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

How to input objects into the bottom of a water tank.

  1. Jun 16, 2012 #1
    Hello, im 17 years old and studying Mechanical engineering an i started thinking of creating some sort of game ( a bit boring) where children have to grab different coloured balls from a tank of water, i was wondering how to create a dispensing system to allow balls from a dispenser in which they are in air, to be inputed by gravity, into a container which can be filled with air or water, once filled with water, a door leading to a water tank will be opened and the ball will rise to the top of the open tank to be picked out of.

    (picture Attached) Sorry for the poor quality, but it should help convey my message.

    This multi atmosphere container, atmosphere will be changed by two other smaller contanairs equal or grater than the 2- way container, which one will hold air and the other will hold water. These will hydraulicly suck or fill this 2- atmosphere container, with eaither water or air depending on whether the ball is ready to be dispensed or resetting the system.
    At the start the 2-A-container, will be filled with air to allow the ball to go in it. Then the door conecting to the dispenser will close. Then teh container filled with water, will fill the 2-A-C with water, therefore the other container will be holding this air. Then the door conecting to the main tank will be opened and the ball released to the tank and will float to the surface.
    Then the door will close. The container which recently held the water in it and now empty, will then suck the water out of the 2-A-C and return it full of air for the process to repeat.

    I would prefer all the mechanisms to be electrical from the , dispenser, doors, and containers which suck and input the air/water into the 2-A-C

    What I am asking for is, to tell me whether or not this will work, limitations in this design and any improvments or alternatives.

    Thanks for reading, and i will be gratful for any comments.

    Attached Files:

  2. jcsd
  3. Jun 16, 2012 #2
    Any body?
  4. Jun 16, 2012 #3


    User Avatar
    Gold Member

    Don't see why it wouldn't work. With fine tuning and the right parts it looks like fun for the younsters.
  5. Jun 16, 2012 #4


    User Avatar
    Gold Member

    I see only one potential hang-up, being that there doesn't seem to be any incentive for the ball to enter the tank. That would easily be solved by angling the "airlock" tube slightly upward so buoyancy would take effect.
  6. Jun 16, 2012 #5
    Wouldn't water trying to level (height) both inside the tank and the dispenser be a problem?
  7. Jun 16, 2012 #6


    User Avatar
    Gold Member

    My gut tells me this might be over-engineered. The reason it's only my gut is because I can't tell how much you actually need and how much you presume to need.

    So my question is: separate from the way you've devised it to happen, what parts need to happen?

    For example, is it necessary to artificially keep the container with the balls in it air-filled? If the ball-filled container were above the water-level, the container will remain air-filled by itself.
  8. Jun 17, 2012 #7


    User Avatar
    Science Advisor
    Homework Helper

    I think I see the problem you are trying to solve, which is that if the balls float in water, they would also float in the dispenser if everything was full of water. So the balls woudn't get into the tank.

    On the other hand, a piston or plunger to push the balls down the dispenser tube so they pop into the tank one a time seems a much simpler idea.
  9. Jun 17, 2012 #8
    Thanks for replying, that was only really a rough sketch but that what i was sort of going for.
  10. Jun 17, 2012 #9
    Thanks for reply but I was trying to show that the dispenser was fulled with air, as i attempted to show in the diagram. Thats why the Valve that changes its atmosphere was so important to help it fall in,then sealed off.
  11. Jun 17, 2012 #10
    The ball container(dispenser) full of air, will be in a seperate tank than the water, meaning it is not surrounded by water so it could be open to for easy topping up of balls. Its just the valves and the method of inputing the balls in the bottom of the tank without manually doing so, thats why i was trying to create a hydraulic system to take away and fill the middle chamber (inbetween the dispenser and water tank) with water. To enable a easy wat to make sure the ball will rise from the bottom of the tank, to stop hands getting wet.
  12. Jun 17, 2012 #11
    So, does any body have any better ways to dispense the ball in the tank?

    THanks for previous and future posts and help.
  13. Jun 17, 2012 #12


    User Avatar
    Gold Member

    What Aleph and I are both trying to do is figure out if there is a simpler way of getting the result you want, without necessarily all the complexity in the components you have.

    So, I put this question to you: would you like us to analyze the setup as you have it currently designed, (even if we think there might be easier way of doing it) or would you like us to help analyze and modify your design so that the outcome is what you want?

    The answer to this question will help give you the kind of advice you want.

    Assuming the latter, it seems to me, the criteria are:
    - balls need to be dropped into a hopper
    - balls get released, one at time, by some trigger (as yet to be defined), into the bottom of the tank
    - balls are bouyant
    - everything else that you've drawn in your diagram are one solution, and are simply one means to an end.

    Are there any other components to the problem/solution that are critical or immutable?
    Last edited: Jun 17, 2012
  14. Jun 17, 2012 #13
    Thanks for the quick reply, I suppose the design can change, but the design i though of was the simpliest I could think of. ( I normally over complicate/design stuff).

    The criteria you stated is correct

    '- balls need to be dropped into a hopper
    - balls get released, one at time, by some trigger (as yet to be defined), into the bottom of the tank
    - balls are bouyant'

    The hopper can be released, by any means really, however really prefered to be automatic and mechanical/electrical. Furthermore, once the bouyant ball has risen to the top of the tank, it will eventually needs to be reset. Maybe manually/by hand or automatic as stated above. But will need to be able to continue the game quickly and without the supervising to keep the dryest.

    So really , I was wondering if there is a low cost( including low energy cost) to get an object from an enviroment where it sinks ( by gravity to) where it rises ( convection but preferebly by upthrust and buyancy), and for these principles can be applied to some sort of simple childrens game.

    And to answer your other question, the design can be changed but it still needs to be able to be reloaded more than once and for the water tank to have an open top for children to reach in.

    Thanks for replying in advance.
  15. Jun 17, 2012 #14


    User Avatar
    Gold Member

    Is your drawing more or less to scale? ie: Is the tank just large enough for a hand to enter? It isn't important to the design in general, but might place restrictions upon what components are suitable.

    edit: Having pondered it for a bit, I believe that I have a system in mind that would eliminate any need for valving or doors completely and be fully automatic. The aforementioned question about the scale would be relevant to the energy cost.
    Last edited: Jun 18, 2012
  16. Jun 18, 2012 #15
    The design isnt really to scale however, i will prefer it to be larger than smaller so more kids can play. So what system would this be then? An answer wouldbe greatly welcome.

  17. Jun 18, 2012 #16
    Bear in mind this system is likely for very young children. Attempts should be made to make the opening at the top to be smaller, rather than larger. We wouldn't want any young'n's trying to reach in and accidentally fall in/submerge their faces.

    Furthermore, care should be taken that there are no traps / valves / what have you at the bottom of the tank which would be accessible to a child's hand in any situation (Dry or wet).

    Just some ideas to keep in mind.
  18. Jun 18, 2012 #17


    User Avatar
    Gold Member

    I'm in the midst of drawing it up, since it might be a bit difficult to properly explain. If I can't have it done within a reasonable amount of time (today), I'll try using words. Going by your drawing, I'm scaling it around 3cm ∅ balls. That's very easy to change without affecting the design at all.
    Travis, good input. I'm adding a shield to the mechanism because of it.
  19. Jun 18, 2012 #18
    Iy needs to be bigger for more than one person, but this will need supervison. A paddling pool can be just as dangerous. However im concered about the mechanisms, so can you help me improve the mechanisms?
  20. Jun 18, 2012 #19


    User Avatar
    Gold Member

    What is wrong with something as simple as a tube/hopper slightly wider than the balls, and a stick of wood to push them down the tube?

    You could tilt the tube (delivery system) over 45 degrees so that the end of the tube (hopper) is far way from the watertub (game zone) and it would still work. A quick shove of the stick (dispenser apparatus) would eject a ball into the game zone.

    Attached Files:

    Last edited: Jun 18, 2012
  21. Jun 19, 2012 #20


    User Avatar
    Gold Member

    I must admit, Dave, that you have me beat all sideways on simplicity. I do have a modification of yours in mind already, which would automate the stick-poking very easily. The elegance of your solution almost makes me embarrassed to post the one that I came up with, but I'm going to anyhow because I've been drawing it all day. (Still trying to get used to Inkscape.) Here it is:


    Uploaded with ImageShack.us[/PLAIN]

    I'll post the explanation tomorrow.
    Last edited: Jun 19, 2012
  22. Jun 19, 2012 #21
    That is a really beautiful design, alot what I was thinking. Just two questions though...

    Why the aquarium pump?
    When the balls from the hopper have entered the ball wells and rotated to under the water tank how will it float if there is no water under it to help it float? If it would already have water in it from the previus cycle, how will the ball fit in the ball well in the first place if there is water in it resisting to sit in the well completely. If it is not in completly it will be scraping the top of the sump tank and to the eneterance of the water tank.

    I really do apreciate the time you have spent drawing the diagram, thankyou.
  23. Jun 19, 2012 #22
    I believe that could work quiete well. Thanks you for spending time to make that diagram.
  24. Jun 19, 2012 #23


    User Avatar
    Gold Member

    He is using a sump chamber to hold the leakage. The aquarium pump would come on when the float switch is activated, to pump extra water out of the sump (or bilge, as we sailors might call it).

    You can rest assured the water will rush in and displace the ball. There is no doubt about that.

    This is where I think the design will be challenging. I think that, unless you get deeply into water-tight seals and tolerances and such, the mechanism will hemorrhage water. That's what the sump pump is for, but my gut says the leakage rate will overwhelm the pump.
    Last edited: Jun 19, 2012
  25. Jun 19, 2012 #24


    User Avatar
    Gold Member

    Luckily, it appears that I don't have to explain most of it. I expected questions about the very aspects that you both mention.
    Dave has one thing backward, though; the pump runs until the float drops to the level of the well bottoms (to prevent the pump from running dry). The 2mm drains in the wells let them empty so the balls just drop in. If a ball protrudes a bit, it will simply cam down when it gets to an overhead obstruction. I hadn't actually thought of having a cover on the sump tank, but it's a good idea to keep debris out.
    The input port in the bottom of the main tank is surrounded by a Teflon (or other slippery plastic) gasket. An O-ring could be used, but I don't know how well it would stand up to shear stresses imparted by the carousel rotating. I obviously haven't done any calculations (and don't even know how to), but my gut instinct says that the gph of the pump should exceed the leakage rate. I defer to Dave's gut on that, though, since he obviously has more experience with bilge pumps and sealing. If he's right, I'm pretty much hooped.
    The only reason that I used a ramp (actually more of an inverted trough) as the finger guard, rather than a simpler cover, is to deflect the balls to the centre and thus give better surface distribution.
    The automation idea that I have for Dave's design is to simply cut a slot in the feed tube above the water line to accommodate the spokes of a motorized "paddle wheel" to force the balls down.
  26. Jun 19, 2012 #25


    User Avatar
    Gold Member

    Nah. It's possible that your idea is more water-tight than I'm getting from the diagram.

    My gut is simply synonymous with 'I have no idea, but can't resist having an opinion'.

    Yeah. That would be good.

    The crux of my idea is simply that there's no need for the water-zone and air-zone to be artificially kept separate. They do that naturally at the waterline.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook