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

Rotary Seal Ideas

  1. Oct 17, 2011 #1
    Hey guys,
    So I need some ideas on how to best make some rotary seals (or a good alternative) for a school project I'm working on (which has a very tight budget).

    I'm creating some small PVC pressure vessels to put some medium sized RC servos in, and I need to run the servo's drive shaft (or an extension of it) outside one of the bulkheads of the housings, which then attach to a thicker rod above it. My plan so far was to create the PVC housings, and then go from there on designing the seals...

    So I was hoping to get some ideas on some different types of seals for this shaft. I guess I can always use some design out of parker o-ring's handbook, but I wanted to ask around before I do that. Any ideas?


    - System is going to be operating at 20-30ft in fresh water (and possibly a few pools), and I was hoping for a safety factor of around 3, so I'd like it to be good for 80-90 ft.

    - I have access to a machine shop with 2-axis CNC mills, and also some decent lathes...I know the mills are good to around .002" ish, but I'm not sure on the lathes...so i'm really hoping that there is good alternative that doesn't require a lot of precise machining (although not out of the question).

    - Cheap

    Any suggestions on a good/cheap alternative to the PVC pressure housings would also be welcome.

    Lemme know if I need to clarify something,

    Last edited: Oct 17, 2011
  2. jcsd
  3. Oct 17, 2011 #2
    A neat idea my robotic team over here had for a submersible was instead of having moving parts extend out of their sealed container, have them move powerful magnets inside the container which in turn move magnets outside of the containers to operate whatever needs to be done.... Of course with this method there is a slip factor (sometimes the torque of the servo will overpower the force of the magnets).

    Aside from that, I don't know.
  4. Oct 18, 2011 #3
    Hmmm, that's an interesting idea. How much did you guys look in to it? Like did you look in to it enough to find enough issues to not develop this system?
  5. Oct 19, 2011 #4
    It was more of an after-the-fact sort of idea. They already built a submersible before coming up with it. I have no idea what kind of seals they used for it though.

    I do know that automotive seals probably would not work as they would need to be installed backwards (because of the orientation of the "sealing flaps" for lack of a better term) and they aren't meant for such pressure differences. There are some vehicles that have seals that would be oriented correctly for your application though (such as the rear wheel bearing seals on some AWD subaru foresters/outbacks (there are 2 per wheel, 1 has the flaps oriented normally, the other is backwards) The ones that have this have drum brakes on the back.). This is not to say that they will be the right size or even remotely suitable for your application though.
    Last edited: Oct 19, 2011
  6. Oct 19, 2011 #5


    User Avatar
    Gold Member

    Have you entertained the idea of making multi-stage seals? If space is not a deal-breaker, you might be able to make seals that reduce pressure-differentials from outside to inside by ~X% at each sealing surface until you have the pressure-resistance that you need. In industry, this is called a labyrinth seal. It is very low-tech and robust and relatively easy to design and produce. Good luck.
  7. Oct 19, 2011 #6
    What about graphite packing? run it in properly and it should easily hold to 20-30 feet. I would test it before expecting it to go to 90 feet though.
  8. Nov 1, 2011 #7

    Difference rotary seal design suitable for difference application.
    Therefore 1st, you need to ask yourself some of the questions below:

    What is your function of your rotary seal? Prevent grease leak out from the system or prevent dust/dirty to go in the system?
    What is your working pressure range?
    What is your working temperature range?
    What is your contact surface material?
    What is your rotary shaft speed ( max & Min)
    What is your sealing medium?
    What is the size that you going to make?
  9. Nov 9, 2011 #8
    Sorry for not responding sooner.

    As far as the questions, I think most are answered by my first post but here:

    1. To prevent water from getting into my pressure vessel

    2. 60-80ish psi

    3. 40-80 degrees F

    4. Shaft and cylinder I was planning on making out of aluminum

    5. Its a servo shaft, so under 100 rpm, probably closer to 30, but i'm not entirely sure.

    6. Don't know the medium yet.

    7. 1/4 - 1/2" shaft

    And remember that friction is a pretty big issue.

    Either way, I've been doing a lot of reading, and I have two main questions at this point, which I hope someone will be able to help me with.

    1. Would an oring loaded lip seal (like the rectangular ones on http://www.mcmaster.com/#mechanical-seals/=ev3uon part #:9514K119 ) work for this application? It seems like they would, but I don't know how well they can deal with pressure from both directions. Anyone have experience with these?

    2. I've been trying to create an oring groove design to seal a standard shaft size like 1/4" or 1/2", but I haven't been able to find an oring that will meet these standards. So far I've been using a combination of the parker oring handbook and their software inPHorm, but the software sucks, and, at least as to what i've seen, the handbook doesn't help much for grooves that aren't the ideal size for a specific oring. How else should I go about selecting an oring or designing the groove for a standard shaft size?
  10. Nov 9, 2011 #9


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi Sideways,
    Those seals are only intended to prevent leaks in one direction. They are pressure energized so they need pressure on the O-ring side to provide a leak tight seal. Pressure on the back side will leak through. Also, they are primarily for axial motion. To be used for a rotating shaft I'd think they would need to be fixed in a bore somehow. It might work but you'd be better off getting a seal that is designed for the application.

