See discussions, stats, and author profiles for this publication at: [Link]

net/publication/371568153

Fan Type Inducer for a Centrifugal Pump by Wire Arc Additive Manufacturing
and Machining

Article in International Journal of Fluid Machinery and Systems · June 2023

DOI: 10.5293/IJFMS.2023.16.2.184

CITATIONS READS

0 80

1 author:

Shinichiro Ejiri
Nikkiso
10 PUBLICATIONS 6 CITATIONS

SEE PROFILE

All content following this page was uploaded by Shinichiro Ejiri on 28 July 2023.

The user has requested enhancement of the downloaded file.

International Journal of Fluid Machinery and Systems DOI: [Link]
Vol. 16, No. 2, April-June 2023 ISSN (Online): 1882-9554

Original Paper

Fan Type Inducer for a Centrifugal Pump by

Wire Arc Additive Manufacturing and Machining
Shinichiro Ejiri1
1
Nikkiso Co., Ltd.
2-16-2, Noguchicho, Higashimurayama-shi, Tokyo 189-8520, Japan, [Link]@[Link]

Abstract

The wire arc additive manufacturing (WAAM) is one of the metal additive manufacturing technologies. WAAM can
be additive manufactured to existing parts, and when combined with machining, has a high affinity for the fabrication of
blades, which are important parts of turbomachinery. However, almost all related research has been limited to the
fabrication of parts such as turbine blade or propeller while investigations have not been conducted by incorporating the
fabricated parts into turbomachinery. In this study, a fan type inducer made of stainless steel was developed using
WAAM and machining. The developed inducer was installed to a centrifugal pump, and pump performance tests were
conducted to investigate the advantage of this fabrication process to turbomachinery.

Keywords: WAAM, Machining, Additive Manufacturing, Fabrication process, Inducer, Centrifugal Pump

1. Introduction
The wire arc additive manufacturing (WAAM), a type of additive manufacturing, has been the subject of active research and
development in recent years [1][2]. The basic idea of WAAM can be traced back to a patent by Baker about 100 years ago [3].
The patent describes the invention of shaping containers by stacking weld beads. In order to make WAAM work, proper control of
welding conditions is necessary. Therefore, this made it difficult to achieve by using the welding machines of the time, and it took
time for the technology to develop. In recent years, the development of control technology has made it possible to perform
additive manufacturing with WAAM by using welding robots [4] and hybrid multitasking machines [5], this has created an area of
various research and development activities.
WAAM is not suitable for very small and complex fabrication, however, it is suitable for relatively large fabrication. The
characteristics of WAAM are described. If WAAM is performed in the atmosphere, there is no limit to the size of the additive
manufacturing. Another advantage is additive manufacturing at a relatively low cost, since commercially available welding wire is
used. It is also possible not to use support during additive manufacturing by adjusting the position of the workpiece during
processing. Furthermore, the Lamination pitch of WAAM is only a few millimeters, which is much faster than additive
manufacturing that uses metal powder. However, the surface roughness of WAAM with weld bead stacking has the problem
several hundred micrometers. Therefore, considering fluid performance such as friction loss, etc., It is unrealistic to install parts
made only with WAAM inside a turbomachinery. But this problem can be resolved by applying a fabrication process in which the
shape is finished by machining after WAAM. However, this solution is difficult to apply to some turbomachinery parts. For
example, it may be difficult to finish by machining the narrow flow paths enclosed by the side plates of a closed impeller, thus
WAAM and machining processes may not be suitable for it.
The space around the blade parts, which are the key components of turbomachinery, must have flow paths. Therefore, the
process of fabricating only the blade area, rather than the entire flow path, can be expected to improve the efficiency of the
fabrication process. One way this can be accomplished is through WAAM and machining, which is the subject of this study.
There are instances of research and development using WAAM and machining for turbomachinery blade components. For
instance, an impeller for a compressor [5], a turbine impeller [6], a turbine blade [7], a turbopump impeller [8], and a propeller [9]
have been fabricated. However, most of these studies have been limited to fabrication of parts. One of the few demonstration tests
on turbomachinery using WAAM and some kind of finishing process is the case of a ship propeller, called WAAMpeller [10],
made with WAAM and grinding, installed and operated on a ship. Although, it is difficult to decide everything based solely on the
limited few results of the demonstration tests. Therefore, more demonstration tests are required for the industrial application of
WAAM-based technology.
Received October 27 2022; revised February 13 2023; accepted for publication April 9 2023: Review conducted by Hideaki Tamaki (Paper
number O22042J)
Corresponding author: Shinichiro Ejiri, [Link]@[Link]

184
 In this study, firstly, a fan type inducer installed in an industrial centrifugal pump developed using WAAM and machining,
and it will be verified that the inducer can be fabricated without any problems. In addition, a demonstration test conducted by
operating with this inducer installed on a centrifugal pump and measuring the pump performance curve and suction performance.
Furthermore, the applicability of this fabrication process to industrial turbopumps is discussed through analysis focusing on the
removal material weight.

2. Fan type inducer

Industrial turbo pumps that require low NPSHR are generally installed with an inducer to improve pump suction performance.
In particular, centrifugal pumps for pumping liquefied gases such as liquefied natural gas (LNG) use fan type inducers of low
solidity to improve pump suction performance over a wide flow range [11]. Figure 1 shows the outline of the fan type inducer
developed in this study. Table 1 is shown on the fan type inducer main specifications.

Fig. 1 Outline of fan type inducer

Table 1 Fan type inducer main specifications

Number of blades 4
Tip diameter [mm] 125.3
Hub diameter [mm] 51
Hub length [mm] 40

3. Fabrication process
WAAM was performed using stainless steel (SUS316L) for the base metal, which is mainly use for the hub of the fan type
inducer, and stainless steel welding wire (YS316L) for the blade. WAAM was performed using VARIAXIS j-600/5X AM by
Yamazaki Mazak under the conditions shown in Table 2. WAAM conditions were set based on the research results of Anzai et al
[12], referring to the conditions that would achieve mechanical strength equal to or greater than that of JIS equivalent materials
and that would minimize the occurrence of blowholes inside the metal. The machining process after WAAM was performed using
INTEGREX e-1600V/10 by Yamazaki Mazak and the typical stainless steel machining conditions shown in Table 3. Figure 2
shows an overview of these fabrication processes. This process consists of WAAM on the rough machined base metal, then
finishing by machining.
After WAAM and machining, an X-ray CT scan was performed using a TOSCANER-24500AVFD by Toshiba IT Control
Systems to observe internal defects. As shown in Fig. 3, only one 0.7 mm × 1.0 mm elliptical blowhole was observed inside one
of the four blades. The location of the blowhole corresponds to the first layer of WAAM. No other internal defects were observed.
Since the defect occurred in only one blade, it is assumed that the internal defect was not mainly caused by the WAAM conditions
shown in Table 2, but by the effect of temperature distribution during metal lamination. Regarding this deficiency, it may be
possible to solve it by applying WAAM controls, including monitoring of the temperature distribution [13]. Since small amount
and fine blowholes cause no significant effect on mechanical properties [14], it was decided to install the developed fan type
inducer on a centrifugal pump and measure the pump performance. However, large number of internal defects affects not only
mechanical properties but also fatigue strength [15]. Therefore, sufficient consideration should be given when using it for
industrial applications.

185
 Table 2 WAAM conditions
Layer 1 to 2 After 3
Current [A] 243 88
Voltage [V] 13.8 10.4
Wire feed [m/min] 9 2
Torch feed [mm/min] 400 100
Shield gas Ar Ar

Table 3 Machining conditions

Tool Ball endmill R3
Cutting speed [m/min] 75
Rotation speed [rpm] 3000
Feed rate [mm/min] 800
Cut depth [mm] 0.2

(a) Welding base metal (b) After WAAM (c) After machining
Fig. 2 WAAM and machining processes

(a) Transverse cross section (b) Longitudinal cross section

Fig. 3 Result of X-ray CT scan of a blade

4. Pump performance test

4.1 Test facility and test pump\
As shown in Figs. 4 and 5, the experimental apparatus mainly consists of suction pipes (Ds=127 mm), discharge pipes (Dd=78
mm), a control valve, an ultrasonic flow meter, a vacuum pump and a tank. A pump is placed at the test section and motor is
electronically controlled to maintain the rotational speed constant at any set value. The working fluid is fresh water. The pump
performance is evaluated from shaft rotational speed, torque, flow rate, and the difference of static pressure between the section A
and B. The flow rate is adjusted to any set value with a control valve and is measured by an ultrasonic flowmeter. When
measuring pump suction performance, a vacuum pump was used to change the pressure of the liquid surface in the tank, which
changed the pressure of the entire piping system.
As shown in Figs. 6 and 7, test model is a centrifugal pump with specific speed of 138 (m3/min, m, min-1) having a 4-blade fan
type inducer, a 9-blade impeller and an 8-vane diffuser with diffuser vane slit [16]. The centrifugal pump used in the present study
is a vertical pump that is mainly used for transporting LNG (Liquified Natural Gas). The discharge port was modified by turning
sideways for applying to the test rig as shown in Fig. 4. The design flow coefficient of the test pump is =0.099 and rotational
speed is 3000 rpm.

186
 Fig. 4 Test facility

Fig. 5 Overall of Test facility

Fig. 6 Test pump Fig. 7 Test pump cross section

4.2 Test results

The pump performance curves obtained from the operation are shown in Fig. 8, and the pump suction performance at the
design flow coefficient =0.099 is shown in Fig. 9. To compare the performance with the developed inducer, measurement results
obtained when the pump was operated without the inducer are shown along with each figure. In the pump performance curve
shown in Fig. 8, it can be seen that the head coefficient increased with the addition of the inducer. In the pump suction
performance shown in Fig. 9, it can be confirmed that the pump suction performance is clearly improved by the installation of the
inducer. These measurement results demonstrate that the fan type inducer developed by WAAM and machining process fully
functions as an inducer during pump operation. The appearance of the inducer after the pump performance test is shown in Figure
9. No obvious damage was observed on the entire surface of the inducer, confirming that it was operated without any problems.
The total operation time was approximately 2 hours.

187
 Fig. 8 Pump performance characteristics

Fig. 9 Pump suction performance characteristics

(a) Suction side (b) Pressure side

Fig. 10 Fan type inducer after pump performance test

5. Discussion
In order to discuss the industrial advantage of this fabrication process, an analysis of the removal material weight was
conducted. The two cases of fabrication by machining only are used for comparison. The machining allowance for the WAAM
areas was estimated as 2 mm, which is required during a typical machining process. The outline is shown in Fig. 11(a). Regarding
the first case of machining only, finish machining is performed from the shape of a rough machined blades welded to a round bar
that is the hub, as shown in Fig. 11(b). The blades of the fan type inducer in this study are twisted, thus it is structurally difficult to
weld a flat plate of sufficient thickness for form the blades directly to a round bar. Therefore, it was decided to form a blade shape
with a finishing allowance from a flat plate by machining. The dimensions of the flat plate shown in Fig. 11(c) are width: 80 mm,

188
length: 230 mm (workpiece: 200 mm, gripping allowance: 30 mm), and thickness: 15 mm. Regarding the second case of
machining only, the selection of workpiece material from marketable materials based on the shape of the inducer, a round bar with
a diameter of 130 mm and a length of 70 mm (workpiece 40 mm, gripping allowance 30 mm) is a candidate. In the present
analysis, machining from this round bar is considered. The outline is shown in Fig. 11(d). First, the weight after WAAM minus
the weight after finishing by machining is set to 1. Similarly, the removed material weight when fabricated only by machining
from the round bar with weld rough machined blades is calculated to be 4.1 and the selected round bar is calculated to be 6.7. The
result of this analysis is shown in Fig. 12. In this case, it was found that the removal material weight could be significantly
decreased approximately 76% and 85% by using WAAM.
The inducer in this study has a relatively low design flow rate and small size for pumping such as LNG. Therefore, in order to
analyze further industrial advantages from this process, an additional analysis was conducted for a similar fabrication process. The
effect of fabricating a large inducer with a design flow rate in the high flow range several times larger than the inducer used in this
study was calculated. The calculated result is shown in Fig. 13. The horizontal axis in Fig. 13 shows the inducer tip diameter ratio.
The inducer tip diameter ratio is defined the tip diameter of the fabricated fan type inducer in this study in the denominator and the
tip diameter of the fan type inducer for analysis in the numerator. In Fig. 13, the analysis targets tip diameters up to approximately
630 mm, which is the inducer tip diameter ratio: 5. The dimensions of the fan type inducers for analysis assume the same specific
speed and use a similarity rule. As the shape increases, the difference in the removal material weight ratio increases, showing that
a large amount of removal material weight reduction can be expected.
Based on the results of these analyses, the process has industrial advantages. The fabrication process using WAAM and
machining investigated in this study is also applicable to axial flow impellers of axial flow pumps [17], which are similar in shape
to fan type inducers, and the process is applicable to a wide range of blade components. Furthermore, various studies have shown
that a blade shape of an inducer affects the occurrence of cavitation and pump suction performance [18] [19] [20]. Since only one
model case was studied in this study, future studies should be conducted for different blade shapes than those covered in this study.

(a) WAAM & Machining model (c) Rough machined blades form flat plate

(b) First case of machining only model (d) Second case of machining only model
(Round bar with welded rough machined blades) (Round bar)

Fig. 11 Outline of workpiece for removal material weight analysis

189
 Fig. 12 Comparison of removal material weight by machining

Fig. 13 Estimation of removal material weight by machining

6. Conclusion
In the study, a fan type inducer was developed by WAAM and machining. In addition, the inducer was installed in a
centrifugal pump to conduct pump performance tests. Furthermore, analysis of removal material weight at fabrication is conducted.
The obtained conclusions are follows:
(1) It is possible to fabricate a fan type inducer using WAAM and machining.
(2) The developed inducer was installed to a centrifugal pump and operated for approximately 2 hours, indicating that it
functioned adequately as an inducer.
(3) The fabrication process using WAAM and machining provides industrial advantages over the machining only fabrication
process in terms of lower removed material weight. Then these advantages increase with the size of the part shape.

Acknowledgments
The author thanks Mr. Takanobu Miura (Miura Co., Ltd.), Mr. Minoru Tanaka (Miura Co., Ltd.), Mr. Hiroki Anzai (Industrial
Technology Institute Fukushima Prefectural Government) and Mr. Daiju Anazawa (Industrial Technology Institute Fukushima
Prefectural Government) for the cooperation of the fabrication and Prof. Masahiro Miyabe (Osaka Institute Technology) for the
cooperation of pump performance tests.

190
 Nomenclature
b width of flow passage [mm] f flow coefficient =Q/(3600pb2d2u2)
d diameter [mm] η efficiency [%]
D pipe diameter [mm] ρ density [kg/m3]
H total head [m] τ shaft power coefficient =2L/(ru23pb2d2)
L shaft power [kW] Ψ head coefficient =H/(u22/2g)
P static pressure [Pa=N/m2] Ns specific speed [m3/min, m, min-1]
Q flow rate [m3/h] Pv Saturation vapor pressure [Pa=N/m2]
u peripheral velocity [m/s] σ cavitation number =(Ps-Pv)/(ρu2/2)

Subscripts
s suction pipe 2 impeller outlet
d discharge pipe BEP best efficiency point

References
[1] Rodrigues, T. A., et al., 2019, “Current status and perspectives on wire and arc additive manufacturing (WAAM),” Materials,
Vol.12, No.7, 1121.
[2] Kumar, N., et al., 2022, “Wire Arc Additive Manufacturing–A revolutionary method in additive manufacturing,” Materials
Chemistry and Physics, Vol.285, 126144.
[3] Baker, R., 1925, US Patent No. US1533300A.
[4] Kulikov, A. A., et al., 2020, “Process Design for the Wire Arc Additive Manufacturing of a Compressor Impeller,” IOP
Conference Series: Materials Science and Engineering, Vol. 969, 012098.
[5] Miyake, R., et al., 2021, “Wire Arc Additive and High-Temperature Subtractive Manufacturing of Ti-6Al-4V,” Applied
Sciences, Vol. 11, No.20, 9521.
[6] Cunningham, C. R., et al., 2017, “Cost modelling and sensitivity analysis of wire and arc additive manufacturing,” Procedia
Manufacturing, Vol.11, pp. 650-657.
[7] Nagamatsu, H., et al., 2020, “Development of a cooperative system for wire and arc additive manufacturing and machining,”
Additive Manufacturing, Vol. 31, 100896.
[8] Chandra, E., 2021, “Design and Fabrication of Al-Si (Er4043) Aluminium Impeller Using Wire Arc Additive Manufactured
(Waam),” International Journal of Modern Agriculture, Vol. 10, No.2, pp. 4176-4189.
[9] Eyercioglu, O., et al., 2021, “A Propeller Blade Manufacturing by Hybrid Additive Manufacturing System,” International
Journal of 3D Printing Technologies and Digital Industry, Vol.5, No.3, pp. 560-568.
[10] Damen Shipyards, 30 November 2017., “World’s first class approved 3D printed ship’s propeller unveiled”
[Link]
(accessed 23 October 2022).
[11] Kurokawa, J., et al., 1993, “Performance Prediction of Multistage Diffuser Pump of Low Specific Speed and Viscous Effect,”
ICOPE, Tokyo, Japan, pp. 293-298.
[12] Anzai, H., et al., 2022, “Understanding characteristics of components fabricated by Additive Manufacturing,” Industrial
Technology Institute Fukushima Prefectural Government Test Report (in Japanese).
[13] Carter, W., et al., 2019, “Thermal process monitoring for wire-arc additive manufacturing using IR cameras, International
Solid Freeform Fabrication Symposium,” Texas, USA, pp. 1812-1817.
[14] Japanese Industrial Standards, 1995, “Methods of Radiographic Examination for Welded Joints in Steel,” JIS Z 3104.
[15] Biswal, R., et al.,2019, “Criticality of porosity defects on the fatigue performance of wire+ arc additive manufactured
titanium alloy,” International Journal of Fatigue, Vol. 122, pp. 208-217.
[16] Takao, S., et al., 2021, “Suppression of Diffuser Rotating Stall in A Centrifugal Pump by Use of Slit Vane,” ASME FEDSM,
Online, FEDSM2021-65519.
[17] Watanabe, S., et al., 2011, “Internal Flow and Limiting Streamlines Observations of Contra-Rotating Axial Flow Pump at
Partial Flow Rate,” International Journal of Fluid Machinery and Systems, Vol. 4, No. 2, pp. 235-242.
[18] Kang, D., et al., 2009, “Inducer design to avoid cavitation instabilities,” International Journal of Fluid Machinery and
Systems, Vol. 2, No. 4, pp. 439-448.
[19] Krátký, T., et al., 2016, “CFD and surrogates-based inducer optimization,” International Journal of Fluid Machinery and
Systems, Vol. 9, No. 3, pp. 213-221.
[20] Morii, T, et al., 2019, “Suction performance and cavitation instabilities of turbopumps with three different inducer design,”
International Journal of Fluid Machinery and Systems, Vol. 12, No. 2, pp. 128-135.

191

View publication stats