Hou Peijin,Lu Qiling,Zhang Yungang and Zhao Kemei
Shanxi Provincial Institute of Hydraulic Research, China
No.112, 4th Lane, Taoyuan Road,Taiyuan,China.
Tel:+86-
351- 4040043,Fax:+86- 351- 4047196,E-mail: wam@public.ty.sx.cn.
Abstract:
Based on hydraulic model test, the process and the
pattern of sediment deposit in high pressure water conveyance pipe, the
YRDP(Yellow River Diversion Project) Shanxi, China, were studied, the
non-deposit critical water velocity and the blockage problem in the pipe were
analysed. The test result shows that the sediment deposit will appear in the
high pressure water conveyance pipe, if the water velocity becomes small, even
if the sediment concentration is small and the sediment grain-size is fine.
However, the thicker the sediment deposit, the higher the water velocity in the
pipe, therefore provided that the system works in good order, the pipe will not
be blocked up. The non-deposit critical water velocity being 0.9~1.0m/s.
Keywords: high pressure water conveyance pipe, sediment deposit, deposit and blockage in pipe, non-deposit critical water velocity
In the water conveyance engineering, we must think over the sediment deposit problem especially in the water conveyance pipe engineering. As the deposited sediment is hard to dislodge in the water conveyance pipe, once it was blocked, the whole water conveyance system would come to a standstill, so when water contains silt in it, we should consider it sufficiently and should put forward some effective measures to treat it in design. The Yellow River Division Project (YRDP) is a large hydraulic engineering project. Its conveyance discharge is 48m3/s. Its total conveyance length is 452km. It is of the multi-stage pumping with a very high lift of pump. the geologic condition being very complicated. Although it takes water from Wanjia-Zhai Reservoir, there continues to be a certain amount of sediment in flood season. According to the calculation, the sediment concentration can attain 4.08kg/cm3 in August, the sediment grain size is very fine, about 70 percent of it，the grain size is smaller than 0.025mm. As only one or two pumps are to be put into operation at the beginning period and the corresponding water velocity in the general pipe is as low as 0.3~0.6m/s, so whether there exist the sediment deposit problem in the pipe has been becoming the focus of interest in the YRDP. Therefore this hydraulic model test was made.
A 1:20 undistorted model was built. The model contains the general pipe, the main pipe, and the third-stage pumping station outlet (see Fig.1). The model pipe is made of transparent organic glass.Because of the Sediment deposit is the main problem we study, the model sand used must satisfy the condition of the settling similarity. From Stokes's formula, the scale of the model sand's grain size can be obtained as follows:
Fig. 1 The sketch of the model
Fig. 2 Comparison between model and prototype grain-size grading
Analysing the sample taken from various sections, along the pipe, we can see that: (1) whatever the operating condition is, at the beginning of the pump running, the distribution of the sediment concentration along the vertical line in the pipe's cross section is fairly uniform ,The grain size of the sediment along the vertical line is a little larger near the bottom than that near the top.(2)Along with the elongation of the running time, the sediment deposit will occur at the bottom of the pipe, where the sediment concentration will obviously heighten, however, the sediment concentration along the vertical line still remains uniformity above the surface of the deposited sediment, only with a gradient near to the surface.
Along
the pipe, the variation of the sediment deposit in the general pipe(φ=5200mm)is shown in Fig.3.The accumulation curve of the sediment
deposited amount is shown in Fig.4.
From these figures we can see:
(1) The change of the sediment deposit, along
the pipe, is biggish in the general pipe with a visible rising and falling in
deposit surface. At the junction between the main pipe and the
general
pipe, only a little deposit with an unsteady deposit surface can be found as
there seems to be a phenomenon of the spiral flow with a strong turbulent
motion, whereas there are a lot of deposits at the bend where the horizontal
pipe is turned upward to the vertical direction.
(2) At the beginning, the deposit velocity is very high, along with increasing of the operating time, the deposit velocity becomes low bit by bit. At the end, it goes to the silt -stable state. The different running condition produces the different deposit process. Under the condition that the sediment concentration is the same, the deposit amount in unit-time is in inverse proportion to the discharge, where as the deposit amount increases with the increasing of the sediment concentration, with the discharge being constant.
(3) From the accumulation curve of the deposit
amount (Fig.4), it can be seen that: the types of the accumulation curve of the
deposit amount in pipe are sameness basically under different running condition.
But at the time when the sediment concentration is the same, the deposit amount
decreases as the discharge increases this indicated that the selection of the
running condition has a great influence on preventing and on decreasing of the
sediment deposit in the pipe.
Fig. 3 Variatcion of the sediment deposition along the general pipe.
Fig. 4 Accumulated deposition in general pipe
From the data of the experiment, it can be seen
that there is almost not deposit in the main pipe(φ=3800mm) when two
pumps (Q=12.9m3/s)work, with the average water velocity of 0.91m/s.The
relation between the deposit amount and the average water velocity in the pipe
under different running condition is shown in Fig.5, from this figure, it can be
seen that when the deposit amount approaches zero, the average water velocity in
the pipe is 0.9~1.0m/s.
At present, there is a few result on the researching of the sediment which has a low sediment concentration and a fine grain-size in high pressure water conveyance pipe. In the paper of 《the primary discussion on the characteristic of pipe carrying clay slurry》[1], presented by our Institute, the data concerning the test are as follows: pipe diameter d=100mm,clay slurry median grain-size d50=0.009~0.013mm, and operation head H=7.05~10.01mm.When the slurry concentration cv=8%,the measured non-deposit critical velocity Uc=1.04m/s. The other experiment data on hydraulic suction pipe obtained from the Tian-jia Wan Reservoir[2] are: pipe diameter d=254mm, pipe length L=48~63m, median grain size of the clay slurry d50=0.006~0.008mm,when the slurry concentration cv=5~7%, the non-deposit critical water velocity Uc=2.3m/s. Those experiments show that the non-deposit critical water velocity of the fine grain sediment is very complicated.In addition the following two formulae are taken to calculate, the calculation results show that the non-deposit critical water velocity calculated is in approximate agreement with that measured.
(1) KHopoЗ's formula[3]
This formula is adapted for the sediment grain-size d<0.07mm. For the pipe, it is expressed as
Where, Uc is the non-deposit critical water velocity (m/s). P=Cw×100,Cw is the solid concentration of the weight ratio. Cw=0.0038.βis the coefficient.β=(gs–1)/17=0.97.D is the pipe diameter D=5.2m.By computation Uc=0.97m/s.
Fig. 5 Relation between the deposition and the water velocity
(2) Fei Xiang-jun's simplified formula about non-uniform flow[4]:
where
,
fm=0.01035(coefficient of drag, from Moady's diagram)
ωmove is the settling velocity of sediment in moving water. ωmove =ω0(1-Sv)7
ωo is the settling velocity of sediment in still water. ω0=(Υs-Υ)d2/(1800μ)
Υm=1.0025(unit weight of muddy water).D=5.2m(pipe diameter).Uc is the non-deposit critical water velocity(m/s).When the temperature t=23℃，the computed result is shown as follows:
|
d(mm) |
0.03 |
0.025 |
0.02 |
0.01 |
|
Uc(m/s) |
0.92 |
0.81 |
0.70 |
0.44 |
From the above table, it can be seen that the result computed closely agrees with that measured, especially with that by KHopo3's formula.
Based on the deposit data in pipe, either main pipe or general pipe, the maximum thickness of the deposit by meter is listed as follows:
|
Discharge(m3/s) Diameter(mm) |
6.45 |
12.9 |
19.35 |
|
Φ3800(main pipe) |
1.7 |
0 |
0 |
|
Φ5200(general pipe) |
3.41 |
2.16 |
0 |
This table shows that, the deposited thickness in general pipe will be over one half of the pipe, if only one unit runs for a long time, however, when three units run, there is no deposit.
From
deposited pattern, it can be shown that the thickness of deposit at the bend
where the horizontal pipe turns to vertical direction are slightly bigger than
other places, while the maximum thickness of the deposit at the pipe's bend, is
only about 1.0mm, as the fine grain sediment's rest angle under water, according
to related information, is 4.5-5°.
In addition, there is a centrifugal force acting on the outer surface at the
bend of the pipe, therefore, the pipe will not be deposited and blocked up, as
long as the running is kept in good order. From the bulk density of the deposit,
it can be found that the bulk density of the deposit is so low that it is
initiated easily in the early days of the deposit. During the period of running
the bulk density of the deposit in the pipe change from 10.7KN/m3 to
12KN/m3 or 13KN/m3 at certain portion, showing that the
increasing of the sedimentation's bulk density is very slow, when the deposit
surface is not exposed.
In high pressure pipe, the deposit having fine grain and lower sediment concentration are a new problem we met with in the practice. From the experimental result, it can be seen that its deposit process and deposit pattern are similar to the sedimentation in the settling basin .As the water velocity is small and the sediment concentration is low at the time when one pump or two pumps work.. We had a try for computing the deposit amount in the pipe at every the interval by the following formula used for settling basin and made a comparison with the measured results.Adopting formula: hi/ωmove = L/V
Where hi—the deposit height of sediment of a certain grain-size group in time interval of L/V
L—the pipe length computed. V—the average water velocity in pipe (Q/Ai).
ωmove—the settling velocity of the sediment of a certain grain size group in moving water
A: Using formula from Yellow River Institute for computing ωmove .
Supposing:
It follows that :
ω0—settling velocity of the sediment in still water.
B: Using formula from
Sediment Hand Book to find ωmove
Where, L-the space between two neighboring grains
η=1.14(coefficient). It follows that :
The settling velocity in moving water is shown as follows:
|
Diameter (mm) |
d>0.03 |
0.02<d<0.03 |
0.02<d<0.03 |
d<0.01 |
|
Yellow River Institute (cm/s) |
>0.105 |
0.105-0.046 |
0.046-0.012 |
<0.012 |
|
Sediment Hand Book (cm/s) |
>0.088 |
0.088-0.039 |
0.039-0.01 |
<0.01 |
When Q=6.45m3/s,ρ=4kg/m3,the accumulation curves computed and measured about deposit amount at each time interval are shown respectively in Fig .6.From this figure, we can see that:
(1) The deposit amount at each time interval computed and measured is the same basically except at the initial running stage, proving that the regime of the pipe is almost the same as that for the open channel, when the water velocity is small in the pipe.
(2) At the early time of the running, the deposit amount measured is higher than that computed, whereas at the later days, the value computed is higher than that measured. This phenomenon maybe due to the fact that the boundary condition of the pipe is not consistent with that of the open channel.
Fig. 6 Comparison of accumulated deposition between measurement and computation
In high pressure pipe, there will be the deposit in the pipe, if the water velocity is lower than the non-deposit critical velocity, even if the sediment concentration is small, However, if a correct running way is adopted the pipe will not be blocked up. Under the condition that there is no drainage, the bulk density of the deposit of the sediment having a fine grain size is low, the deposit may be swept away once the water velocity become larger than the starting velocity of the sediment. But the sediment concentration in the water will heighten suddenly, causing an unfavorable influence to the set's running. Therefore, in choice of the running pattern of the high pressure pipe, we should consider the sediment's characteristic and the pipe's conditions, and should try our best to make the water velocity in pipe be greater than the non-deposit critical velocity of sediment, in order to guarantee that the deposit of sediment doesn't occur during the process of the water conveyance.
References
[1] Dai Jilan, Wan Zhaohui, etc. An Experimental Study of Slurry Transport in pipes, First International Symposium on River sedimentation, Beijing, China.1980.
[2] The Experimental Study on Hydraulic Suction Pipe in Tian-jia Wan Reservoir, Journal of Shanxi Hydrotechnics, China.No.1,1977.
[3] Fei, Xiang-jun(1994), The Clay Slurry and Grain Material Transport Hydraulics, Tsinghua University Press, Beijing, China. PP 253~254.
[4] Fei, Xiang-jun(1994), The Clay Slurry and Grain Material Transport Hydraulics, Tsinghua University Press, Beijing, China. PP 262~264.