Turkish Journal of Electrical Engineering & Computer Sciences Turk J Elec Eng & Comp Sci

(2013) 21: 26 37
http://journals.tubitak.gov.tr/elektrik/
ITAK
c TUB
Research Article doi:10.3906/elk-1105-45

The vortex effect of Francis turbine in electric power generation

Veli TURKMENO
GLU

Electrical and Energy Department, Ordu University, 52200 Ordu, Turkey

Received: 23.05.2011 Accepted: 15.08.2011 Published Online: 27.12.2012 Printed: 28.01.2013

Abstract: In this study, the vibration effects of a vortex that occurred in high-head Francis turbines and an alternator
are examined. The vortex effect, which directly affects the efficiency and the quality of the energy, was tested at the
Darca-1 hydroelectric power plant (HPP) located in Ordu Province, Turkey. Formed by undissolved oxygen in the water,
the vortex effect, which is parallel to the alternator load, causes tremendous vibration within the alternator and Francis
turbine bearings. This problem, which has a direct negative effect on the alternator capacity, was solved by adding an
air-admission system. In doing so, power production was increased by 11.11%, from 44 MW to 49.5 MW. This caused a
significant head loss, specifically in the Francis turbines. Vortex optimization was successfully established at the Darca-1
HPP.
Key words: Francis turbine, vortex effect, alternator, hydroelectric power plant

1. Introduction
Hydroplants and various hydromechanisms generally have vibration-related issues. Structural problems are
likely to be caused by the extensive amount of water stored in places such as pipelines, tunnels, penstocks,
and draft tubes. When the frequency of the oscillations coincide and resonate with the natural frequency of
a hydraulic, mechanical, or structural component, the amplitudes of the oscillations are reinforced, and the
potential for damage is maximized.
The most damaging cause of excitation is the draft tube vortex core in Francis turbines. The intensity
of the air significantly affects the excitation of the natural frequencies of a turbine water passage system and
the response of the system. It is crucial to consider the airwater flow through the turbine draft tube. Keeping
turbine loads within the optimal operating zone, far from the zone of vortex core resonance, will increase plant
profitability and reliability [1].
There are many studies on turbine vortex effect [2,3]. Researchers have proposed various techniques to
lessen the impact of vortex and vibrations [4]. Air admission into the draft tube is the most commonly used
technique among them [5]. Meanwhile, a new method has been offered by Susan-Resiga et al. [6,7], which uses
an axial jet made from the crow tip to reduce the amount of vortex rope.
Empirical studies have looked at the irregularities of the pressure caused by the vortex rope in the draft
tube of Francis turbines [8,9]. Another study tested the use of scale model Francis turbines at the EPFL
Laboratory for Hydraulic Machines [10]. Another study offered information about the operation of Francis
turbines and their complex hydrodynamic system [11].
Correspondence: turkmen67@hotmail.com

26

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

At the upper part of load operation under low cavitation numbers, scale models of high-specific-speed
Francis turbines may present pressure fluctuation in the range 2 to 4 times f n , the rotating frequency. In the
framework of the FLINDT project, pressure measurements on the draft tube wall of a Francis turbine scale
model, at 104 locations, revealed such a phenomenon at a frequency of almost 2.5 f n . The phase shift analysis
of the measured compression instabilities in the draft tube at this frequency points out a pressure source located
in the inner part of the draft tube elbow [12].
An analysis of the dynamic pressures causing danger during operation in the Francis turbine runner was
carried out and it was found that the main cause for the damage in the runner blade was the hydraulic forces
creating dynamic pressure [13].
In another study, researchers used wavelet analysis to investigate the static pressure pulsation in the
cavitated vortex core in the draft tube of a Francis turbine and found that there was a significant correlation
between the processes [14].
In order to increase the amount of dissolved oxygen (DO level) in hydroelectric plants, researchers have
also used models where they injected air into the Francis turbines [15].
This study presents our ongoing developments of the jet control technique for swirling flows in hydraulic
turbines. In order to determine the best jet configuration for mitigating the vortex rope, researchers have
developed a special swirling jet apparatus that is able to generate swirling flows similar to the ones downstream
from Francis turbine runners operated at partial discharge [16]. In electricity generation, studies of network
security, diagnostics, and the alternator are all very important [17,18].

2. Theoretical approach
Because it is in line with the theoretical infrastructure of the optimization at the Darca-1 hydroelectric power
plant (HPP), located in Ordu Province, Turkey, one theoretical approach used in a study at the Laboratory
for Hydraulic Machines (LMH; Lausanne, Switzerland) is presented here, related to undissolved oxygen-formed
vibration [15].
Turbine tests regarding the DO levels were carried out at the LMH laboratory, where knowledge pa-
rameters governing gas dissolution in liquids and the analysis of each of the parameters were presented as a
prerequisite.
As stated in Ficks law [19], gas dissolution is mainly affected by the following parameters: exchange area
A, control volume V , contact time t, saturation concentration Cs , and initial and final concentrations C1 and
C2 , stating a coefficient of diffusion KL and meeting the requirements of the differential equation below.

C A
= KL (Cs C) (1)
t V

Presuming a continuous coefficient of diffusion, Eq. (1) can be incorporated as follows:

 
Cs C2 A
ln = KL t. (2)
Cs C1 V

The final concentration is reached as follows:

C2 = Cs (Cs C1 ) e(KL V t) .
A
(3)

Eq. (3) can be used to find the final concentration in the river downstream from the power plants.

27

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

In Henrys law [20], the increase in the liquid temperature decreases saturation concentration Cs of gas
in liquids, while it increases with pressure. In a study by Thompson and Gulliver [21], the relationship between
the prototype (p) and the model ( m) about the exchange area A is stated as follows:

  3/5  3/5  6/5  4/5

Ap p p m np Dp
= , (4)
Am m m p nm Dm

where = the gas volume ratio, = the water density, = the surface tension, n = the rotational speed,
and D = the runner diameter. The existing experiments have used various volumes of gas during the tests;
however, with p = m and p / m, the ratio continues and equals unity. Moreover, because the water
temperature will be similar between the model and the prototype, Eq. (4) can be reduced as follows:

  4/5
Ap np Dp
= . (5)
Am nm Dm

According to Ficks law, the oxygenated water is represented by volume V , which is the through flow in a
turbine. In terms of volumes, there is a proportion between the ratio of the prototype ( p) and the model (m)
to the square of the ratio of the 2 diameters ( D) and the square root of the corresponding heads ( H) :

 2 r
Vp Dp Hp
= . (6)
Vm Dm Hm

As with the contact time and the pressure, the 2 phases of dissolution must be differentiated from each other,
the diffuser being the first and the exit channel or tail race the second. Approximations need to be done in
accordance with the difference between a models section downstream and the prototypes tail. For the current
test, the coefficient of diffusion has been assumed to be the same as in the diffuser and the exit channel. The
ratio of the proportion between the contact time and the geometric scale and the square root of the heads is as
follows:
s
tp Dp Hm nm
= = . (7)
tm Dm Hp np

Because only the decrease in the speed of the bubbles has significance in the tail race, the contact time is
determined by the geometric scale and the level of upstream water. Similarly, there is a proportion between the
pressure ratio with an impact on the saturation concentration level,Cs , and the geometric scale and the level of
upstream, and the square root of the heads in the turbine, as well as the geometric scale and the level of water
upstream in the exit channel or tail race, which is as follows:

s
Pp Dp Hm Z2p nm Z2p
= = . (8)
Pm Dm Hp Z2m np Z2m

Figure 1 shows a simulation of the vortex rope motion in the draft tube.

28

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

Figure 1. Vortex rope motion in the draft tube [13].

3. Materials and methods

3.1. Vibration measurement
To measure the vibration, a DTM10 distributed-vibration transmitter-monitor, as shown in Figure 2, is used.
With proximity probes and a Modbus interface to a programmable logic controller or distributed control system,
it is recommended to use a DTM10 distributed-vibration transmitter-monitor to check the vibration in the
machine via integration. Redundant power supplies and redundant 420 mA transmissions are also provided
with the DTM10. It is possible to create an interface between the DTM10 and any other proximity probe
system, which can be done with the existing hardware.

Figure 2. DTM10 distributed-vibration transmitter- Figure 3. TM180 proximity probe transducer [22].
monitor [22].

In order for the DTM10 vibration sensor to perceive the vibrations, as shown in Figure 3, TM180 proximity

29

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

probe transducers check the static and dynamic distance between the target and the probe. The architecture
of the TM series proximity probe system is in line with the standards of API 670. There are probes, extension
cables, and drivers in every system. Figure 4 shows the DTM10 system installation.

4-20mA+(1)
BUF
Butter
Out

COM

COM
PWR

BUF
SIG
TM0200 or
PLC, DCS

TM0200 or
PLC, DCS

COM
RST/BYP
TIRP/M
COM
COM
4-20mA+(2)
Inp/Multiply
or phase
reference
input/output
PROVIBTECH
POW-24V

Remote
COM
COM

Reset
SIG

TM 182 Driver OK
ALT

TM0181 DNG BUF

Cable
BYP
TRX

TM0180
Proximity Probe RS-485+
RS-485- PLC/DCS or
BUS(+)
BUS(-)

DTM-CFG
ARM2
COM

NO2
NC2
To
Danger

To
Alert
24V+(1)
GND
24V+(2)
NC1
ARM1
NO1
24V+(1)

24V+(2)
GND

Power Power
Supply 1 Supply 1

Figure 4. DTM10 system installation [22].

3.2. Air-admission system

Figure 5 shows the air-admission system used to prevent the vortex effect at the Darca-1 HPP. With part 3 of
the slip ring cover valve, shown in the upper section of Figure 5, the amount of air is admitted through the
shaft. Normally, this valve is left fully open for maximum air admission.
The Francis turbine section of the air-admission system is presented in Figure 6. Here, owing to the check
valve in the lower part, airflow is directed upwardly in a one-way manner. Using a check valve in the lower
section, pressured water is not permitted to flow through the turbinesgenerator shaft to the HPP building for
any reason. Due to centrally injected air, the vortex oscillation amplitude is decreased; thus, hitting the draft
tube is prevented. As a result, the turbinegenerator group can operate more efficiently.
Figure 7 shows the data from the turbine and generators used at the Darca-1 HPP. Either of the 2 units
was operating while the vibration was being measured.

30

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

Air admission system

slip ring cover part 3
SECTION B-B

480
440 z=8+2
8x
M10x33
Hex Head Screw
8x
Hex Nut

Air admission system

slip ring cover part 1
420

8x
M12x25
Hex Head Screw
590 z=8
622

Air admission system

slip ring cover
641

O-ring d6

380 6x
M16x50
340 z=6 Hex Head Screw

Generator shaft

Figure 5. Generator section [courtesy of Voith Hydro GmbH & Co. KG, Austria].
Flow direction

Turbine shaft
O-ring d6 General Data:
-Quantitiy 2 units
- Speed: 600 rpm
Air system - Runaway speed: 964 rpm
drawing runner - Rotation: clockwise
Runner
Check valve
Generator Data:
Air system drawing - Rated output: 56 MVA
- Rated voltage: 13.8 kV
slip ring cover 2
- Rated current: 2343 A
- Frequency: 50 Hz
- Power factor: 0.90 cos phi

Turbine Data:
- Rated head: 302 mwc
- Rated power: 49.5 MW
A - Rated discharge: 18 m3/s

Figure 6. Francis turbine section [courtesy of Voith Hy- Figure 7. Darca-1 hydropower turbine and generator
dro GmbH & Co. KG, Austria]. data.

31

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

Figure 8. Water flow computer monitoring.

Figure 8 presents the HPP SCADA system, software that was developed by Voith. The immediate
turbine regulator controls, power, voltage, current, frequency, power factor, source water level control, main
suction hatch controls, and check valve controls of the respective units (unit 1 and unit 2) can be seen and
performed by the monitor.

Figure 9. Francis turbine computer monitoring.

32

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

In Figure 9, functions such as the monitoring and control of the spherical valve, tracking of the helix
pressure, measuring of the turbine flow, positioning of the turbine hatch, speed check, and general monitoring
can be performed.

Figure 10. Francis turbine upper and lower bearings.

In Figure 10, the control and monitoring of the generators upper and lower bearings, turbine bearings,
and stator temperature can be observed.
Figure 11 shows the upper section of a generator without an air-admission system. Figure 12 shows the
upper section of a generator with an air-admission system mounted onto it. With the valve located on the top,
the air-admission rate can be adjusted manually. Based on the experiments performed, this valve is adjusted
to the fully open position in the case of the Darca-1 HPP.

Figure 11. Generator without an air-admission system. Figure 12. Generator with air admission system.

33

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

4. Results
As in Darca-1, a generator that operates without an air-admission system can produce only 44 MW of power
despite the fact that its capacity in the safe-vibration range is 49.5 MW. However, after installation of the
air-admission system, the generator is now able to produce power under 49.5 MW at full load, within the safe
vibration range.
Figure 13 shows the power production of a generator without an air-admission system and bearing
vibrations.

Figure 13. Power production of a generator without an air-admission system and bearing vibrations.

Figure 14 shows the measured rates of a system without an air-admission tool. These rates were obtained
from the Darca-1 HPP databank, during a period of 3 min and 14 s, on 17.11.2009 at 1050 hours. Here, the
significant rates shown in the cursor 1 and cursor 2 columns are circled. In the case of the system without air

Figure 14. Measurement panel without air admission.

34

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

admission, for the turbine bearings relative radial vibration, cursor 1 = 551.8519 ym and cursor 2 = 533.3333
ym; for the actual power, cursor 1 = 47.3047 MW and cursor 2 = 49.7583 MW. The steady-state actual power
was measured as 44.0837 MW and the turbine bearings relative radial vibration was 475.9259 ym.

Figure 15. Power production of a generator with an air-admission system and bearing vibrations.

Figure 15 shows the vibration rates of the turbinegenerator group with an air-admission system and
the actual power produced by the generator. The data used in the study was obtained from the Darca-1 HPP
databank during a period of 3 min and 42 s, on 07.05.2010 at 0845 hours. Based on the results, for the turbine
bearings relative radial vibration, cursor 1 = 355.5555 ym and cursor 2 = 370.3704 ym; for the actual power,
cursor 1 = 49.4023 MW and cursor 2 = 49.3393 MW. The steady-state actual power was measured as 49.5406
MW and the turbine bearings relative radial vibration was 372.2224 ym.
Figure 16 shows the measured rates of a system with an air-admission tool.

Figure 16. Measurement panel with air admission.

35

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

5. Conclusion
In this study, the effect of the Francis turbine air-admission system on both the efficiency of the generators
power production and the vibration of the turbine has been investigated. Due to the air-admission system, it
was determined that the actual power production increased by 11.11% and reached 49.5 MW, and the relative
radial vibration of the Francis turbine decreased by 27.68%. Because the generator operates at full capacity, the
power production was increased about 4,752,000 KWh annually. Considering that the vibration of the Francis
turbine decreased by 27.68%, its contribution to the life cycle of the turbine is clear. This improvement will
decrease power cutoffs and positively affect the periodic maintenance of the turbine. One of the most important
parts of the turbines is their bearings; the lower the amplitude, the better the operating time of the bearings. It
is obvious that the decrease in vibration will improve the operating time of the bearings for both the generator
and the turbines.

Acknowledgments
The author would like to express his gratitude to Voith Hydro for granting permission to study the Darca-1
air-admission system and for sharing the technical photos of the system. Thanks are also given to Yapsan Enerji
A.. for allowing the author to conduct a study at the Darca-1 HPP facilities and for their contributions and
valuable information.

References

[1] S. Pejovic, Understanding the effects of draft tube vortex core resonance, Hydro Review Worldwide, Vol. 8, pp.
2833, 2000.
[2] O. Kirschner, A. Ruprecht, E. G
ode, Experimental investigation of pressure pulsation in a simplified draft tube,
Proceedings of the 3rd IAHR International Meeting of the Workgroup on Cavitation and Dynamic Problems in
Hydraulic Machinery and Systems, pp. 5564, 2009.
[3] J. Paik, M. Sotiropoulos, M.J. Sale, Numerical simulation of swirling flow in complex hydroturbine draft tube
using unsteady statistical turbulence models, Journal of Hydraulic Engineering, Vol. 131, pp. 441456, 2005.
[4] R.H. Thicke, Practical solutions for draft tube instability, Water Power & Dam Construction, Vol. 33, pp. 3137,
1981.
[5] B. Papillon, J. Kirejczyk, M. Sabourin, Atmospheric air admission in hydro turbines, Hydrovision, paper 3C,
2000.
[6] R. Susan-Resiga, T.C. Vu, S. Muntean, G.D. Ciocan, B. Nennemann, Jet control of the draft tube vortex rope
in Francis turbines at partial discharge, Proceedings of the 23rd IAHR Symposium on Hydraulic Machinery and
Systems, paper F192, 2006.
[7] R. Susan-Resiga, S. Muntean, V. Hasmatuchi, F. Avellan, I. Anton, Analysis and prevention of vortex breakdown in
the simplified discharge cone of a Francis turbine, Journal of Fluids Engineering, Vol. 132, doi:10.1115/1.4001486,
2010.
[8] A. Bosioc, C. Tanasa, S. Muntean, R.F. Susan-Resiga, Unsteady pressure measurements and numerical investiga-
tion of the jet control method in a conical diffuser with swirling flow, Proceedings of the 25th IAHR Symposium
on Hydraulic Machinery and Systems, 2010.
[9] A. Baya, S. Muntean, V.C. C ampian, A. Cuzmo, M. Diaconescu, G. Balan, Experimental investigations of the
unsteady flow in a Francis turbine draft tube cone, 25th IAHR Symposium on Hydraulic Machinery and Systems,
Vol. 12, doi:10.1088/1755-1315/12/1/012007, 2010.
[10] C. Nicolet, A. Zobeiri, P. Maruzewski, F. Avellan, On the upper part load vortex rope in Francis turbine:
experimental investigation, 25th IAHR Symposium on Hydraulic Machinery and Systems, doi:10.1088/1755-
1315/12/1/012053, 2010.

36

TURKMENO
GLU/Turk J Elec Eng & Comp Sci

[11] M.S. Iliescu, G.D. Ciocan, F. Avellan, Analysis of the cavitating draft tube vortex in a Francis turbine using particle
image velocimetry measurements in two-phase flow, Journal of Fluids Engineering, Vol. 130, doi:10.1115/1.2813052,
2008.
[12] C. Nicolet, J. Arpe, F. Avellan, Identification and modeling of pressure fluctuations of a Francis turbine scale
model at part load operation, 22nd IAHR Symposium on Hydraulic Machinery and Systems, 2004.
[13] R. Xiao, Z. Wang, Y. Luo, Dynamic stresses in a Francis turbine runner based on fluid-structure interaction
analysis, Tsinghua Science and Technology, Vol. 13, pp. 587592, 2008.
[14] B. Sirok, B. Blagojevic, T. Bajcar, F. Trenc, Simultaneous study of pressure pulsation and structural fluctuations
of a cavitated vortex core in the draft tube of a Francis turbine, Journal of Hydraulic Research, Vol. 41, pp.
541548, 2003.

[15] R. Fraser, N. Desy, E. Demers, C. Deschenes, J.P. Fau, S. Hamel, Improvements to self-
venting Francis turbines, http://lamh.gmc.ulaval.ca/fileadmin/docs/Publications Recentes/Improvements to Self-
Ventinf Francis Turbines.pdf, accessed January 2011.
[16] R. Susan-Resiga, S. Muntean, V. Hasmauchi, S. Bernard, Development of a swirling flow control technique for Fran-
cis turbines operated at partial discharge, 3rd German-Romanian Workshop on Turbomachinery Hydrodynamics,
2007.
[17] A. Khan, O. Ozgonenel, M.A. Rahman, Wavelet transform based protection of stator faults in synchronous
generators, Electric Power Components and Systems, Vol. 35, pp. 625637, 2007
[18] B. Vural, A. Kzl, M. Uzunolu, A power quality monitoring system based on MATLAB server pages Turkish
Journal of Electrical Engineering & Computer Sciences, Vol. 18, pp. 313325, 2010.
[19] G. Degremont, Memento technique de leau, Technique et Documentation, Paris, 1972.
[20] J. Montgomery, Water Treatment Principles & Design, New York, John Wiley & Sons, 1985.
[21] E.J. Thompson, J.S. Gulliver, Oxygen transfer similitude for vented hydroturbine, Journal of Hydraulic Engi-
neering, Vol. 123, pp. 529538, 1997.
[22] ProvibTech, http://www.provibtech.com, accessed January 2011.

37