Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Water Distribution - Rain Barrel System

  1. Jul 7, 2008 #1
    I had an idea to collect rain in large amounts in rain barrels for use in irrigation for my lawn and garden. I want to create a network of 30 rain barrels at 55 gallons each. I will put one rain barrel at each of my down spouts on my gutters, to catch the rain. Then I will have the rest, 28 barrels, 100+ ft away from the catching barrels at the back of my property. I was planning to connect the 30 barrels to a shared 1" backbone pipe that will connect underneath each barrel and run under ground.

    Code (Text):
     b                                                         b  b  b  b  b  b
     |                                                         |  |  |  |  |  |
     -------------------------------------------------------------------------- etc
                                |                              |  |  |  |  |  |
                                |                              b  b  b  b b  b
               HOUSE            |

    My question is, will gravity evenly disperse the water to all the barrels, or will I end up with more water in the catching barrels vs the network in the back, or vice versus? Basically my hope is that each barrel will fill up evenly with the others to maximize water collection. Any thoughts would be greatly helped.
  2. jcsd
  3. Jul 7, 2008 #2


    User Avatar
    Gold Member

    Water always levels out right? Think about this.
  4. Jul 7, 2008 #3


    User Avatar
    Science Advisor
    Gold Member

    I'd like to see a much better diagram if possible.
  5. Jul 7, 2008 #4
    Here is a "better" diagram. The point I'm curious about is the effect that the large distance between the network in the back and the catcher barrels will have. Note: The network in the back will be 28 barrels. It is drawn as 10 to save space on the diagram. The catcher barrels by the house will indeed just be one on each side as shown.

    For this question, I think we can just assume the barrels are all approximately at the same level.

    Attached Files:

    Last edited: Jul 7, 2008
  6. Jul 7, 2008 #5
    The previous diagram had a transparent background and was showing up funny in my browser. Here it is with a white background.

    Attached Files:

  7. Jul 7, 2008 #6


    User Avatar
    Science Advisor
    Gold Member

    Hmm, not quite what I was after. Could you do a drawing showing the vertical levels of all the necessary pipework? What level/gradient is the guttering at? How about the long runs of pipes?
  8. Jul 7, 2008 #7
    I dont have very good software to make diagrams. So I will try to describe it a little better. Assume all of the barrels are at vertical level 0ft. The pipe is 1" pipe that goes directly out the side of the base of the barrel for 3 in then down at a 90 deg angle for 2 ft, where it connects to the backbone pipe. The backbone pipe will be completely at -2 ft. The house is 75 ft wide and 40 ft deep. So the two catcher barrels are 115 linear feet apart on the backbone pipe. The network of barrels in the back are all less than 2 feet from each other making a rectangular grid. The grid is 200 feet from the closest connecting point of the backbone of the two catcher barrels. All the barrels will be hooked to the back bone in the way described before, 3 in out, 2 ft down.

    If it makes a large difference, there is a slight grade up from the house to where the network of barrels will be in the back. Might be 1 deg or less.
  9. Jul 7, 2008 #8


    User Avatar
    Science Advisor
    Gold Member

    If all the barrels are at exactly the same level, as you've described it the water will level itself out across all barrels. However, a 1 degree incline over 200 feet will raise those barrels by 3.5 feet, so (if these barrels are higher than the ones at the house) they'll all be 3.5 feet emptier.
  10. Jul 14, 2008 #9
    I wonder if venting of the pipe may be required, as in house drain/waste pipe. If it;s only 1" diameter, could run slow which may be fine anyways. So long as the collection barrels all have open tops my guess would be that the 1" pipe would be fine unvented, but I don't really know.

    BTW, if I was doing this myself I would look into putting valves at the inlet of each barrel, or a valve for a bank of 3 barrels or so, just so you could control which barrels will fill, leaving the others empty, which will allow less stagnant water and mosquito breeding.

    I would also be concerned that the long line would eventually plug up with silt and debris, as the water would be stagnant and dirty. I would look into a method to provide hooking up a hose to back flush as required.
    Last edited: Jul 14, 2008
  11. Jul 15, 2008 #10
    I would say that the 1-inch pipe is too small. First, you need to decide how heavy a rainfall (in say, inches per hour) that you want your system to handle without the downspout barrels overflowing. Then you need to figure out how much of the roof surface will go to the downspouts, so you can figure out the flowrate (in gallons per minute) through the pipe to the storage barrels. Then you can figure out the pressure drop this flow will create, allowing you to see how much higher the level will be in the downspout barrels, relative to the level in the storage barrels. If the downspout barrel level you calculate is higher than the top of the barrel, that means the barrel will overflow (wasting rainwater you are rying to collect). And that means the pipe is too small.

    If I understand your system, a 1 inch per hour rain will produce almost 30 gpm thru the pipe. If it is one inch pipe, the head would need to be 24 feet for each 100 ft of pipe. You say you have 200 ft of pipe, so the level in the downspout barrels would need to be 48 feet - obviously thats too high, so your one inch line is too small. for a reasonable head (around one foot) use 2 inch or even better, 2 1/2 inch pipe. The pressure loss goes as the fifth power of diameter - a huge effect. But remember - it all depends on the flowrate you want the system to accommodate without overflowing.
  12. Jul 15, 2008 #11
    Thought about doing this myself. You are going to want to put screens on the lines from the house barrels out to the storage barrels. Shouldn't need to vent anything as long as the barrels are all open to air. One problem may be the slope of the land, but other than that all the levels should be equal due to hydraulics. Might be a better idea to use larger piping and a cattle trough for the collection if you expect to get heavy rain. Also you may want to think about putting cheap sump pumps in the barrels that way you can keep them empty while keeping your water inventory full.
  13. Apr 5, 2010 #12
    :smile:If Pascal's Law Works, Yes it will work:smile:
  14. Apr 6, 2010 #13
    'Pascal's Law' is not applicable if there is flowing fluid in the system (like the water moving from the downspout barrels into the storage barrels).
  15. Apr 6, 2010 #14


    User Avatar
    Science Advisor
    Gold Member

    Any slope of land shouldn't matter as long as your catcher barrel is as high as the tops of the storage barrels and the connecting pipe is not vented. Having the catcher barrels mounted so high could look odd - you might be better with a small 'hopper' mounted higher up the wall, rather than a barrel - with the whole system 'under a small pressure'. If your pipe it wide enough, there shouldn't be a problem getting the water to flow fast enough into the storage barrels.
    It would be a good idea to have all the barrels connected 'in series' using pipes with taps so you could control which was first to empty and maintain a reasonable height / head of water in the remaining ones.
    Mosquitos can be controlled using a few drops of oil on the surface of the water in each barrel.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Water Distribution - Rain Barrel System
  1. Water Pump System (Replies: 0)