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

Help on cylinder and seals

  1. Aug 17, 2010 #1
    I am designing a prototype of my patent and I am a little stuck. The main part of my prototype involves the use of a cylinder and a seal that will slide inside up and down with minimal friction. I have already discussed on a previous thread the specs of the seal, the pressure environment, speed at which mating surface slides, etc...

    The problem is that my prototype firm claims it will be hard to make a good cylinder that is 1 foot in diameter and make it "perfectly round" so that the seal does the job in keeping the water from "leaking" through it. I didn't think that this would be that hard but the two firms i saw insisted it would be tougher than i think. Now I already talked to some sealing firms and they have no problems giving me the seal i want. I want a spring energized teflon seal with a small area mating surface (less friction).

    Can someone tell me where or who i can contact to attain a cylinder with these characteristics:

    1. perfectly round - to ensure the seal always is in contact with the wall? (thought that would be easy enough but...)
    2. with the inside cylinder wall with a low low friction surface (i.e teflon lined, chrome, stainless steel, just steel?) - is asking for teflon overkill? is teflon - steel close enough to 0.04 in friction co-efficient??

    3. up to 18 inches in diameter?

    As well, i also need the seal to be housed onto a disc so I was wondering what kind of firm/entity could build the disc with "wear" rings inside to keep the seal in line.

    As mentioned, the most important part is to have minimal friction between the seal and the inside cylinder wall. It is low pressure, no heat, slow moving so the seal doesn't have any real challenges other than to keep the water on one side while allowing movement of the cylinder around the seal.

    Thanks for your time and any suggestions you can make. I really thought the prototype firm would be able to handle something simple like this but it hasn't been fruitful.

    I tried to attach a file but for some reason it won't let me - sorry
    Last edited: Aug 17, 2010
  2. jcsd
  3. Aug 18, 2010 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi gloo,
    A hydraulic cylinder with a 12" ID and a tolerance of +/- 0.001" and a surface finish of ~12 shouldn't be a big deal. It'll be expensive, but it's not that unusual and it's not hard to manufacture as long as it's not very long. Send your drawing to some local machine shops and see what you come up with. If you'd like a recommendation for shops in Pennsylvania, PM me.
  4. Aug 18, 2010 #3
    Can't a spring energized teflon seal make up for the .0001 tolerance differences by pressing harder into the .0001 gaps?
  5. Aug 18, 2010 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Not sure what you mean. For a cylinder 12" in diameter, a typical tolerance for a small diameter U-cup Teflon seal will be +/- 0.001" not 0.0001". And yes, it can float sufficiently that a few thousandths of an inch radial motion won't break the seal.
  6. Aug 18, 2010 #5
    Sorry, i got a little carried away typing the zeros :). Ok, so the spring is not for pushing the seal into the imperfections? I thought the springs were meant for that kind of function.

    Who would be making the casing to house the seal and the wear rings? Would it be the seal company or some other speciality shop? Sorry, I just thought my prototype firm was a little more versatile to help me but it seems they will just want to slap this thing together for me but i have to get the parts.
  7. Aug 18, 2010 #6


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Generally, we don't think of the seal being pushed into imperfections by the spring, but yes, that's essentially what's going on. A seal works by providing a contact stress between it and the sealing surface that is higher than the differential pressure being sealed.
  8. Aug 18, 2010 #7
    Time to learn the word 'tolerancing'. It will become the bain of your life until it becomes second nature.

    You need to have an idea of what 'perfectly round' means. As I can tell you that 'perfect' does not exist. When you can draw a perfect circle, then you can machine one.

    The spring seal will have a design guide in the catalogue, they have it in all Oring and other types of elasomer seal books. This will basically tell you what tolerances and sizes to make your grooves.

    eg James Walkers Radial Seal catalogue
    http://www.jameswalker.biz/docs/jwco/walkersele guide.pdf

    As you can see, it contains a wealth of information. There will be a similar catalogue for a seal you are looking for, that will contain all the sealing design information you'll need.

    EDIT: Also if you are really serious about this get yourself the Machinerys handbook NOW! It'll set you back rather a lot of cash, but it is the single best investment you could make. Absolutely every engineering design office in the land will have at least 1 copy.
    Last edited: Aug 18, 2010
  9. Aug 18, 2010 #8
    haha.. sorry - I am not an engineer and was just sloppily using the term that Qgoest gave me. I do appreciate the heads up - that will be useful information when i take my cylinder and it's specs to the sealing firm.

    I won't be able to make use of buying an entire book as this is my only project/prototype. If it works, iwon't be making this size ever again --hopefully.
  10. Aug 18, 2010 #9
    Hi Goo, this may not be what you asked for or need, but for ease of manufacture and high degree of seal quality, you might be interested in a "rolling sock" type of seal.
    Hope this is helpful. Good luck
  11. Aug 18, 2010 #10
    i can kind of picture what you are saying but i can't find any info on this - even when i google it. Can you give me a link or something?
  12. Aug 19, 2010 #11
    The hand book will tell you about almost anything design related. Even if you only use it for this project, it will be invaluable. You can but a secondhand copy off ebay and then resell it. So in effect 'hiring' the book for the duration of the project. I doubt you'd find it in a library, but you could try that.

    Definately you'll need the seal guide that you are using, they will be in pdf format on sealing companies websites.

    What exact seal are you thinking of using?
  13. Aug 19, 2010 #12
    I understand what you are saying about buying the handbook. I have a fear that it would be lost on me due to my the fact that I am not an engineer and the technicality would overwhelm me.

    I have been talking to a contact at AHP Seals and looking to get a spring energized teflon seal that has a small mating surface since my most important goal is to have the lowest friction possible while keeping the water from moving past the seal. I am going to find a premade seal of the closest diameter i want and get the precision parts company to make the cylinder to the working specs of the seal.

    That is my most important element -low low friction between the seal and inside cylinder wall. The pressure of the water is extremely low (2 to 3 feet of water max) and the movement is slow. The seal doesn't have to be ultra robust because the demand on it (heat, movement, pressure, chemical) is very low. I am going to make sure the seal company will understand this requirement and hope they can give me the ideal seal.
  14. Aug 19, 2010 #13


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I like the "rolling sock" idea but it might be difficult to locate something off the shelf. A diaphragm would also work well if the movement isn't very large, and friction would be nil.
  15. Aug 19, 2010 #14
    Really? how would a diaphragm work? Can you show me a pic or link??
  16. Aug 19, 2010 #15
    Ok, i don't think the diaphragm will work. The set up doesn't allow enough movement . Thanks.
    Last edited: Aug 19, 2010
  17. Aug 20, 2010 #16
    As others have said, ask the seal manufacturer for their recommendation on bore tolerances.

    This is another case of "ask the seal manufacturer for their recommendation" (well, partially.) They may give you just a range of acceptable surface roughness, but they should be able to recommend some preferred coatings.

    Any decent machine shop should be able to handle that. They might hit you up for tooling charges. If nothing else, a shop that deals with large industrial or marine engines likely has the capability to help you.

    Not too sure what you mean by "wear rings" in this context.

    I also highly recommend you pick up a copy of Machinery's Handbook. It really isn't that expensive -- new copies are less than https://www.amazon.com/Machinerys-Handbook-Toolbox-Oberg/dp/0831128003". You don't need to be an engineer to use it. It's full of extraordinarily useful information for the kind of work you're doing.
    Last edited by a moderator: May 4, 2017
  18. Aug 21, 2010 #17
    lots of good info here - thanks Pantaz.

    actually going to obviate that demand for teflon inner walls. I see the coefficient of friction on even dry surface btw steel and teflon is 0.04 (lowest), which is the same is teflon-teflon.

    I only heard of it throught one of the contributors above who made a good suggestions to use "wear rings" which hold the seal in place above and below and stabilize the seal so that it won't wobble as it moves up and down. Good suggestion by him b/c i wondered how the seal would stay in line with the walls.
    Last edited by a moderator: May 4, 2017
  19. Aug 21, 2010 #18
    Also minor point, to sound like a cool kiddie when talking to sealing companies, call teflon by it's real name - PTFE. :P

    Also the more I hear you talk about this, the more I get the image that you have the seal just sat on the outside of the cylinder. Is this the case?
    The seal should be sat in either a groove in the housing, or a groove in the cylinder.

    Can you draw a schematic of the sealing arrangement, as you've been talking about it freely it doesn't sound critical to security, so won't give the game away. Basically you want to get across what the product does, but not how it does it. So if there is something critical about the seal arrangement (ie something new and exciting, just put a box marked x in it's place or something.

    It'll make it much easier to give advice. It'll be also easier for the sealing comanies to help, you can take the drawing and go "I want to seal like this, what do I need?". They will tell you a) if oyu can, b)which of their products to buy c) the geometry needed to house and seal properly d) the tolderances and surface finish needed.
    Last edited: Aug 21, 2010
  20. Aug 21, 2010 #19
    too late Chris.. the jig is already up with the seal company from the first conversation i had (lol). I will try to throw the technical jargon guys have given me here (load, tolerance, contact stress, bore diameter etc.... :P )

    I actually tried to insert an image and can't figure it out. I drew a simple one on Word and saved it and tried to insert but it kept saying "invalid document" after i try to upload. Can you tell me how?
  21. Aug 21, 2010 #20
    You can upload to photobucket (free registration) and then link to that.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook