MODELLING AND SIMULATION OF CORIOLIS
FLOW METER USED FOR BUNKER FUEL OIL
FLOW
by
IVAN KURNIA
3269978
Submitted in accordance with the requirements for
MECH4841 A – Mechanical Engineering Project
The University of Newcastle
Faculty of Engineering and Built Environment
School of Engineering, Discipline of Mechanical Engineering
Supervisor:
Date of Submission:
Dr. Luo Rongmo
10 December 2018
DECLARATION
I declare that this thesis is my own work unless otherwise acknowledged and is in
accordance with the University’s academic integrity policy available from the
Policy Library on the web at
http://www.newcastle.edu.au/policylibrary/000608.html
I certify that this assessment item has not been submitted previously for academic
credit in this or any other course.
I acknowledge that the assessor may, for the purpose of assessing this thesis:
 Reproduce this thesis and provide a copy to another member of the Faculty;
and/or
 Communicate a copy of this assessment item to a plagiarism checking
service (which may then retain a copy of the item on its database for the
purpose of future plagiarism checking).
 Submit the assessment item to other forms of plagiarism checking.
I certify that any electronic version of this assignment item that I have submitted
or will submit is identical to this paper version.
Student name:
Date:
Ivan Kurnia
13-10-2018
i
ACKNOWLEDGEMENTS
I would like to thank you Dr Luo for his guidance for this report throughout the semester.
ii
TABLE OF CONTENTS
DECLARATION ........................................................................................................................ i
ACKNOWLEDGEMENTS .......................................................................................................ii
TABLE OF CONTENTS ......................................................................................................... iii
NOMENCLATURE .................................................................................................................. v
CHAPTER 1 INTRODUCTION .............................................................................................. 1
1.1
Background ................................................................................................................. 1
1.2
Project Aim and Scope ................................................................................................ 1
1.3
Coriolis Flow Meter .................................................................................................... 1
1.3.1
Working Principle ................................................................................................ 1
............................................................................................................................................ 2
1.3.2
Design Breakdown ............................................................................................... 3
1.3.3
Dimensions .......................................................................................................... 3
Figure 4: Specification summary of commercially available CFM. ...................................... 3
1.4
Governing Equations ................................................................................................... 4
1.4.1
Reynolds Number ................................................................................................ 4
Figure 5: Formula for Reynolds Number........................................................................... 4
1.4.2
Beam Theory........................................................................................................ 4
1.4.3
Arbitrary Lagrangian-Eulerian (ALE) Formulation ............................................ 5
Figure 8: ALE formulation used to find low Reynolds number effect. ................................. 5
1.5
Methodology ............................................................................................................... 5
Figure 9: ANSYS modelling of single bent tube for the project’s future experiment. .......... 6
1.6
GANTT Chart ............................................................................................................. 6
CHAPTER 2 LITERATURE REVIEW ................................................................................... 1
2.1
CMF Performance on Bunker Fuel ............................................................................. 1
2.2
Cause of Measurement Error....................................................................................... 1
2.2.1
Low Reynolds Number ........................................................................................ 1
2.2.2
Pressure ................................................................................................................ 1
2.2.3
Vibration .............................................................................................................. 2
2.2.4
Bubble Theory ..................................................................................................... 2
2.3
Fluids Passing Through CFM ..................................................................................... 2
iii
2.3.1
Heavy Oil ............................................................................................................. 2
2.3.2
Water .................................................................................................................... 2
REFERENCES .......................................................................................................................... 3
APPENDIX A – GANTT CHART ............................................................................................ 5
iv
NOMENCLATURE
𝑅𝑒
Reynolds number
Greek Letters

Angular velocity
Abbreviations
CFD
Computational Fluid Dynamics
CFM
Coriolis Flow Meter
MPA
Maritime and Port Authority of Singapore
GVF
Gas Volume Fraction
cSt
centistokes, unit of kinematic fluid viscosity
Pa s
Pascal second, unit of kinematic viscosity, which translates to 1000 cSt
cP
centipoise, unit of dynamic viscosity, translates to 1 mPa s
v
CHAPTER 1
INTRODUCTION
1.1
Background
In the past 20 years, Coriolis flow meter (CFM) has become one of the most researched type
of flow meter and gained wide usage in many industries, in tandem with continuous
improvement in areas of computational fluid dynamics (CFD). It was predicted to attain the
largest demand of all flow meters, overtaking differential pressure flow meter (Wang & Baker,
2014).
Singapore has the largest bunkering port in the world and is a global leader in bunker fuel sales.
Its MPA has mandated all bunker companies to use CFM for high viscosity oil since January
2017, citing improvement in quantitative transparency and transportation and storage
efficiency. MPA also predicted that the variance in fuel quantity can be lowered from 0.7% to
0.5% by using CFM (Liang, 2016).
Moreover, due to sheer scale of Singapore’s bunker industry, even small error can lead to
serious money loss. A mere 2.26 M ton of disputed quantity can translate to $ 690 million
dollars in cost according to ExxonMobil Asia Pacific representative (Liang, 2016).
1.2
Project Aim and Scope
The project aims to identify factors related to oil viscosity that lead to inaccuracy in CFM
reading and mitigate them. Other factors will also be addressed for CFM applications beyond
the bunkering industry. The project will mainly focus on CFM application on high density,
high viscosity bunker fuel flow. By reducing uncertainties on the already highly accurate CFM,
more time and money can be saved in bunkering industry, keeping Singapore ahead of the
bunkering competition.
1.3
Coriolis Flow Meter
1.3.1 Working Principle
Below is the diagram for a bent U-shaped flow meter. Although there are many different types
of CFM including straight tube ones, the mechanism is generally the same for all CFM.
1
Figure 1: Diagram for simple bent, double tube Coriolis flow meter.
:
The flow tube is vibrated at resonant frequency by the driver and when there is no flow, the
inlet and outlet part are in phase. However, when flow is present, the frequency interaction
with the flow leads to Coriolis acceleration of fluid particles. Such acceleration creates a
Coriolis force directly proportional to the flowing fluid’s density and velocity. This creates
phase difference between inlet and outlet in terms of time delay, which is then read by the
motion sensor to obtain mass flow (Sultan & Hemp, 1989). The graph below shows how the
phase difference is observed.
Figure 2: Phase shift measurement for bent, double tube CFM.
2
1.3.2 Design Breakdown
CFM design can be differentiated by their shapes and configurations, and the two are not
mutually exclusive. There are straight and bent pipe design, which can have either single pipe
or double pipe configuration. Across all industries, including bunkering, bent pipe is much
more commonly used although single tube finds its niche in food processing industry (Anklin,
Drahm, & Rieder, 2006).
Figure 3: Available CFM models in the market.
1.3.3 Dimensions
Below is the summary of CFM available for sales as of 2014, as found by Wang and Baker.
They noted that the majority of CFM for sale have bent tube shape or its variations e.g. Vshaped and Ohm-shaped.
Figure 4: Specification summary of commercially available CFM.
3
1.4
Governing Equations
1.4.1 Reynolds Number
Because the report mainly focuses on oil viscosity, Reynolds number becomes an important
parameter to consider. It is inversely proportional to viscosity, hence in the report’s case low
Reynolds number will be investigated, like research done by Anklin and Kumar. ρ stands for
oil density and µ for kinematic viscosity. V is fluid velocity and D is tube diameter.
Figure 5: Formula for Reynolds Number
1.4.2 Beam Theory
For measurement of deflection, the tube can be assumed as a beam and thus subject to beam
theorem. Research on CFM goes back to 1958 (Wang and Baker, 2014) and so researchers
have had used different theorem over the years. Figure 6 shows Euler beam formulation for a
bent tube from Sultan and Hemp in 1989, while Figure 7 shows Timoshenko beam theory used
by Wang and Hussain to find pressure effects on CFM in 2010.
Figure 6: Euler beam theory for bent tube.
4
Figure 7: Timoshenko beam theory for straight tube.
1.4.3 Arbitrary Lagrangian-Eulerian (ALE) Formulation
Below is the ALE formulation used by Anklin and Kumar during their experiment with
Reynolds number. It is lauded for its high meshing accuracy and ability to split the fluid mesh
from the tube, making the simulation stable and accurate.
Figure 8: ALE formulation used to model the fluid.
1.5
Methodology
Simulation of bunker oil flow will be carried out using CFD and physical experiment. The CFD
model will be created using ANSYS student version. If possible, CFD experiments carried out
in the journals will be replicated to measure their relevance to the project’s model. Should the
model fail to deliver, it will be redone be it in terms of geometry or initial condition until a data
trend is observed. Real life experiment will be carried out using water as the passing fluid,
given the unviability of actual bunker fuel.
5
Figure 9: ANSYS modelling of single bent tube for the project’s future experiment.
1.6
GANTT Chart
The GANTT chart for MECH4841A for the entire semester can be found under Appendix A.
6
CHAPTER 2
LITERATURE REVIEW
2.1
CMF Performance on Bunker Fuel
CFM is very viable for high viscosity oil/gas mixtures, with mass flow error of 5% and 2.5%
for gas and liquid respectively. Their experiment utilized oil viscosities of 50 to 500 cSt and
GVF of 0-90%, with high accuracy reading observed even at 90% GVF. Although the margin
of error slightly increased at lower viscosities, the overall accuracy for the two-phase flow is
still outstanding (Tombs et al., 2018). Similarly, Henry et al. (2006) suggested that as viscosity
tends to infinity, the measurement error for mass flow and density tends to zero.
2.2
Cause of Measurement Error
2.2.1 Low Reynolds Number
Kumar and Anklin (2011) discovered that at low Reynold number, CFM reading may fluctuate
due to presence of fluid-dynamic forces. Such effect is caused by periodic shear mechanism,
which on encounter with the oscillatory Coriolis force, decreases the tube deflection. Part of
the Coriolis force was spread out thin in the secondary circulation and does not contribute to
deflection. The authors suggested compensation method using a function of Reynolds number,
reducing error to under 0.2%, down from 0.5 – 1%.
2.2.2 Pressure
Pressure also affects the reading but to a negligible extent. Wang and Hussain (2010), using
linear damping model to investigate the pressure effects on CFM, found that newer CFM has
pressure sensitivity of less than 0.01% per bar in contrast to older models averaging at 0.13%.
An interesting finding is that most bent tube flow meter has negative pressure sensitivity while
straight tube ones have positive pressure sensitivity. The mechanism behind such behaviour
and pressure effects in general is currently unknown.
1
2.2.3 Vibration
Study made by Ridder et al. (2014) shows that external vibrations at the CFM’s drive frequency
gives rise to measurement error, regardless of algorithm used by the CFM. They proposed to
minimise such influence with strong balancing system such as twin-tube configuration.
However, this model does not work with CFM designed for low mass flow.
Moreover, the magnitude of error greatly varies by algorithm used and it is still unclear how
vibration creates error, hence it is hard to measure the extent of vibration influence on the
reading (Clark & Cheesewright, 2003).
2.2.4 Bubble Theory
The measurement error due to particle entrainment can be described under bubble theory
assumptions. When acceleration of the two phases are different, phase decoupling occurs which
causes negative error i.e. the reading is always below the real value. Two-phase flow is
especially prone to large decoupling phase error (Nils, 2014).
2.3
Fluids Passing Through CFM
2.3.1 Heavy Oil
The oil used in this report is heavy crude oil used for bunker fuel. It has density of at least 920
kg/m3 and can go above 1000 kg/m3 for very heavy crude oil (Anton Paar Wiki). The dynamic
viscosity is around 10 000 cP. (Alaska Gas and Oil Association).
2.3.2 Water
Water has density of about 1000 kg/m3 with slight fluctuations depending on temperature.
Researchers commonly use this fluid in experiments aiming to identify and compensate causes
of measurement error that are not viscosity related.
2
REFERENCES
Wang, T., & Baker, R. (2014). Coriolis flowmeters: A review of developments over the past
20 years, and an assessment of the state of the art and likely future directions. Flow
Measurement and Instrumentation,40, 99-123. doi:10.1016/j.flowmeasinst.2014.08.015
Liang, L. H. (2016, March 31). Singapore moves closer to mass flow meter bunkering.
Retrieved December 10, 2018, from http://www.seatrade-maritime.com/news/asia/singaporemoves-closer-to-mass-flow-meter-bunkering.html
Liang, L. H. (2016, December 29). Singapore enters 2017 with mandatory use of mass flow
meter bunkering. Retrieved October 8, 2018, from http://www.seatrademaritime.com/news/asia/singapore-enters-2017-with-mandatory-use-of-mass-flow-meterbunkering.html
The Basics of Flow Measurement with Coriolis Meters: Part 2. (2017, April 18). Retrieved
from https://www.flowsolutionsblog.com/blog/the-basics-of-flow-measurement-withcoriolis-meters-part-2/
Sultan, G., & Hemp, J. (1989). Modelling of the Coriolis mass flowmeter. Journal of Sound
and Vibration,132(3), 473-489. Retrieved October 10, 2018, from
https://www.sciencedirect.com/journal/journal-of-sound-and-vibration/vol/132/issue/3.
Anklin, M., Drahm, W., & Rieder, A. (2006). Coriolis mass flowmeters: Overview of the
current state of the art and latest research. Flow Measurement and Instrumentation,17(6),
317-323. doi:10.1016/j.flowmeasinst.2006.07.004
Viscosity of Crude Oil – viscosity table and viscosity chart :: Anton Paar Wiki. (n.d.).
Retrieved from https://wiki.anton-paar.com/en/crude-oil/
Kumar, V., & Anklin, M. (2011). Numerical simulations of Coriolis flow meters for low
Reynolds number flows. Mapan,26(3), 225-235. doi:10.1007/s12647-011-0021-6
3
Wang, T., & Hussain, Y. (2010). Pressure effects on Coriolis mass flowmeters. Flow
Measurement and Instrumentation,21(4), 504-510. doi:10.1016/j.flowmeasinst.2010.08.001
Ridder, L. V., Hakvoort, W., Dijk, J. V., Lötters, J., & Boer, A. D. (2014). Quantification of
the influence of external vibrations on the measurement error of a Coriolis mass-flow
meter. Flow Measurement and Instrumentation,40, 39-49.
doi:10.1016/j.flowmeasinst.2014.08.005
Clark, C., & Cheesewright, R. (2003). The influence upon Coriolis mass flow meters of
external vibrations at selected frequencies. Flow Measurement and Instrumentation,14(1-2),
33-42. doi:10.1016/s0955-5986(02)00065-1
Henry, M., Tombs, M., Duta, M., Zhou, F., Mercado, R., Kenyery, F., . . . Langansan, R.
(2006). Two-phase flow metering of heavy oil using a Coriolis mass flow meter: A case
study. Flow Measurement and Instrumentation,17(6), 399-413.
doi:10.1016/j.flowmeasinst.2006.07.008
Basse, N. T. (2014). A review of the theory of Coriolis flowmeter measurement errors due to
entrained particles. Flow Measurement and Instrumentation,37, 107-118.
doi:10.1016/j.flowmeasinst.2014.03.009
Heavy Oil Vs. Light Oil. (2011, March). Retrieved December 10, 2018, from
http://www.aoga.org/wp-content/uploads/2011/03/HRES-3.10.11-Lunch-Learn-BP-HeavyOil1.pdf
4
APPENDIX A – GANTT CHART
5