1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Flow rate of water dropping out of a pipe

  1. Jan 6, 2013 #1
    I need to figure out the best way to improve drainage of spring water flowing through my property.

    The water from my neighbour’s spring flows 10m through a pipe into a manhole on my side. The water level in this manhole is always 10-20 cm higher than the in and out pipes. The water then flows 30m to a second manhole on my side and then through a 10m pipe into the lake. When I replaced this last 10m pipe, I set it lower so that the water from the 30m pipe dropped 20 cm instead of 5 cm. This appeared to increase the flow considerably, but I’m not satisfied because the water level in the first manhole is still too high, even though the water falling out of the 30m pipe is only a third of its diameter.

    My questions:
    1) When water drops out of a pipe, is the flow rate affected by how big the drop is?
    2) If the pipe has an increasing gradient, does the faster water increase the speed of the following water by pulling on it or by reducing the flow resistance?
    3) How does the flow work, do the water molecules stick together or are they independent like marbles and lose energy by jostling each other?
    4) If I increase the diameter of the pipe only in the later half of its length, will this increase the flow?
    5) If I replace the concrete pipe with a lower friction pvc pipe only in the later half of its length, will that increase the flow?

    As you see, I am looking for a way to improve the situation without replacing the whole 30m of pipe.
     
  2. jcsd
  3. Jan 6, 2013 #2
    flow rate has a simple formula which is flow rate=velocity of fluid X cross sectional area of pipe.
    there is a term chock flow..which is the max flow that can take place throw a pipe or nozzle at specific inlet pressure..so increasing the diameter will not give u increased flow unless choked flow has occurred.
    replacing the concrete with pvc pipe will decrease the pressure drop occurring in pipe,which will decrease the head loss..but it is irrelevent,because i think u are not going to extract energy from this drain water..
     
  4. Jan 6, 2013 #3
    Please draw a sketch showing the cross section of the 'manholes'.

    I suspect that the first 'manhole' is what is known as a silt pit which needs cleaning out.
     
  5. Jan 6, 2013 #4

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    1. No. The flow rate is governed by the amount of gravity head at the exit. Once the water leaves the pipe, it has no effect on the stream still in the pipe.
    2. By increasing the gradient of the outlet pipe, you are moving the exit to a lower position than originally, thus increasing the gravity head. This would account for the increased flow rate.
    3. Liquid flow resistance depends on the viscosity of the fluid, or the resistance to shear internal to the fluid. This is a complex subject with no simple one-line answer.
    4. Increasing the diameter of the outflow pipe partially might produce a small change in outflow rate, due to having a smaller length of original pipe, but it is hard to say without calculation.
    5. Ditto with changing material for the last part of the length of pipe.
     
  6. Jan 6, 2013 #5

    Low-Q

    User Avatar
    Gold Member

    Sounds like you have some kind of siphon.

    1. The increased drop will in general increase the flow rate if the drop is still inside the pipe. However, the flow rate is not increasing at the same scale as the increased drop. Higher drop will in general pull harder on the water in the manhole due to increased force from the weight of the water - even if the weight is the same, the angle is increased.

    2. Not sure what you mean by increasing gradient. Do you mean that the pipes cross section is increasing with length?

    3. It is known that troubled water is heating up. Rivers tend to not freeze to ice due to the turbulence in the water which force the molecules jostling each other. It means mechanical loss.

    4. Increasing the cross section after the half length will in general not increase the flow. However, the friction/mass relationship is less in a bigger pipe.
    Imagine if you put a bunch of 30 m drinking straws inside a big pipe. The cross section is the same, but the friction has increased reducing the flow rate. Remove the straws and the flow rate increase. So in practice: Yes, the flow rate will increase.

    5. I would suggest that a rough surface might have more friction than a smooth surface. However, if you let air bubbles enter the pipe from small grooves inside, the flow rate will increase considerably.
    Also, my own experiment with smooth surfaces will actually increase friction compared to a more coarse surface. But that is because the coarse surface allow air to be trapped inbetween the surface and the water.
     
  7. Jan 7, 2013 #6
    Thanks everybody for your comments. I would reply as follows:

    To Studiot: Yes, the 'manholes' are silt pits and I keep the bottoms clean, so this is not a factor.

    To Low-Q:
    I don't see any syphon effect. There is just the gravity effect, as each of the 3 pipes has 1 end open.
    By 'gradient' I meant fall or drop along the length of the pipe.

    From your point 1. I conclude that I can just replace the last section of the 30m pipe and increase the fall of the new section. This will reduce the head in the first manhole and increase the flow rate. But when the water leaves the new section into the second manhole, there will be no penalty from a shorter drop.
     
  8. Jan 7, 2013 #7
    Well, I only commented about the manholes so far and it seems I guessed correctly.

    A long section would greatly facilitate a proper assessment of the setup. You have not mentioned pipe diameters or if the final outfall is above or below the lake surface.
     
  9. Jan 7, 2013 #8
    The hydraulics of pipes running full bore and part bore are quite different.
    This is often not realised even amongst 'experts' who then implement an inappropriate channeling and then wonder why it does not work.
    I have seen this in practice many times.
     
  10. Jan 7, 2013 #9
    I made a nice diagram of the setup, but it doesn’t copy, so I am giving it to you as text.

    Source (1 metre below surface) --> Pipe P1 --> Silt Pit S1 --> Pipe P2 --> Silt Pit S2 --> Pipe P3 --> Lake


    Pipe length, diameter, material and capacity usage at each end of pipe are:

    P1 = 15m x 15cm concrete, 100% at source, 100% at S1
    P2 = 30m x 10cm concrete, 100% at S1, 33% at S2
    P3 = 10m x 10cm pvc, 33% at S2, 33% at lake (outfall above surface)

    Estimated flow at source = 1.5 litre/sec (1 litre into P1 + 0.5 litre overflow)
    Estimated flow into lake = 1 litre/sec

    I assume that by replacing P2 the problem would be solved and the flow would then be 1.5 litres/sec.
    But as P2 is 30m long and runs under the lawn and I do not want to use a digger (very messy!) I plan to replace only the end part of P2 with a 5m x 10cm pvc pipe, increasing the fall at S2 by 20 cm.
    Do you agree that this would increase the flow and eliminate the 20 cm head of water in S1?
     
  11. Jan 8, 2013 #10
    No I'm afraid that your proposal will not have the desired effect.

    The only way to force more water through a pipe at full bore is to increase the pressure at the inlet. This is why water is building up above the inlet to P2. So long as any section of P2 remains at 100 diameter you will have this problem.

    There are things I can suggest however, but I have a full day's wok ahead of me today so I will not be able to get back here until this evening.

    :smile:
     
  12. Jan 10, 2013 #11
    Or shorten the pipe. If I introduced an additional manhole with silt trap partway along P2, then I would shorten the distance which water has to flow and thereby reduce the resistance due to friction. It would also help the cleaning of the pipe. Then I would expect a greater total flow and elimination of the head in S1, without causing another head in my new manhole.

    The only problem is the installation of the manhole. A concrete manhole, which is the only type used around here, can be installed in sections, but the heavy bottom section has to be lowered into position and it has to be big enough to give access for making the watertight connections. If you agree that this solution should solve the problem, I will enquire about the manhole sizes and weights. If the manholes are too heavy, then I could pour my own manhole in situ, but this is a lot of work.

    I have seen that manholes or traps exist in lighter materials, but I don't know if they are stable. My doubt is concerning the watertight inlet and outlet. I could make my own manhole with a pvc rainwater tank, but I doubt that I would get a good and permanent join to the concrete pipe ends. The advantage of the concrete manhole is that it doesn't move and you can easily make the watertight joins with cement.
     
  13. Jan 10, 2013 #12
    You seem over concerned with the length of pipe.

    The plain fact is that the pipes are too small for the job, you have a 6 inch inlet pipe feeding a four inch outlet pipe. Really it should be the other way round. This is apart from the fact that 4 inches is too small for such a duty and even 6 inches is suspect.

    So the logical approach is to find a way to increase the effective pipe diameter.

    Since you do not want to lay a larger one, how about a second one, either above the first or via a different route around the grass?

    Alternatively how about a french drain?

    A problem with any pipe, even if it is theoretically adequate when new, is due to blockages and wall encrustation.

    As a for instance when Colebrook was doing the original research for the famous published tables that everyone uses for design of these things he found that

    The friction factor for new concrete pipe and plastic pipe to be 0.03, but for what he called
    'slimed concrete' it increased to 6, and increase of 200 fold.

    No I'm sory an extra pit will just be a good deal of work but will not solve th basic problem.

    Incidentally pipe runs are designed to develop what is known as 'self cleansing velocity' Their grade is normally set to achieve this and it occurs when the fluid velocity and that of the suspended solids is the same. Tilt the pipe too steeply and the fluid will flow too fast and leave the solid behind. Tilt it too little and the fluid will not have enough energy to transport the sediment, so it will scrape along the bottom and build up into a blockage.
     
  14. Jan 11, 2013 #13
    So far, you have demolished all my short cuts! But of course, I don't want an easier solution which does only half the job.

    I am now considering your idea of a second pipe. Indeed, the pit S1 had an original outlet pipe which lies under the end of the lawn, but this old concrete pipe is full of roots. P2 was laid later to feed a pond.

    So I am now considering to replace the original pipe which runs 20m direct to the lake. It's a lot of digging, but it's more out of the way, so I can do the work in stages.

    I will lay a 15cm pipe and I will put a sliding flap gate in S1 so that I can divert all or some of the water into this new pipe, as required. The normal recommended fall is 1-2%, which would be 20-40 cms in this case.
     
  15. Jan 11, 2013 #14
    I'm not trying to be a wet blanket, if you will excuse the pun, just help you achieve a workable solution.

    You mention digging. It is perfectly possible, and quite cheap these days, to thrust bore pipes of this size. This saves most of the digging. I have successfully inserted similar under/through motorway embankments.

    Secondly when I asked for a long section I also meant to include the falls or gradient information. That is the elevation or height differences betwen the various points. You have already provided estimated flows. The gradients allow formal calculation of required pipe sizes using these. Most pipe manufacturers publish these in their catalogs - or you could ask your pipe supplier.
     
  16. Jan 11, 2013 #15
    Have looked at some videos on thrust boring. Looks good. Not sure what you mean by quite cheap, but a firm supplying a technique like this in Switzerland is highly unlikely to be cheap. Most construction in Switzerland is piecemeal stuff, so there are less benefits of scale. Costs are kept down by good organization and control.

    The point is, as soon as my costs go up, I would logically have to consider renovating my neighbour's spring too. The water is just flowing uncontrolled out of the ground, because the original concrete containment has been smashed to pieces by farm machinery. The farmer is only a tenant, so he won't pay anything and the subsoil is porous, so there is no flooding. I get dead animals flowing through with the water.

    The ground area we are talking about has only a slight fall towards the bank of the lake, then it drops 1-2 metres. So the depth of the trench can be kept constant and this will give a fall of about 2-3%.

    So thank you for all your advice and now I have a pretty clear picture. Don't want to misuse PhysicsForums! Will let you know when I start work.
     
  17. Jan 11, 2013 #16
    I'm in the UK, not Switzerland and we have plenty of firms doing thrust boring. This is not a tunnel throught the Alps, or an underground railway in London!
    My gas supply pipe was recently replaced by this method under my concrete drive and reinforced brickwork planters. Magic. And cheap. One of the utilities contractors did it.

    :biggrin:
     
  18. Jan 12, 2013 #17

    anorlunda

    User Avatar
    Science Advisor
    Gold Member

    Are the pipes filled with water? If the water merely drips from the end, it suggests that much of the pipe volume is actually filled with air. If so, then changing pipe diameter will have little effect.

    If the slope of the pipes is not uniform, it is also possible to trap an air pocket in a high spot of the pipe. that will reduce the effective diameter.
     
  19. Jan 12, 2013 #18
    Have you read post#9?

    Your point about a local hump in the pipe trapping an air bubble is a good one, however.
     
  20. Jan 14, 2013 #19
    After 30m the water flows out of the pipe quite fast, considering that the fall is only 1-2%. But the diameter of the pipe used by the outflowing water is only about 30%. So yes, you are right that a lot of the pipe is filled with air and that's the frustrating part.

    My original question had a lot to do with why the pipe is delivering so little water. It's possible that there is an air pocket, but it's also possible that there are roots in the pipe. I do clean the pipe every year and in doing so I remove a small amount of silt and fine roots from the pipe, but this only increases the flow marginally.

    The conclusion so far is that there is no sense in dividing the pipe into 2, which would facilitate better cleaning and inspection, because that would not increase the flow of the 2 halves. Why not, is not clear to me, but I accept the advice on this point.

    Perhaps you could explain, if the head of water is caused by the dimensions of the pipe, why can't I remove the head by reducing the dimensions? I can see that the diameter of the pipe is a restriction to the flow, but I don't see why the length of the pipe is not relevant. I can't imagine the water not flowing through the pipe if it were only 30cm long instead of 30m, because the friction loss would be so much smaller.

    If I had an open pit 30m long with a 10cm inlet and outlet, are we saying that the outlet would not allow any more water to pass than it does now?
     
  21. Jan 14, 2013 #20
    My advice is to cross a local farmer's palm with silver.

    Since you live in an area of good surface and groundwater many farmers will have a land drainage plough such as this.

    You could get a new P2 pipe in plastic installed easily and cheaply this way, but go for at least 6 inches. A 30 mete run is almost nothing, This pipe comes on 100m+ rolls.

    Much smaller machines are also in use but you get the idea from here.

    http://www.mastenbroek.com/land%20Drainage-agricultural.htm [Broken]
     
    Last edited by a moderator: May 6, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Flow rate of water dropping out of a pipe
  1. Water flow into a pipe (Replies: 2)

Loading...