    Check out: 5154T48
    This is a rotary seal that's pressed into a bore and can handle up to 90 psi static and 50 psi at relatively high rotational rates, so it should be adequate for your pressure needs. Like the other seal it only works in one direction though. You could put two of these back to back but I wonder why you want a bidirectional seal.

    Regarding O-rings, they're really not intended for rotational sealing either. Like the cup seal you pointed out, they can rotate in the bore, so some means of preventing rotation would be advisable. You might consider clamping them in a bore. The ID and OD of the gland would be fixed but the axial length of the gland would be adjustable using a threaded arrangement. By increasing the clamping force with the threaded arrangement, you could crush the O-ring axially and it would extrude out radially, making contact with the shaft to a degree dependent on the axial load. That way you could tailor how much friction you had between the O-ring and shaft by adjusting the threaded part and still maintain a seal.
  11. Nov 10, 2011 #10
    March Pumps uses cup-shaped magnets over a cup-shaped nylon housing with a magnet inside. Thus torque is transfered via the magnets.

    If you're going slow and have some bellows material handy (or possibly hose?), there's always the wobble drive.

    Essentially, your motor or servo drives a bent shaft (i.e. 30-45 degrees) that terminates in a bearing at the end of a bellow / hose. The bellow / hose is sealed at the bulkhead and at the bearing, so the shaft is inside and your environment is sealed outside. This method eliminates concerns over small leaks through packed bearings.

    Of course, if rotation is what you desire, it is fairly complicated method. On the other hand, if you are looking to tilt something, it's a good candidate.

    Best Luck

    - Mike
    Last edited: Nov 10, 2011
  12. Nov 10, 2011 #11

    Based on my understanding, one of the important factor for design or select rotary seal is the PV value. P mean pressure ( MPa) x V mean speed ( m/s). However, based on your feedback, the PV value of your application is quite low. Due to your low friction requirement, you may consider the rotary seal design similar to Zurcon Roto Glydring S as per this website : http://www.tss.trelleborg.com/com/www/en/products/rotary_seals_shaft_seals/Product-Detail_9_5708.jsp

    Rubber O-Ring as the energizer ring and PTFE Ring as the sealing ring. It is a bi-direction rotary.
    As per your feedback, you may going to use aluminium as the shaft material. Aluminimum is a soft material. I think you need to select the correct PTFE ring material. PTFE based material in carbon fibre filled may be is a good choice for the seal ring as this material is good for soft material surface sealing and good to seal water. Another suggestion for the seal ring material is you may consider the Zurcon Z80 material which mention in the catalogue for the above website.

    I hope my feedback help you.

    Thank You.
  13. Nov 13, 2011 #12
    Are you saying I don't need a bidirectional seal? I was just worried about the internal pressure because I didn't know how it would effect a lip seal. Do you think it'd be fine with one?

    Do they let you buy in small quantities? Or where would I buy one of these? And how much do these generally run?

    Thanks for the replies
  14. Dec 5, 2011 #13
    *Bump* :)

    Anyone else have experience with metal-rubber shaft seals (and their ability to deal with internal pressure)?
  15. Dec 6, 2011 #14
    you should probably be looking for something made out of NBR rubber and exclusively for rotary shafts (o-rings are usually for static elements).

    I suggest you take a look at the simmerings on simrits online catalog

    Now from all those options i suggest u take a look at BA and BASL types. Once you locate the simmering matching your bore and shaft diameters click on the link and open the "product data sheet", there u will find 2 important facts, one of them being maximum pressure load and the other surface treatment requirements.

    Im afraid of one thing and thats you using an aluminum shaft, see the problem is that not only does aluminum have a higher coefficient of thermal expasion than steel, the simmerings are made of polymers, which contain crystal structures. These crystals and friction cause wear on the shaft, thats why you also need to calculate your materials surface hardness.

    anyways hope i helped at least a bit.
  16. Dec 7, 2011 #15
    You may want to look into some motorcycle suspension seals. There are some internal seals that are made for small aluminum shafts, that have to hold high pressure fluid on both sides. Although they aren't really designed for a rotary use, they are very cheap and easy to get. Might be worth a try.
  17. Dec 9, 2011 #16
    Assuming your submersible doesn't collapse, there will be much greater pressure outside of it than inside. Thus, I doubt a bidirectional seal is needed. Water will be trying to get in with much more force than the air trying to get out. You could use a bidirectional seal if you want, but it's not necessary.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook