
59
60
 Artificial Organs Page 2 of 24

1
2
3
4
5
6
7
8
9
10 Abstract
11
12
13 A magnetically suspended axial flow blood pump is studied experimentally in this
14
15 paper. The pump casing enclosed a three-blade straightener, a two-blade impeller
16
17 shrouded by a permanent magnet embedded cylinder, and a three-blade diffuser. The
18
Fo
19 internal flow fields were simulated earlier using Computational Fluid Dynamics
20
21 (CFD), and the pump characteristic curves were determined. The simulation results
22
r

23 showed that the internal flow field was basically streamlined, except the diffuser
24
Re

25 region. Particle Image Velocimetry (PIV) measurement of the 1:1 pump model was
26
27 conducted to validate the CFD result. In order to ensure the optical access, an acrylic
28
vi

29 prototype was fabricated with the impeller driven by a servomotor instead, since the
30
31 magnet is opaque. In addition to the transparent model, the blood analog fluid with the
ew

32
33 refractive index close to that of acrylic was used so as to avoid refraction. According
34
35 to the CFD results, the axial flow blood pump could generate adequate pressure head
36
On

37 at the rotating speed of 9,500 rpm and flow rate of 5 L/min, and the same flow
38
39 condition was applied during the PIV measurement. Through the comparisons, it was
40
41 found that the experimental results were close to those obtained by CFD and had thus
ly

42
43 validated the CFD model, which could complement the limitation of the measurement
44
45 in assessing the more detailed flow fields of the axial flow pump.
46
47
48
49
50
51
52
53
54
55 2
56
57
58
59
60
Page 3 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9 Introduction:
10
11
12 Each year, approximately 30,000 patients at risk of dying from severe heart failure are
13
14 awaiting heart transplant, but only around 3,500 cases are performed (1). It is mainly
15
16 constrained by the limited heart donors available. Ventricular Assist Devices (VADs)
17
18 have been developed for decades as an alternative therapeutic modality and can be
Fo
19
20 produced in the quantity needed. VADs are used for bridge-to-transplantation, bridge-
21
22 to-recovery, or even long-term cardiac support (2, 3). The continuous flow VADs
r

23
24 usually include a rotating impeller to generate adequate flow rate and pressure head,
Re

25
26 so are also named as rotary blood pumps. Typically, the rotating speed of a centrifugal
27
28 blood pump is 2,000~4,000 rpm (4-7), and that of an axial blood pump is
vi

29
30 6,000~12,000 rpm (8-11). The hemolysis induced by a rotary blood pump is the main
31
ew

32 concern besides the hydraulic performance (12). It is found that the hemolysis is
33
34 significantly related to the flow behaviour, and the scalar shear stress and exposure
35
36 time are two main parameters used to estimate the hemolysis (13). Therefore, the
On

37
38 rotary blood pump poses a challenge, especially for axial flow blood pump with
39
40 higher rotating speed.
41
ly

42
43 Computational Fluid Dynamics (CFD) has been widely used as a powerful tool to
44
45 investigate many blood contacting medical devices including blood pumps (14, 15).
46
47 With the aid of CFD, the design phase can be shortened significantly from several
48
49 years to several months (16). It can also support and explain experimental results.
50
51 Compared with the traditional design, CFD can predict the detailed flow fields, pump
52
53 characteristics curves and other parameters such as hemolysis (12, 17). However, the
54
55 3
56
57
58
59
60
 Artificial Organs Page 4 of 24

1
2
3
4
5
6 numerical results still need validating using the nonintrusive measurements, though
7
8 they are usually difficult and limited by the optical access. For the whole-flow-field,
9
10 Particle Image Velocimetry (PIV) technique is widely adopted (18-20), as compared
11
12 with the point-wise hot wire (21) or Laser Doppler Anemometer (LDA) (22, 23). The
13
14 internal flow seeded with particles is exposed by double frame laser sheets, and the
15
16 velocity vector map is calculated using a cross-correlation technique in PIV
17
18 measurement.
Fo
19
20
21 Due to the complexity of the structure of an axial flow blood pump, which usually
22
r

23 includes a straightener, impeller and diffuser, and its high rotating speed, there are
24
Re

25 only few studies using PIV to measure its internal flow (14, 20) as compared with the
26
27 centrifugal blood pump (18, 19, 24-26). Therefore, in this study, the axial flow blood
28
vi

29 pump was analyzed by both CFD and PIV techniques, and both numerical and
30
31 experimental results were compared for validation.
ew

32
33
34 Methods and Materials
35
36
On

37
Physical Model
38
39
40
The axial blood pump comprises of three components: a three-blade straightener, an
41
ly

42 enclosed two-blade impeller and a three-blade diffuser as shown in Figure 1. Its

43
44 overall axial length and outer diameter of the inner components are 60 mm and 19
45
46 mm respectively. The impeller was the only rotating component enclosed inside a
47
48 cylinder, which was driven by the Lorentz force. As a result, the impeller and the
49
50 cylinder rotated simultaneously as a single component (27). Since the magnet was
51
52 installed in the rotor cylinder, there was a clearance gap between the rotor cylinder
53
54
55 4
56
57
58
59
60
Page 5 of 24 Artificial Organs

1
2
3
4
5
6 and the pump casing. The current diffuser has only three blades for the ease of
7
8 fabrication and PIV measurement.
9
10
11 CFD Method
12
13
14 The axial flow blood pump was generated using Computer-Aided Design (CAD)
15
16 software Pro/Engineer Wildfire 4.0 (Parametric Technology Corporation) and was
17
18 then exported into commercial grid generator software ICEM (ANSYS, Inc). The
Fo
19
20 numerical model consisted of three domains, i.e. straightener, impeller and diffuser,
21
22 with both straightener and diffuser prolonged to make the flow develop. The grid
r

23
24 independence test was conducted based on the pressure head across the blood pump
Re

25
26 as shown in Figure 2. It can be found that the pressure head difference with respect to
27
28 the finest model was less than 1% as the grid number of the numerical model
vi

29
30 increased to 1.3 million. Therefore, the model with 1.3 million grids was selected as a
31
ew

32 compromise between the accuracy and computational resource. In between the

33
34 clearance gap between the pump casing and the impeller shroud, 10 layers of grid
35
36 cells were generated in the selected model. The skew angles of these grids were
On

37
38 between 20 and 160 degrees with the y+ value of grids at near wall region maintained
39
40 around 1, in order to ensure the quality of grids. Figure 3 (a) shows the surface grids
41
ly

42 on the inner components and the longitudinal cross section with the complete
43
44 computational domain shown in Figure 3 (b).
45
46 The numerical simulation was conducted using the finite-volume CFD code FLUENT
47
48 6.3.26. The blood was assumed to be Newtonian with a viscosity of 0.0035 Pa·s and a
49
50 density of 1,050 kg/m3 (28). At the rotating speed of 9,500 rpm, the Reynolds number
51
52 based on the impeller tip was about 3.4×104, so the internal flow was turbulent. Shear
53
54
55 5
56
57
58
59
60
 Artificial Organs Page 6 of 24

1
2
3
4
5
6 Stress Transport (SST) k-ω turbulence model developed by Menter (29) was selected,
7
8 because it combines the advantage of k-ω and k-ε models in the near wall regions and
9
10 outside the boundary layers, respectively (26, 30, 31). In addition, it predicts better
11
12 results in adverse pressure gradients and separating flow (32). Sliding mesh model
13
14 was applied to accurately predict the rotor-stator interaction with 300 time steps in
15
16 one period (33). Velocity inlet and pressure outlet were applied at the pump inlet and
17
18 outlet, respectively. The static pressure head across the numerical model was used as
Fo
19
20 the macro parameter to monitor the convergence, and the iteration was stopped when
21
22 the convergence criterion of 10-4 was satisfied (34).
r

23
24
Re

25 Experimental Method
26
27
28 Figure 4 (a) shows the schematic drawing of the constructed experimental test rig.
vi

29
30 The pump model constructed is the exact size of the prototype in order to avoid any
31
ew

32 scaling problem that alters the results. When the pump operated, the flow rate was
33
34 controlled by the throttle valve located at the downstream of the flow meter, and the
35
36 pressure head across the pump model and the flow rate could be obtained from the
On

37
38 pressure transducer and the flowmeter, respectively.
39
40
In order to reduce the possibility of hemolysis, magnetic suspension technique is
41
ly

42
widely adopted in rotary blood pumps including the present study. However it is
43
44 impossible for the PIV measurement due to the opaque magnets. Therefore, the
45
46 impeller is driven by a servomotor through a shaft. The test model is partially
47
48 fabricated from acrylic for the illumination of the laser sheet and image pairs
49
50 capturing as demonstrated in Figure 4 (b). In order to minimize the vibration of the
51
52 operating impeller, a bearing holder, which is a hollow step cylinder, is designed as
53
54
55 6
56
57
58
59
60
Page 7 of 24 Artificial Organs

1
2
3
4
5
6 shown in Figure 4 (c). Its front end is designed to place the hollow diffuser in position
7
8 with the diameter of the diffuser hub equal to that of the bearing holder at Section B.
9
10 Note that the influences of the vibrations of the inner components such as impeller
11
12 and diffuser are not considered in this study. The straightener, impeller and diffuser
13
14 are all made of acrylic as well as the impeller shroud that is glued together with the
15
16 impeller. The impeller is driven by a steel shaft, which is inserted through the
17
18 bearings and bearing holder. These components are enclosed in the chamber, which is
Fo
19
20 capped by two components connected to the test rig.
21
22
r

23 The PIV system used in the present study was the Dantec FlowMap PIV system
24
(Dantec Measurement Technology, Denmark). The Gemini PIV 200-15 (New Wave
Re

25
26
27 Research, USA), a double cavity pulsed Nd:YAG (Neodymium Yttium Aluminum
28
Garnet) laser system was used as the illumination source. The two pulsing beams
vi

29
30
from a double-cavity Q-switched Nd:YAG laser are combined and formed into
31
ew

32
pulsing light-sheets using optics. Q-switched pulse can effectively freeze the motion
33
34
of particle, and two cavities ensure the widest possible ranges of time between pulses.
35
36
The CCD camera used in the FlowMap PIV system is HiSense MkII camera
On

37
38
(Hamamatsu C8484-05 digital CCD chip; resolution 1280×1024) equipped with
39
40
Nikon AF Micro-Nikkor lens 60/2.8. FlowManager software controlled the
41
ly

42 synchronization sequence to acquire data, and stored the recording data. The ‘once-
43
44 per-revolution’ signal from the shaft encoder was sent to the processor, which then
45
46 fired the laser and acquires the images from the PIV camera as shown in Figure 3 (a),
47
48 so images could be taken at certain angular position of the impeller. Moreover, the
49
50 timing signals are used to control the camera to make sure that the CCD sensor is only
51
52 exposed to the light from the illumination system, since the ambient light will reduce
53
54
55 7
56
57
58
59
60
 Artificial Organs Page 8 of 24

1
2
3
4
5
6 the signal-noise ratio. In order to measure the flow at different angular position, a
7
8 delay was set between the external signal and the image acquisition. Two successive
9
10 images are then transferred to the FlowMap processor and stored finally. As a result, a
11
12 pair of seeding particle images with time delay was captured at certain impeller
13
14 angular position, and the velocity vector map was derived from the movement of
15
16 particle during the time delay.
17
18
Fo
19 In order to avoid refraction, the working fluid involved mixing 40% of glycerine and
20
21 60% of aqueous ammonium thiocyanate (NH4SCN) solution by volume (35). Its
22
r

23 refractive index was 1.488 to match with that of the acrylic model with a density of
24
1,155 kg/m3 and a dynamic viscosity of 3.8 mPa·s. The kinetic viscosities of the
Re

25
26
27 working fluid and blood were 3.29 and 3.33 µm2/s respectively so as to maintain
28
Reynolds number similarity. The solution was seeded with neutral buoyant hollow
vi

29
30
glass spheres with the mean diameter of 10 µm. In the present PIV measurement, 100
31
ew

32
image pairs were captured at each measured area. The measurement area was
33
34
10.76×8.26 mm2 with the scale factor, S, of 1.25. The recording aperture, f#, was
35
36
5.6~8, and thus the camera image of seeding particle had a diameter 2~3 pixels under
On

37
38
the wavelength of 532 nm. The interrogation area of 32×32 pixels was used with 50%
39
40
overlap. The pulse duration of this PIV is 10 ns, so the particle movement during this
41
ly

42 time duration was negligible. In the straightener, the pulse separation varied 10~25 µs
43
44 at different measurement area according to the local flow velocity. In order to
45
46 minimize the effects of out-of-plane velocity component, the thickness of the laser
47
48 sheet was 0.5 mm, and the time between light pulses was 10 µs so as to satisfy the
49
50 equation below (36):
51
52
53
54
55 8
56
57
58
59
60
Page 9 of 24 Artificial Organs

1
2
3
4
5
6 vz ⋅ ∆t
7 ≤ 0.25
zl
8
9
10 where vz is the out-of-plane velocity; ∆t is the time between pulses; and zl is the
11
12 thickness of the laser sheet. Post-processing validation methods including peak height
13
14 validation and velocity range validation were applied on the raw vector. Finally, the
15
16 mean phase-averaged velocity profiles over 100 recordings were obtained in each
17
18 region.
Fo
19
20
21 Results and Discussion
22
r

23
24 Figure 5 compares the pump performance curves of both CFD and experimental
Re

25
26 results at different rotating speeds. Note that each point corresponds to a specific flow
27
28 rate and rotating speed under steady flow condition. Both the experimental and CFD
vi

29
30 simulation results show that the static pressure head decreases with the increasing
31
ew

32 flow rate. It can be found that CFD simulations overestimate the performance, mainly
33
34 due to the obstruction of the bars as indicated by black colour in Figure 4 (c).
35
36 However, they follow the trends of the experimental results. In order to match the
On

37
38 operating conditions used in the numerical simulation, the impeller rotating speed and
39
40 the pump flow rate were 9,500 rpm and 5 L/min respectively in the PIV experiment.
41
ly

42
43 In order to show the flow field between the hub (R=5.5 mm) and pump casing (R=7.5
44
45 mm), Figure 6 shows the longitudinal cross sectional planes through Z axis in each
46
47 region. Figures 7 (a), (b) and (c) show the comparisons of velocity profiles between
48
49 the CFD and PIV results in the straightener, impeller and diffuser regions,
50
51 respectively. Note that the solid line and dotted line denoting PIV and CFD results
52
53 respectively in these figures. In Figure 7 (a), the flow in the straightener region is
54
55 9
56
57
58
59
60
 Artificial Organs Page 10 of 24

1
2
3
4
5
6 smooth in positive axial direction. At the region close to the walls of the hub and
7
8 casing of the pump, the fluid has lower speed due to the boundary layer effect as
9
10 compared to that at the middle of the channel. As fluid flows forward, the smooth
11
12 curvature expansion of the straightener hub has accelerated the flow due to the
13
14 reduced cross sectional area (perpendicular to axial axis), and guides the flow
15
16 smoothly with a parabolic velocity profile into the space between the pump casing
17
18 and the hub constant diameter region, which begins at Z=-10 mm.
Fo
19
20
21 As the fluid flows into the impeller channel, the dominant flow is in positive Z
22
r

23 direction, and the high axial velocity is located slightly close to the impeller hub
24
(R=5.5 mm), especially at the further downstream until Z=18 mm, as shown in Figure
Re

25
26
27 7 (b). As a result, the fluid velocity close to the impeller shroud (R=7.5 mm) is much
28
smaller than that close to the hub (R=5.5 mm). From Z=18 mm onwards, it can be
vi

29
30
observed that the forward velocity at the impeller shroud has increased in magnitude,
31
ew

32
while near the impeller exit, reverse flow is detected close to the impeller hub.
33
34
Because the diffuser placed at the impeller downstream induces an adverse static
35
36
pressure gradient at the impeller exit, the fluid near the rotating impeller hub, which
On

37
38
has lower momentum than that near the impeller shroud, cannot overcome it and
39
40
flows reversely.
41
ly

42
43 Figure 7 (c) shows the velocity vector distributions in the diffuser. Compared with the
44
45 flow patterns in both straightener and the impeller, the flow pattern in the diffuser is
46
47 the most irregular with strong reverse flow close to the diffuser hub (R=5.5 mm). It is
48
49 because the fluid there has lower velocity and cannot overcome the adverse pressure
50
51 gradient. However, the forward flow becomes more dominant from Z=26.5 mm
52
53
54
55 10
56
57
58
59
60
Page 11 of 24 Artificial Organs

1
2
3
4
5
6 onwards where the blade ended. The similar flow pattern is predicted by the
7
8 numerical simulation.
9
10
11 A qualitative agreement between the numerical and experimental results can be
12
13 observed, especially in the straightener followed by impeller and diffuser. The
14
15 discrepancy becomes relatively obvious at the regions close to blades, hub and inner
16
17 casing surfaces due to a number of reasons. Firstly, the trapped seeding particles on
18
Fo
19 surfaces could induce relatively strong reflection, resulting in error in PIV
20
21 measurement. Secondly, the fluid close to the interfaces usually has higher velocity
22
r

23 gradient in radial direction, and it could cause error due to the difficulty in PIV
24
measurements. In order to calculate the estimated velocity accurately, the seeding
Re

25
26
27 particles within an interrogation area, where the camera image is usually divided into
28
32 pixels by 32 pixels in size, should be homogenous, and have exactly the same
vi

29
30
displacement. However, the displacement of individual particle is always different
31
ew

32
from each other under high velocity gradient. Thirdly, the out-of-plane motion of the
33
34
flow field will introduce signal drop-out. However, it has been minimized by selecting
35
36
proper laser thickness and the time between pulses as mentioned earlier in
On

37
38
experimental method. The L2 relative error norm of the differences in axial velocity
39
40
between CFD and PIV with respect to the PIV in the straightener, impeller and
41
ly

42 diffuser are 11%, 19%, and 27% respectively. Note that the L2 relative error norm is
43
44 N

∑(f )
2
45 E
− fi N
i
46 defined as i
, where f i E and f i N represent the values at node i for PIV
N
47
48 ∑(f )
i
i
E 2

49
50 experiment and numerical simulation respectively. These values are consistent with
51
52 the observation earlier.
53
54
55 11
56
57
58
59
60
 Artificial Organs Page 12 of 24

1
2
3
4
5
6 Conclusion
7
8
9
10 CFD was initially used to study the internal flow field of the axial flow blood pump.
11
12 The experimental test showed that the numerical simulation could predict the pump
13
14 characteristic curves with similar trends. Furthermore, the PIV measurement was
15
16 conducted for the validation at the flow conditions of 9,500 rpm rotating speed and
17
18 5L/min flow rate. Generally, the CFD results agree fairly well with the PIV results,
Fo
19
20 especially in the straightener region. It is also found that the matching between
21
22 numerical and measured results is related to the flow pattern, i.e. a better agreement is
r

23
always obtained in a smooth flow field such as in the straightener region. The major
24
Re

25
differences between the experimental and simulation results are usually found near
26
27
interfaces, due to the high velocity gradients, strong reflections from the seeding
28
vi

29
particles trapped there, and the high out-of-plane velocity component. Basically, a fair
30
31
agreement of the simulations and measurements can be observed, and it has
ew

32
33
reasonably validated the simulation results. Therefore, SST k-ω turbulence model is
34
35
able to predict the flow phenomenon in the axial flow blood pump.
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55 12
56
57
58
59
60
Page 13 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24
Re

25
26
27
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55 13
56
57
58
59
60
 Artificial Organs Page 14 of 24

1
2
3
4
5
6
7 References
8
9
10 1. Mielniczuk L, Mussivand T, Davies R, Mesana TG, Hendry RG, Mastersand
11 PJ, et al. Patient selection for left ventricular assist devices. Artif Organs
12 2004;28:152-7.
13
14 2. Borovetz HS, Ramasamy N, Zerbe TR, Portner PM. Evaluation of an
15 implantable ventricular assist system for humans with chronic refractory heart failure.
16 ASAIO J 1995;41:42-8.
17
18 3. Tayama E, Olsen DB, Ohashi Y, Benkowski R, Morley D, Noon GP, et al.
Fo
19 The debakey ventricular assist device: Current status in 1997. Artif Organs
20 1999;23:1113-6.
21
4. Anderson JB, Wood HG, Allaire PE, Olsen DB. Numerical analysis of blood
22
flow in the clearance regions of a continuous flow artificial heart pump. Artif Organs
r

23
2000;24:492-500.
24
Re

25 5. Song X, Throckmorton AL, Wood HG, Antaki JF, Olsen DB. Quantitative
26 evaluation of blood damage in a centrifugal vad by computational fluid dynamics. J
27 Fluids Eng 2004;126:410-8.
28
vi

29 6. Chua LP, Song G, Yu SCM, Lim TM. Computational fluid dynamics of gap
30 flow in a biocentrifugal blood pump. Artif Organs 2005;29:620-8.
31
ew

32 7. Legendre D, Antunes P, Bock E, Andrade A, Biscegli JF, Ortiz JP.

33 Computational fluid dynamics investigation of a centrifugal blood pump. Artif Organs
34 2008;32:342-8.
35
36 8. Mitamura Y, Nakamura H, Okamoto E, Yozu R, Kawada S, Kim D-W.
On

37 Development of the valvo pump: An axial flow pump implanted at the heart valve
38 position. Artif Organs 1999;23:566-71.
39
40 9. Zhu X, Zhang M, Zhang G, Liu H. Numerical investigation on hydrodynamics
41 and biocompatibility of a magnetically suspended axial blood pump. ASAIO J
ly

42 2006;52:624-9.
43
10. Throckmorton AL, Untaroiu A, Allaire PE, Wood HG, Lim DS, McCulloch
44
MA, et al. Numerical design and experimental hydraulic testing of an axial flow
45
ventricular assist device for infants and children. ASAIO J 2007;53:754-61.
46
47 11. Zhang Y, Zhan Z, Gui X, Sun HS, Zhang H, Zheng Z, et al. Design
48 optimization of an axial blood pump with computational fluid dynamics. ASAIO J
49 2008;54:150-5.
50
51 12. Arvand A, Hormes M, Reul H. A validated computational fluid dynamics
52 model to estimate hemolysis in a rotary blood pump. Artif Organs 2005;29:531-40.
53
54
55 14
56
57
58
59
60
Page 15 of 24 Artificial Organs

1
2
3
4
5
6 13. Paul R, Apel J, Schugner S, Shwindke P, Reul H. Shear stress related blood
7 damage in laminar couette flow. Artif Organs 2003;27:517-29.
8
9 14. Triep M, Brucker C, Schroder W, Siess T. Computational fluid dynamics and
10 digital particle image velocimetry study of the flow through an optimized micro-axial
11 blood pump. Artif Organs 2006;30:384-91.
12
13 15. Zhang J-M, Chua LP, Ghista DN, Zhou T-M, Tan YS. Validation of numerical
14 simulation with PIV measurements for two anastomosis models. Med Eng Phys
15 2008;30:226-47.
16
16. Burgreen GW, Antaki JF, Wu ZJ, Holmes AJ. Computational fluid dynamics
17
as a development tool for rotary blood pumps. Artif Organs 2001;25:336-40.
18
Fo
19 17. Goubergrits L, Affeld K. Numerical estimation of blood damage in artificial
20 organs. Artif Organs 2004;28:499-507.
21
22 18. Sankovic JM, Kadambi JR, Mehta M, Smith WA, Wernet MP. PIV
r

23 investigation of the flow field in the volute of a rotary blood pump. ASME
24 Transaction of ASME 2004;126:730-34.
Re

25
26 19. Day SW, McDaniel JC. PIV measurements of flow in a centrifugal blood
27 pump: Steady flow. Transaction of ASME 2005;127:244-53.
28
vi

29 20. Triep M, Brucker C, Kerkhoffs W, Schumacher O, Marseille O. Investigation

30 of the washout effect in a magnetically driven axial blood pump. Artif Organs
31 2008;32:778-84.
ew

32
33 21. Chua LP, Ong KS, Yu CMS, Chan WK, Wong YW. Gap velocity
34 measurements of a blood pump model. Artif Organs 2002;26:682-94.
35
36 22. Chua LP, Ong KS, Song G. Study of velocity and shear stress distributions in
the impeller passages and the volute of a bio-centrifugal ventricular assist device.
On

37
Artif Organs 2008;32:376-87.
38
39 23. Chua LP, Ong KS, Song G, Ji W. Measurements by laser doppler velocimetry
40 in the casing/impeller clearance gap of a biocentrifugal ventricular assist device
41
ly

model. Artif Organs 2009;33:360-72.

42
43 24. Day SW, McDaniel JC, Wood HG, Allaire PE, Landrot N, Curtas A. Particle
44 image velocimetry measurements of blood velocity in a continuous flow ventricular
45 assist device. ASAIO J 2001;47:406-11.
46
47 25. Day SW, McDaniel JC, Wood HG, Allaire PE, Song X, Lemire PP, et al. A
48 prototype heartquest ventricular assist device for particle image velocimetry
49 measurements. Artif Organs 2002;26:1002-5.
50
51 26. Song X, Wood HG, Day SW, Olsen DB. Studies of turbulence models in a
52 computational fluid dynamics model of a blood pump. Artif Organs 2003;27:935-37.
53
54
55 15
56
57
58
59
60
 Artificial Organs Page 16 of 24

1
2
3
4
5
6 27. Lim TM, Zhang D. Development of lorentz force-type self-bearing motor for
7 an alternative axial flow blood pump design. Artif Organs 2006;30:347-53.
8
9 28. Chua LP, Su BY, Lim TM, Zhou TM. Numerical simulation of an axial blood
10 pump. Artif Organs 2007;31:560-70.
11
12 29. Menter FR. Two-equation eddy-viscosity turbulence models for engineering
13 applications. AIAA J 1994;32:1598-605.
14
15 30. Zhu L, Zhang X, Yao Z. Shape optimization of the diffuser blade of an axial
16 blood pump by computational fluid dynamics. Artif Organs 2010;34:185-92.
17
31. Kido K, Hoshi H, Watanabe N, Kataoka H, Ohuchi K, Asama J, et al.
18 Computational fluid dynamics analysis of the pediatric tiny centrifugal blood pump
Fo
19 (tinypump). Artif Organs 2006;30:392-9.
20
21 32. Fraser KH, Taskin ME, Griffith BP, Wu ZJ. The use of computational fluid
22 dynamics in the development of ventricular assist devices. Med Eng Phys
r

23 2011;33:263-80.
24
Re

25 33. Chua LP, Song GL, Zhou TM. Numerical analysis of the inner flow field of a
26 biocentrifugal blood pump. Artif Organs 2006;30:467-77.
27
28 34. Throckmorton AL, Untaroiu A, Allaire PE, Wood HG, Matherne GP, Lim DS,
vi

29 et al. Computational analysis of an axial flow pediatric ventricular assist device. Artif
30 Organs 2004;28:881-91.
31
ew

32 35. Doyle MG, Vergniaud JB, Tavoularis S, Bourgault Y. Numerical simulations

33 of blood flow in artificial and natural hearts with fluid-structure interaction. Artif
34 Organs 2008;32:870-9.
35
36 36. Dantec-Dynamics. Flowmanager software and introduction to PIV
instrumentation (software user's guide) 2002.
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55 16
56
57
58
59
60
Page 17 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24 Figure 1: The demonstration of the axial flow VAD
Re

25
26
27
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50 Figure 2: Grid independence test
51
52
53
54
55
56
57
58 1
59
60
 Artificial Organs Page 18 of 24

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24
(a) Detailed surface grids
Re

25
26
27
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
(b) Complete numerical domain
46
47
48
Figure 3: Surface grids on the axial flow blood pump
49
50
51
52
53
54
55
56
57
58 2
59
60
Page 19 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24
Re

25
26
27
28
vi

29
30
31
ew

32 (a)
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52 (b)
53
54
55
56
57
58 3
59
60
 Artificial Organs Page 20 of 24

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24
Re

25
26
27
28
vi

29
30
31
(c)
ew

32
33 Figure 4: Schematic view of (a) the test rig and (b) test model mounted on the traverse
34 stage in PIV measurement. (c) CAD drawing of the bearing holder
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 4
59
60
Page 21 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21
22
r

23
24
Re

25
26
27 Figure 5: Experimental and numerical pump characteristic curves
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 5
59
60
 Artificial Organs Page 22 of 24

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20
21 (a) straightener
22
r

23
24
Re

25
26
27
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41 (b) impeller
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 6
59
60
Page 23 of 24 Artificial Organs

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Fo
19
20 (c) diffuser
21
22 Figure 6: Demonstration of the cross sectional planes in each region
r

23
24
Re

25
26
27
28
vi

29
30
31
ew

32
33
34
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 7
59
60
 Artificial Organs Page 24 of 24

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18 (a) Straightener
Fo
19
20
21
22
r

23
24
Re

25
26
27
28
vi

29
30
31
ew

32
33
34 (b) Impeller
35
36
On

37
38
39
40
41
ly

42
43
44
45
46
47
48
49
50 (c) Diffuser
51
52
53
54
Figure 7: Comparison of the velocity profiles obtained from PIV (solid line) and CFD
55 (dotted line) results in (a) straightener, (b) impeller and (c) diffuser
56
57
58 8
59
60