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

Making a vacuum seal with a screw?

  1. Feb 8, 2012 #1
    I apologize for my lack of mechanical term knowledge. I'm looking for a type of screw I could use in a project. Sorry if the thing is unrealistic or if the part is obvious.

    I attached a sketch of what I'm looking for. The screw would never be completely out of the hole like the sketch shows, but rather it would be screwed all the way in, and then you would screw it up partially until you get the desired pressure, much like a syringe.

    Does anyone know what this part would be called aside from a generic screw pump? I would like to use one to attach one component to another with a vacuum seal. The hole size could be about 5 cm in diameter ideally, and the length of the screw would be about the same.

    And I have a calculation, but I'm more interested in what I've asked about above.

    Using the 5cm height and 5cm diameter, I calculated the force I would get if the screw starts with a volume in the hole of 2mm height, (2*pi*.025^2)*.002 or 0.00000785 m^3. and is screwed until the volume is at a hole with height 5cm (2*pi*.025^2)*.05 or 0.00019625 m^3.

    Using boyle's law, with a starting pressure of 101,325 N/m^2 (1ATM), I calculated the pressure after the volume change to be 405 Pa. This gives a change in pressure of 100,920 Pa, and using the surface area of the screw hole of (2*pi*.025^2), I get an applied force of 396N.

    Attached Files:

  2. jcsd
  3. Feb 8, 2012 #2


    User Avatar
    Gold Member

    I doubt that you can maintain a good vacuum seal in the screw thread. Even if it held initially, it would likely leak rapidly with time, so the seal would break at an unpredictably time.
  4. Feb 8, 2012 #3


    User Avatar
    Gold Member

    Exacting tolerance on the threads and some thick silicon grease. Maybe talk to some scuba divers about that.

    Would also serve to lube the threads, which will need it, under pressure.
  5. Feb 8, 2012 #4
    Thanks for the info, I'll be keeping this in mind. Do you know of any hardware that exists that I can already buy like this? If not, do you think this is feasible? I know I calculated a force of about 400N, which I think is comparable to how much a screw torqued down would place on a surface.
  6. Feb 9, 2012 #5
    Include a straight-sided section below the threaded area -- like the shaft of a hydraulic or pneumatic cylinder. Your seal solution will be much easier.
  7. Feb 9, 2012 #6


    User Avatar
    Gold Member

    Good point. Much easier to keep a seal there than on the threads.
  8. Feb 9, 2012 #7
    having exacting tolerances on a screw would be expensive.
    consider instead using the screw to drive a smooth-sided cylinder. this would have a better seal for cheap.
    I like your idea, what is the final application?
    == is smooth
    //// is threaded



    O====O////// (low pressure)

    ==O==///O/// high pressure

    ====O////O highest presure

    these are two holes. One is the hole in the object that's being pressurized. The other is a pilot hole for the thread- does that make sense, or should i make a more intricate sketch?

    edit: you might want to consider some kind of locking mechanism- maybe drill holes up the threaded body, and put a pin just below the guide hole (so the threads don't work their way upwards)

    edit: picture!
    http://desmond.imageshack.us/Himg716/scaled.php?server=716&filename=50474693.png&res=medium [Broken]

    there's a hole in the side of the chamber/room/object, where the smooth cylinder goes in and out (air tight)
    the cylinder is moved by the threaded portion, which feeds through the guide hole

    I'm half tempted to make this myself =0P
    Last edited by a moderator: May 5, 2017
  9. Feb 9, 2012 #8


    User Avatar
    Gold Member

    I sure don't get it. Why are there holes in anything?

    Perhaps I'm misunderstanding, but I assume what he's trying to pressurize is a chamber, not an object.
  10. Feb 9, 2012 #9
    added a picture. whether it's a chamber or an object, i don't think it matters- we'd just need a bigger/smaller screw/cylinder
  11. Feb 9, 2012 #10


    User Avatar
    Gold Member

    Well, your sketch is missing a piece. The screw needs a piston head that fills the chamber's height and width.
  12. Feb 9, 2012 #11


    User Avatar
    Gold Member

    You can further improve the situation by making a labyrinth seal on the straight-sided cylindrical portion, that is, having either the external or internal member not be a simple cylinder, but rather a stepped cylinder. This could provide a very good seal.
  13. Feb 10, 2012 #12

    The application is to mount something to my snowboard ;) I don't want to drill holes or ruin my snowboard, and it has a flat smooth surface so I think it is ideal for a vacuum seal. I'm happy you have interest and enthusiasm for the idea.

    However, I've sort of moved past the screw for the time being, and I am more interested in using a lever to pull the "plunger" (what the screw is supposed to be). Still considering the screw method tho. Also, I agree, I had thought a locking mechanism might be needed.

    Also, does anyone know if there some kind of hardware like this that I can buy that already exists?
    Last edited by a moderator: May 5, 2017
  14. Feb 10, 2012 #13
    Thank you for that idea, I will need to study what this means a bit.
  15. Feb 10, 2012 #14
    Also, how would I make this? I don't own a milling machine or lathe, and never used one before. I also don't know how to mold rubbers or other materials to make a gasket seal.

    I have drawn a prototype component of it in google sketchup, along with some other components, but isn't that expensive to have someone make prototypes?
  16. Feb 10, 2012 #15


    User Avatar
    Gold Member

    Why wouldn't you just buy one? It will be far easier.
  17. Feb 10, 2012 #16
    I want to, but I couldn't find a part. I'm not even sure what they would be called or who sells them.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook