4 82

Development of a Positive Displacement Micro-Hydro Turbine'

Dousith PHOMMACHANH**, Junichi KUROKAWA***,

Young-Do CHOI"' and Noboru NAKAJIMA""

The objective of this study is to develop an efficient turbine that can be used to extract
micro hydropower potential of a water supply system. For the case of high head and critical
low flow rate rangae of micro hydropower resources, it requires very low specific speed tur-
bines which are lower than conventional impulse turbines' specific speed. For this purpose,
we develop a new Positive Displacement Turbine (PDT). In order to reveal the performance
characteristics of the new turbine, one conventional impulse turbine, which is used for auto-
matic water faucet system, was tested for comparison. The test results show that the PDT was
much more efficient than a conventional turbine and it can sustain high efficiency under the
wide range of operating conditions. In addition, the ef ciency of the PDT is much improved
when reducing its side clearance. The pressure pulsations at the inlet and outlet of the PDT
can be considerably minimized by using simple dampers.

Key Words : Turbomachine, Water Turbine, Positive Displacement Turbine, Water Supply
System, Performance Improvement

able. In the case of high head and critical low fiow rate
1. Introduction micro hydropower resource, it requires very low specific

There are many available micro-hydropower re- speed turbines which are lower than conventional impulse
sources surrounding us for extraction, such as rivers, mu- turbine particularly the Pelton type, which works best at
nicipal water supply systems of towns, drainage water high head and low fiow rate (known as lowest specific
from houses, irrigation canal and so on. The least expen- speed turbine).

sive small hydropower systems are those that utilize the The pressure in the water supply system (WSS) is
pipe systems(1). These systems only require the placing normally much higher than required. There is a lar.ae
of a turbine on the pipe and no major/no civil works are amount of dissipated energ'y due to a high pressure dif-
required. Recently,' several types of turbines have been ferential at the inlet and outlet of water faucets. A simple

used for extracting the energy from micro hydropower re-
sources(2) (4). The selection of the best turbine for any
particular hydro site depends on the site characteristics(4) ,
the dominant factors being the head and fiow rate avail-
Received 26th September. 2005 (No. 05-4 1 68)
" ivision of Systems Integration, Graduate School of Engi-
neering, Yokohama National University, 79-5 Tokiwadai,
Hodogaya-ku. Yokohama 240 8501, Japan.
E-mail: dousith @ mach.me.ynu.ac jp
"" ivision of Systems Research, Faculty of Engineer-
ing, Yokohama National University, 79-5 Tokiwadal,
Hodogaya-ku. Yokohama ('-40-850 1 . Japan.
E-mail: kuro@post.me.ynu.acjp; ydchoi @mach.me.ynu.
ac jp
""' ngineerin_,_a Group, CATV Products Division, URO
ELECTRONICS IND. CO., LTD., 5-27-lO Minami-Ohi,
Shinagawa-ku, Tokyo 140-0013, Japan_
E-mail : nakajima-noboru @uro.co jp
calculation reveals, that each time we use a water faucet,
l O to 80 W of power is dissipated. A simple calculation for
selecting an appropriate turbine within the water supply
system can be used. For example: if an overall e :ciency
about 60%, turbine is connected to a generator with a
speed of I OOO min t , for every maximum available power
points of water faucets (for pressure > O. 10 Mpa), the spe-
cific speed (ns = nPl/2/H5/4) wil-1 be in the range of 2.2 to
4.8 and the average specific speed is about 3.7 [min I m,
kW]. For this specific speed, there is no suitable turbine
type, because the impulse turbine has the specific speed
range of 10 to 70(4)-(6). To realize the present requirement
of specific speed, we proposed a new turbine type, which
is a Positive Displacement Turbine (PDT).
The objective of this study is two-fold: First, we want
to study the characteristics of the WSS and estimate its
micro hydropower potential for power generation at the

Series B, V ol.49, No. 2 2006 JS!}/E Internotional Journ.al

 483

I
water faucets. Secondly, we want to develop the new tur-

,
- l -'- , Locationl @ BN0.3 IB
0.5 'It* O¥ -O-MA Location 2 -A-B N.'0.3 IF

t
bine which has very low specific speed, simple structure

¥
-O- LLA Locatbn ! 1- 5 N0.3 2F
¥

and hi_gh efficiency.

t'b
:1L. ' 0
-A-LLA Locatbn ' V-B
-JL-LLA Locatbn S B
N. 0-3
?_?0.5
3P
4F

¥
It I ,
l
-e-B N0.3 5F

'

ol
L
-- t -1-; *¥?o.: IP -JL-B Noi 6F

,,
:

t:
Nomenclature
'L o .4
r -V-B N02 IF -Cl-B N0.3 7F

.
)_' : t ¥ :" t*t -V-3
-:o:
+1t l=
_ _b:BN'a;0.
e" 2F
B No
3F It-e-
-e-BBB N0-4 6F :JL" Ii' ¥ ::¥¥¥ ' _e_B
'_ 4F
-It - B N0.4 2B

-e -B N0-4 7F
B : Height of rotor [m]
c" 1!X-B Nlo'2 -L -O - B + B No') SF

1_ ¥ 6F
-O-B N0.4 IaF

it
0.3 tl :1 ¥ 10 ¥¥It:
: -(: BB No.4 15F

) 5}: 'II t:b :{:

ILL
lb¥ ¥"':

¥
D : Diameter [m]
---
'L -- f4: 't *, H t! ¥f¥ ,Lb *¥¥
o
+ t .L
- J
'
f : Frequency [Hz] ic '¥ ,1F¥ ' : :L, 'L't
; 0.2 $Jb , $ .
' It 1,L¥ )a L o¥¥*,L

¥
( 1!,¥111 I

l
H : Head [m] ' - _l ¥ ' 1'b ¥t ,L tb
L ,L¥ IL., ¥

,L
-_r--O
. _ ¥'
llL
It,L t V : ;L L L¥
LON : Number of the lobe order
- ll: '* -1L
¥'1'1
l',t , ,1Ll,' :L,
L I " 'L 'b
0.1 e*r:"'bllt ; t""': t "L:'t : "I
:'1L:'L-JL t" ,L " :'tlI" ¥I 'L

--_ __ LAO
,L,
1""'I '
-- t'-- L ':L " :,, A
.,. t

;
'L¥
n : Revolution speed [min I J J:'1E V

Jtititt
ILa
-J-

1
ILt :'¥ ' 'L I 'b"' 'b

1*
'L' 'bILIL 'b

'
' l ",L' , ,L ,L,t -It ""b
ns : Specific speed [min 1 m, kW] L ll i¥
'Lr :b" ¥x' "Lt q+ F,,' LL 7

,;
LL
OJ'

o
N,, : Unit speed [min 1 J *

L
tt

o
a
p : Pressure [MPa]
P : Power [W]
Flow Rate. Q [1/minj
B = Buildhlg MA = Mountain Areah LLA - Low Land Area
20 3
P** : Unit power [W] Fig. l Inlet pressure and fiow rate relationship of water faucet
Q : Flow rate [m3/s] of water supply systems
T : Mechanical toque [Nm]
Vd : Volumetric displacement [m3/rad]

t lft t l I I tf , I I , I i I l li ttI t, li
L ':,
, : ;S , -' - MA bocation l

;
t , i' ? 1,, --- -O - MA Lccatiol' 2
Greek symbols l OO
I : 'e':?..,
I F ; fl :S -O - LLA LQcatian I
tJ -A- LLA Location 2
: Efficiency I / / l . 8 .:! "' IL -A- LLA Location 3

l
t '// t
, ･ / -l-BNo.1 Ir

1 t
p : Density [kg/m3]

,
_ 80

t
-V-B INc-2 IF

t t
d ILI 'S: F.
'L
-V- B No.2 ･-F
.*

a, : Angular velocity [rad/s] l j

1'
;$' r L : B N;0.2 4P
B No.'- 3F
:

t
l'll /a .L¥
'bL ,'
-+- B N0.2 SF
Subscripts
Sl
1' .'. /// -1 IL
¥ L -X- 8 N0.2 6F
o J ;. t / t -@-BN.io_3 IB
60

1
-

l : Losses or leakage t'L

t L¥ lL¥ Itl, i
1 -A-BN0.3 IP
i :S:S'

l/ ¥
o i ' .1 ,1 / //' t
,g

:? ,thL
. t1 - l
-BNo_3 ･-F
: : BN0-3 3P
o : Output 1'b

L'L tL
It It B N0.3 4P
l

1 'b a
_ l- Lt L1 l' t -O-BN0.3 5F

tl
1
:

th : Theoretical < 40 l a ¥ l -A-BN0.3 6F

l

1 It t l
l'J!!' t' t t tt Il D-BN0.3 IF
-. t*
'! 1F . SLItht
I'tI tl
:':t ,t?
- r-BN0.42B
,.l / l t IL¥ h It 8 _e_BNo.4 6F
l l

2. Water Supply System & Its Miero Hydropower 'l l'i L IL

l L Ib$¥11 IL tt h , -e-B No.4 IF
IL

* t 'L lL L,LI tt
t -O-Bl¥'o_4 tOf
J-
1

tt
Potential 20 - -t Ll. tA tl { -0-3N0.4 ISF
- t 'LI 1ltt t tt
t e:O'-hi･ $?e
L t I It i t ,
/ tii
*,bI ti ,LI
Lt l t t
The normal fiow rate range at the outlet of the faucets
, ll t
o

¥lL L t t I tt l
o

in houses, ofiices etc. is about four to 10L/min which is : O

the interesting range for this study, as it is shown in Figs. 1
and 2 the range between two vertical dashed lines.
To make water fiovv from the faucets requires only
a small differential pressure between the inlet and outlet.
However, to ensure water fiows to all users, the utilities
always supply a higher water pressure than is required,
In this study, difilerent sources of water supply sys-
tems (residential houses, office buildings, etc.) in the city
of Yokohama were investigated. The results are shown in
Figs. I and 2. Figoure I shows the pressure and fiow rela-
tionship of the water faucets tested. The maximum pres-
sure and fiow rate are in the ran_ge of 0.1 -0.54MPa and
1 1 - 36L/min, respectively, which are much higher than
the normal requirement. All of the houses use water from
the water supply utilities while most of the high-rise build-
ings require a booster system or pump to deliver the water
to the top of building, where it can be distributed to each
fioor.
We can see in Fi_g. I that not all the curves are paral-
lel to each other as they should be. This may be because
of the different pipe diameters. From the data in Fig. I ,
the estimation of the available micro hydropower poten-
20 30
Flow Rate, Q [Vtnin]
B = Building, MA = Mountah Are LLA = Low Land Area
Fig. '- Available power, fiow rate and pressure relationship of
WSS in Yokohama
tial of the water supply system at the end used valves can
be computed by P =pgQH. The result is shown in Fig. 2.
The maximum available hydropower varies between I O W
to 100 W.
3. Test Apparatus and Method
3. I Testing facilities
Figure 3 shows the schematic view of the testing sys-
tem. The designed water supply system has the ability
to test the turbine in the range of 0-60L/min and O-
0.7 MPa of fiow rate and pressure, respectively. Water
is pumped from the water tank (1) and is supplied to the
system by multi-stage pump (2). The effective head H is
varied by adjusting the rotational speed of the pump and
adjusting the pressure control valve (12). The amount of
water fiow to the turbine is controlled by the valve (6).
While the valve (1 1 ) is used for controlling the discharge

JSM:E International Joumal Series B, ¥, ol. 49, No. 2, 2006

484

!
( I )water tank ({o) outlet pressure gauge
(2)Mutti stage pump (1 1) Disc arccre valve (9) (8) Bearings

¥
(3)Filter (12) pressure controi valve 1 L s 1 17)

i
( o)
(4 ) Piow meter
If
(s) pump control unit ! r jvl
!" I n (6)1r'c ._
({ 1)

T: 'tt: i' ___LI

(s) Inlet valve r ?
( 7 ) :nlet pressule ' auge
f i

'i
:1bc･!L; cPhr:sur _
il '4 ; *: ' - fl ! I

; ___
{:)TD: :gau:e j: j '! lJtii:J( 2;?: :;: s f r : r :t: {TJ':L- -= ) {

i
i I

1 I
;Ti LI s :

I t'l-: I t'
t; _ "i _Il J
(;) I !: [] :i :i
f '

i! *

Ij
( )
w ., i '

:i
+.,
'l -- (5 {

I
L

I?1 't
{

i
c* -
Rotors
's :e
Casina_

Fig. 3 Schematic view of testing system Shafts

Cover
Weiglxing Scale (a) Whole view of PDT

Torque meter '

. ,
ii ,;J :r

¥
Ill

)
Turbine

l
' r Ti r'
!

I - T< , '-
Pulley
'1 _ .'
l
'"*

. i iiel-
i
*I ;

:IE

__!__. s {
I

J: __ {
s
i
s

: i I ii (b) Rotor profile

!
i

JJi Fig. 5 Positive displacement turbine design

rr! are made of hard plastic and have a diameter of 2 1 .2 mm

L JWeigllt
ql}

and helght of 1 6 4 mm. The rotors are tightly fixed with

shafts by spring-pins. To improve the torque transmission
Fig. 4 Testing bench between the two rotors, the rotors are designed to have a
spur gear around the curve shape. The gear on each rotor
pressure. has 33 teeth, see Fig. 5 (b).
The turbine-testing bench (9) is designed for measur- In ideal PDT, there is no leakage from inlet to out-
ing the turbine performance, as shown in Fig. 4. The tur- let ports and the rotational speed of the turbine is directly
bine shaft is connected to one end of the torque meter by proportional to the flow rate. Thus, the main interest of
a coupling and the other end of the torque meter is con- the study is to determine the ef iciency of the PDT and the
nected, either to an electric generator or mechanical load, influence of the leakage on ef :ciency.
The mechanical load can measure the torque of the turbine 3. 3 Experimental method
by applying the load to the turbine (called a weighing scale Generally, hydraulic turbines are designed to operate
torque). The weighing scale torque meter gives a very ac- at a certain values of revolution speed, discharge, effec-
curate torque measurement when the generating heat from tive head, power and efficiency(7). However, micro water
the friction between pulley and rope is fully cooled. faucet turbine operates under the wide range of effective
3.2 Turbine design head but narrow range of fiow rate, Therefore, it is nec-
Figure 5 shows a positive displacement micro-hydro essary to test the performance characteristic of the turbine
turbine design_ A rotary three-lobe turbine was selected under the constant fiow rate.
because of its simplicity. The fluid flows into the turbine's Tests are performed by maintaining a constant flow
inlet port under press'ure. Because pressure at the inlet rate and the speed (n) is varied by changmg the load on
port is greater than at outlet port, fluid can cause the rotors the turbine. For each value of n, the torque and differential
to rota.te against the load torque. The whole view of the pressure across the turbine are measured. Similar tests are
turbine is shown in Fig. 5 (a). The turbine consists of two performed at different fiow rates. The value of input power
rotors, two shafts, four bearmgs, casin_a,_, and cover. Bear- P and output power Po can be computed by P =pgQH and
ings support both ends of the shafts with the three-lobe ro- Po = Ta), respectively. These powers determine the overall
tors located between bearings. This resists shaft defip_ction efi:ciency of the turbine by using q = Po/ P.
and contact between the rotors and casing wall. The rotors The performance curves_ of the turbine are normally

Seties B,¥,' oi.49s iNo. 2 2006 JSME In:ternatlo/ al Jour'l al

 485

plotted in the form of ef ciency against revolution speed, Positive displacement turbine

CF D _ . - 6Vm rl

l
o -g 4lfmin
power against revolution speed, flow rate against revolu-
A a _AIQ- A 8linh
tion speed etc. It is common practice in turbo-type ma- :fs ¥ -o l 1'-Vmh
Q I Ollmin

chine to use the unit speed and unit power which are de- a tAt
t Conventio! l turbhe

t
,
. e Q 3Vivlin
rived from the similarity law to reveal the performance
-o ¥ t tlt tO A Q 4V n
characteristics of turbines(7).
o
OF¥ ¥ tt
cf Q 4_SVtnin
IJL l
' v' O:=5VhliD
Unit speed and unit power are expressed as follows:
.
::) c 0.5
S I ttt,) at
L tt
tt 't
t -
Q::6Vhlin
o O : 1Vhlin
Lt LtIIltlLt"
tt
nD L
t I ItQ
4 t
lI LI It
LL tt Itl
tlt
P

vo, 'gr vo lll

y ' eFaF Uo "' 'Ttl l *^ -

O
P,, = D2 3/-? (for conventional turbine) (2) a : 1
L LL

O I O ･_O 40
30
P = Po (for new turbine (PDT)) (3) Unit speed, iV*,

(DB)H3n Fig. 6 Variation of unit power with unit speed

In order to compare the performance characteristics
of the new turbine with a conventional turbine, a small im- 0.8
?osit ive disp lacement turbine
pulse turbine was tested. Conventional turbine is referred -c 2*4Vmin
-' 6Vmin
to a small Turgo impulse type, which has developed and -A I 8Vmin
0.6 -Lr i ! OVmin
used for automatic water faucets for several years now. :a:
-o IQ 1 2Vmin

tOIL
;bo
Conventional turbine
o t( AJL :'
This turbine has two nozzles and each of them has a jet at l
li¥ ! 31/min
area of 3.14mm2. The turbine runner has a diameter of oc 0.4
'S
¥ ¥ lL t JL Q 4Vmin
4.5Vmin
¥4:LJLt L
v Q 5Vrnin
22 mm, which is nearly equal to the PDT's diameter, with Q
¥ L I 6Vmin
14 buckets. L.QtIt ttttl:!
o o 7umin
_
Both the Turgo impulse turbine and new turbine were
0.2
v oI¥¥ ¥LLLq
lL o oo r
t t 01
tested using a constant fiow rate test-method. As it was t 1 IL
L tLL
oovolP
1 1L LItL
L
O

mentioned earlier, the water turbines are design to work at

certain values of revolution speed, discharge and effective
head. However, turbines are often required to work at the
O '_O 3 O 40
lO
Urlit speed. Nu

Fig. 7 EfEciencies of PDT and CT at different fiow rates

conditions far from the designing point. Therefore, it is
important to examine the behavior of the turbines under
the various conditions.
The test on the Turg'o impulse turbine was conducted
for constant flow rate and supply head up to 7 L/min and
25 m respectively. The maximum head and fiow rate for
the PDT test were 70 m and 14L/min, respectively. The
revolution speeds of the Turgo impulse turbine and PDT
are in the range of O - 10 OOO min 1 and 0-2 800 min l
respectively.
is about 5 within the test range. There are two reasons why
this small Turg'o impulse turbine has such low efiiciency:
first, is the size. Normally the efficiency of the larger size
impulse turbines is about 90%, but it becomes much lower
in the smaller size. Second, unlike the usual impulse tur-
bines, the runner of tbe small impulse turbine rotates fully
immersed in water, therefore the efficiency drops drasti-
cally.

Due to the pressure pulsations in the positive dis-
placement turbine, the effective head is calculated by the
time average of the inlet and outlet pressure.
4. Experimental Results (Turbine Performances)
4. I Positive displacement turbine versus Turgo
ilnpulse turbine
Figures 6 and 7 show the comparison of performance
characteristics of the Turgo impulse turbine with the new
turbine. It is clear that the new turbine has much better ef-
ficiency and much higher output power than the Turgo im-
pulse turbine when they are operated under the same con-
ditions. The maximum efficiency of the Tur*"o impulse tur-
bine is 25% when it is operated at a fiow rate and head in
the range of 4 - 5 L/min and 8 - 12 m, respectively. More-
over, the specific speed is 9.5. In contrast, PDT has the
maximum ef ciency of about 62% and the specific speed
In addition, the efficiency of the PDT increases when
it is operated at higher fiow rates. The optimum efiicien-
cies of the PDT and Turgo turbines appear at different unit
speeds. This is primarily because the revolution speed
of the Tur_ o turbine is much higher than PDT (about six
times at the same fiow rate).
The results also show that the highest efficiency of
the PDT is obtained when it operates at a fiow rate of
1 2 L/min, which is 2 times higher than average fiow rate
of water faucets. Because the fiow rate is proportional to
the displacement volume of the rotors, therefore, the new
turbine size should be smaller of current size for use at
the water faucets, and/or be bigger size for using in other
pipin*a systems which have higher fiow rate range.
4. 2 Performance improvement by modification of
the turbine side clearances
The new turbine is desi_ :ned to have both side and

JS VIE International Journal Serie:'- B. V ol_ 49, No. 2, 2006

48 6

l0
['8[ lO lOO
Side cleara ce of60p m Side ciearaR c oflOll t"

17
- if. 0 41!min -o - ,7. 0 _ 4lhnin
3 : ,.,_,d ,tL -"-_"''O" ' d- ,7. 2 61!min -A- if' 0 '_ 61lmia
:lF::; e:' a:1'1Ll" " 'PO-"
8 i a - if. 8Vmin 7. :
EI :-8V'njn
4 1"'Id JL-"!1: !:';:ilO

%
-h i7. O*lOl,min -1r - '7. _-lOtimin 17 ( c=10// m)
., ,.,.., 80 H - p. i2tlmin -<'- 7. Q 12iiTnin
eO
I O OO 9L HI if. l41/min
/. 1 ::.. -X- '7. Q 14Vmiu

ee
c:
., sl. AA'a'L'
r F
JL
:e 0'6L :
1'O"".
.
. I
Ot ' l V
6 ',iA, iJb"t

A. _ c

i
I

60 8 iA
C,,
,

?
1!' s5
cf "bt . .
(, Oa X
oeo

:'
A' l
:: - "0'1' ia
s

t
:'
.5
Q

e:_ e 4
le . . JO"'

,
:::
:: 0'4

tol ' I t'b

I i; ".'d"'

t
I't s) 40
v
i77( C=6 m) t

!:

i
)!Cl I
JO
Ql
2

n
0.2 -c-Q-41'trin
- dF - Q--6rhlir
- d - 0+8Vtrirt 20

t It t,
oo

-1b-O'iOy ll
o

*} - Q* 12 R
t

0.1 0.3 0.4 0.2 0.5 0.6

I
l
o
500 1500 2000 2 OO
Differential pressure, Ap [Mpa] l OOO 3000
Revolution speed, n Imin t]
(a) Side clearance of 60 um
Fig. 9 Variation of eficiency with revolution speed
l

lO
e :: 't･' el'
IELtda',A'AO:':a:"'O"e :-t 1'1 IF'
oQ.t,t ' 'epJ"
':iA"JL' a" _ _ dt ' a 0.8
8

oo -e. .･･････,sr"_
_ .

n
0.8 '. _,,jt?tO:_1 t .1.
1 : ::'J LCF ' - 'Jl" " ' Sidecieara!'ce Qf 01lm
.. . OR t'r . . V 1'
:p c-- :'r_- ie lF_:
i a,1,'_ _ _ ･--;;
< IL'-"-- -F. Q [!inir
dF IJ": J9' t. ' -- _ : -e-p. Q liuia

S
d":' .J'- "". "t-....._ c rlf. QnjViuh
5 s

1
, ';'f cr 'r ' $. 4f . _A _ lr -Lp. Q*lOVinh
. 06 6 ' /' l'l." 17 t _ 706 i.-- l - -if. Q-i2P:,1'i,
s'
t4!Cf I "- nlF . L
*1 1 dh ,r. O*Wlftnin
S
-*
tl

s "I ' j r 4 yinh

- A
l

: A T. 6uhl ]
)

,1; s } rQ 8L'h'h
./
?

:'4Ft'h.
:5

4 ,,,t O O :_ l' n ': Q IOVolti

0.4 al o /
e,*

(
h
$5 O*6t'it'h T. Q t2Vt,I l
"til
Il
A

_ .. O r Writlh
,,

ll
8 't,h
d

o:
1' : 0.4
dlt A lOVblir

esf ea A,'A
,,, t l, Side cl earance of IQP m
O 12}'iut ,:;
2

,,tt ,
IAQ
l

,t,1 , - o -J7.
t 4}'nt, $::
l

0.2 /
l

- L-F' Q lm!"
fyL L 'pAqle IP"
// - '-p. Q 8 h'h
s' - -!7. Q:'1QVmh
-C-Jr. Qsl2VhliR
//
o

:. 0.2 Tth= p Vd I Q4:'}4ml]V!Tlaa

iS

/,
O 0.2 0.4 0.6 Dif crential pressure, p tMpa] T
ee f.l
1:!= vy8L"w]PP
1 u'l
o r. Q 10Vl P
oo

A r. Q t2pltil l
QI r. O=14Vtah
(b) Side clearance of 10 t-m
l o 20 30 Effective
40 )ohead,60H Cmj
70 80 90 1 oo
Fig. 8 Variation of eficiency and leakage with effective head
Fig. lO Variation of e ciency and torque with effective head

tip clearances of 60 um. The tip and side clearances are

clearances between the casing wall and the lobe rotors and was decreased from 60 L, tm to I O um to improve the efi -
the clearance between the sides of the rotors and the side ciency of the new turbine.
plates, respectively. The comparison of performance characteristics of
The variation of efficiencies and leakage throu_a,*h the PDT between before and after modifications of the side
clearances of two cases side clearances with efflective head clearances is shown in Figs. 8 - lO. From these results,
are shown in Fig. 8. The amount of leakage (Ql) was es- it is found that the leakage is significantly reduced (about
timated by Ql = Q - Qth, where Q is the discharge volume 2.5 times) and the maximum overall efficiencies have been
measuring at the outlet of the turbine and Qth is the the- improved by 10 to 12 percent depending on the operat-
oretical displacement volume delivered by rotors per unit ing conditions. Figure 10 shows that PDT, with the side
time (Qth = Vd'a)). From Fig. 8 (a), we can see that the high clearance of 10L, tm, operates with the higahest e :ciency
efficiency (n : 95% of nmax) operatinga range of PDT (of of about 72% at the fiow rate of 12L/min and 37m of
60 tm side clearance) is narrow and the efficiency rapidly head. Moreover, unlike the PDT with a big side clearance,
decreases with increasing of head. This is because of the the PDT with smaller side clearance can work under the
amount of water leakage through the clearances between high efiiciency (n ; 95% of nmax) over much wider range
the rotatinga and stationary parts of turbine. In positive dis- of operating conditions. For example, at the fiow rate of
placement turbine, the revolution speed is proportional to l 2 L/min, the PDT with side clearance of 60 L, Lm can oper-
Qth' Figure 8 (a) shows that, increasing of the effective ate with high efficiency in the range of 10 to 25 m of head.
head results more leakage (and reduce Qth, at the same In contrast, at the same flow rate, the PDT with 10 um can
discharge Q). For example, at 6 L/min constant discharge operate under higher efficiency in a much wider range, be-
Q, when turbine operated at 40 m of head about 2/3 of tween 20 to 70 m.
the discharge water comes from Ql' This directly reduces In the positive displacement machine, a theoretical
the revolution speed of the turbine and consequent_ly af- torque is a function of differential pressure (p) and vol-
fects the volumetric efficiency (nv = Qth /Q) and overall umetric displacement(8) (Vd = Qthla,), which can be com-
efficiency ( = t * n.) of the turbine. The side clearance puted by Tth = PVd. The torque losses (Tl) is composed

Series B, ¥*ol. 49, No. 2, 2006 JS.'}fE International Journal

 48 7

of two major torque losses in PDT: first, the torque pro- torque (head) . In contrast, relatively small changes of the
portional to the turbine speed, because torque is required revolution speeds are recognized in the small side clear-
to shear the fiuid in the small tip clearance in relative mo- ance case. At the same torques, the revolution speed of
tion and second, torque required to overcome all the fric- PDT with a small side clearance is much higher than PDT
tion such as seal friction, side disks, bearings, and so on. wrth a large slde clearance. Figure 12 reveals that PDT
The actual torque (Ta) delivered to the load is computed with a smaller side clearance can obtain much higher out-
by Ta = Tth - T! = PVd - Tl' The torque efi ciency is cal- put power and consequently higher efficiency than PDT
culated by qt = T*lT!h' From Fig. 8 (a) and (b) the torque with a lar,_',_e side clearance at the same head and flow rate,
efiioiency of small side clearance is decreased noticeably particularly when turbine operates at high head.
in the range of low pressure region. This may be caused 4.3 Pressure pulsations
by the increasing of the viscous torque of the side ends of In the PDT, the intermittent fiow of the fiuid through
the rotors. However, the improvement of the volumetric the inlet and outlet of the turbine generates pressure pul-
e ciency of the turbine with small side clearance is much sations. Pressure pulsations have two important character-
higher than the torque efficiency reduction. Thus the over- istics: frequency and amplitude. The comparison of fre-
all efficiency is increased significantly, see Figs. 9 and 10. quency and amplitude of pressure pulsations are shown in
As the torque losses are small and change a little, the tur- Fig. 13. The results measured at 60 mm far from the in-
bine's torque is only dependent on the pressure, is shown _let and outlet ports of the turbine. The pressure pulsations
in Fig. I O that the torque of PDT is directly proportional measured at 200 mm, upstream and downstream sides of
to the pressure (effective head) and is independent of the the turbine, they did not show any significant differences
fiow rate. in value.
The measured torques and mechanical output power Figure 13(a) shows the amplitude of the pressure
in the large side clearance case and small side clearance
case are compared in Figs. 1 1 and 12. We can see from 0.2
Fig. I I that, the revolution speeds of the PDT with big･ger Turbine Qpe!ation at aveTage pressure ot

side clearance is reduced significantly with the increase of 0.04MPa 0.40MPa

･ - hrlet pressure pulsation - Inlet pressure pulsatbn
Outlet pressure pulsation utlet pressure pulsation
O.

- - - -
l

- ::$
:

Side clearaoce of60p m Side clearancc of lep m ,

0.6 - h4Vmin - - Q Vmin
05

* Q tanin
- 8Vmin
- -Q Vmiu
-Gh - Q lVmin
-
O

-r ! OL'min - . * Q l Oumin
Q*12V!ni!t -- Q l 2Vmi't
F Q l 41/min -OF- Q l4Vmin

' s_ 0.4
e (P
S

< -O. 1
e

e A
.

E
It,

'
::,

e
･*

A
e

e
0.2
(
Q
*

A Q1 -0.20
0.02 0.04 0.08 0.06 O.
l

tD
e AA
Time [s]
O

(a) Constant speed, n = 800 min l

o

] !
o

500 1000 1 500 2000 _9500

Revoh!ion speedj n [nin l]
3000
0.6
ni2 100lmin 1] =1 OOO[min l]
Fig. I l Torque and revolution speed relationship
* - Iniet
Outlet e- Inlet
--- - - utlet
l ･_O 0.4
Stde clearance of 60 p m Side clearance Qf 10 p m
:

- P* Q* ,Vrnin --1 -' P.. Q=4Vmin sp c

100 -p* Q Vrnin ...e- pl" ( 6Vmin
- - p Q 8Vo'in -
- p* Q 81!min Q
:

- Pp Q*lOVmiu --( -p*Q IOl/min 4L

S

hi2L'!T'in '- -P.. l21/min L

80 F P j 14ymtl ' ::h- P,,. Q l4Vmin
c

c': 0.2
,

Q d .L .
Q
c
:)

-:'cL 60
4

e' t
:s
o
'

O
'

l I
(

l ,

_1,1t -//,N,b
ILIItI_･_l
ll
/lL

g 40 iG q 'L
tt

,l
l , I
(?

,It
tI ,1
J
It
dr

,
,

J: 1 A : A o , )
,

t l

* ... ]
,

oo H * .
Q

･_O
O O 0.02 0.04 0.06
E3

0.08
3
oo

'

Time [s]

500l ooo
2000' 3000
1 500 500
Revolution speed, n [min :] (b) Constant pressure (head), p = 0.2MPa
Fig. 12 Output power and revolution speed relationship Fig. 13 Pressure pulsations

JS"' :E Ir ternational Journal Series B, ¥/ ol. 49, i¥To. 2, 2006

488
pulsations of the turbine operating at different heads but
same speed. Figure 1 3 (b) shows the pressure pulsations
when the turbine works at the same head but different
speeds. The frequency of the pressure pulsation is equal
to f = n * (LON)/30. From Fig. 13(a), we can see that
when the turbine works at higher head, the maganitude of
pressure pulsations of both inlet and outlet of the turbine
increase. From Fig. 1 3 (b), the magnitudes of the pres-
sure pulsations also increase when the turbine operates at
higher speeds .
These pulsations are sometimes very severe and can
cause damage to the piping or other components in a hy-
draulic system. To eliminate and/or minimize pressure
pulsations can be achieved by judicious use of acoustic fil-
ters and side branch accumulators(9) or modification of the
rotor shapes . In this study, small bladder pressure tanks
were used. The pressure tank is 20 mm in diameter and
50 mm in height and charged by air (see Fig. 14(a)). A
rubber sheet was used as the bladder in order to prevent
the contact between fiuid and charged gas. Figure 14(b)
shows the coinparison of the pressure pulsations between
turbines operating with and without dampers. The results
in Fig. 1 4 show that by installing only one damper close to
oaO
(a) Side-branch accumulator
O.
6
Both inlet and outlet have dampers
Inlet Pressure
0.4
-
'S
l References
:
L-,
ca
O 0.2
(e
ea Without dar!1per wjth one damper
e)
e at Inlet
O
Tune [s]
constant speed, n=800tmin'l J
(b) Reduction of pressure pulsations
Fig. 14 Reduction of pressure pulsations by dampers
Series B, ¥･'ol. 49, No. 2, 2006
the inlet port of the turbine, it makes the inlet pressure be-
come nearly stable. However, the pulsation of outlet pres-
sure is increased. Therefore, only 1/2 of the differential
pressure pulsation can be reduced. In contrast, when two
dampers were installed at both the inlet and outlet of the
turbine more than 3/4 of the pressure pulsations were elim-
inated. Consequently, vibration and noise are reduced, re-
sulting smoother operation.
5. Conclusions
From the results of this study the major conclusions
are summarized as follows:
( I ) A new positive displacement turbine was devel-
oped. The ef :ciency of new t rbine is much higher than
the small Turgo impulse turbine (about 2.9 times).
( 2 ) The new turbine, with a smaller side clearance,
has much higher efficiency than the turbine with bigger
side clearance and it can sustain the high efficiency under
the wider range of operation conditions.
( 3 ) The maximum efficiency of the new turbine is
72% when it operates at 12L/min and 37 m of fiow rate
and head, respectively. To fit for the normal size of water
faucets, the PDT size should be smaller than current size.
( 4 ) The pressure pulsations of this turbine can
be eliminated about 80% by using simple construction
dampers .
( 5 ) The turbine torque of the PDT is proportional to
the effective head and independent of the fiow rate. The
leakage is also depended on the effective head and nearly
independent of the fiow rate.
It is quite clear from the study that the positive dis-
placement turbine is suitable for the micro-hydropower
generation which requires very low specific speed much
lower than impulse turbine specific speed. The new tur-
bine is also applicable for extracting micro-hydropower at
the valves of other piping systems .
Acknowledgement
The authors would like to express their sincere grat-
itude to TATSUNO Mechatronics Co., LTD. for proving
the experiment apparatus. Grateful thanks to Dr. Kazunari
Matsumoto for his continuous help during' research study.
( I ) European Renewable Ener*"y Exchange. Small Hy-
dropower Fact Sheet, http://www,earorex.com
Outlet Pressure ( 2 ) Davis, S.. Microhydro: Clean Power from Water,
('_003), New Society Publishers, Canada.
i ,,
( 3 ) Water Recycle Group, Electricity from Water with
PowerPal, Micro and Mini Hydroelectric Generators,
Canberra, Autralia.
( 4 ) Thake, J., The Micro-Hydro Pelton Turbine Manual,
(2000). ITDG Publishing, London.
( 5 ) Harvey. A. and Brown, A., Priyantha Hettiarachi and
Allen Inversin, Micro-Hydro Design Manual, ('_005),
ITDG Publishing.
JS, lE Irbternational Journal

（6） Bama，PS。，FI纏idMechanics forEng三neerings，（1964），
   ButterWo曲s，London．
（7） Nag，PK。，Power Plant Eng1neering，（2002），Mc−
   GrawHiH，Singapore．
（8）Merritt，H』E・，HYDRAULIC CONTROL SYSTEMS，
1S躍E1π‘emα琵oπα」／0琶rπαZ
489
   （1967），JohnW重1ey＆Sons，Ipc．，NewYork．
（9） Karasslk，U，，Messina，J．P，goopeちR and Heald，
   C。C．，PUMP HANDBOOK，（2001），McGraw−Hi11，
   New Ybrk．
Seripq R V湖 4．q．Nn ウ つno6