Radial thrust bearings

Discussion in 'All Things Boats & Boating' started by owene, Sep 14, 2009.

  1. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Hi, I'd appreciate the advice of anyone who has already figured this. Am building a 40' Hartley powered by a 6BT59M/330HP thru a 2.86:1 PRM box. Motor is on rubber and coupled to the 50mm prop shaft with a 30" long sliding spline shaft with a UJ on each end. The inboard end of the prop shaft has a half flange bolted to the rear UJ and it's 'bearing-useable' surface is 80mm OD. The outboard end is swinging a 38x23 4 blader. All OK so far.

    Now, the question relates to axial thrust bearings. Clearly the prop shaft needs one to transfer the thrust into the hull. Short of becoming a mathematician overnight, are there any 'rule of thumb' formulas to determine the actual thrust? Can anyone advise what type of bearing and even a part number range of one of the big bearing makers? Do I need two axial bearings (reverse thrust as well) and one radial bearing all in a set or does anyone make a combined bearing that will stand this sort of thrust? This is all a black magic science to me and I'm receiving totally contradictory advice from all of the local bearing shops.

    Some have suggested that I dump the existing setup and buy a 'pre-made' unit (at huge expence!!) but the whole point of using 'off-the-shelf' parts is to simplify long-trm maintenance. Thanks, Owene
     
  2. marshmat
    Joined: Apr 2005
    Posts: 4,127
    Likes: 149, Points: 63, Legacy Rep: 2043
    Location: Ontario

    marshmat Senior Member

    Hi Owen, welcome aboard :)

    Regarding prop thrust. You want a quick rule of thumb, here's one:
    F = Total thrust in kilonewtons
    P = Shaft power in kilowatts
    V = Speed in m/s
    n = "Efficiency" (well, in a manner of speaking.... use something like 0.5 to 0.7)
    Then P = F * V * (1/n)
    or F = (P * n) / V

    So 330 hp (246 kW), let's say for the moment she tops out at 20 knots (10 m/s), perhaps a bit optimistic so sub in your own estimate.... that gives (246 * 0.7) / 10 = 17 kN thrust, ie. about 3800 lbf.

    This is of course a very rough guess just to get us started. (I would highly recommend picking up a copy of Dave Gerr's "Propeller Handbook"- less than $20- the methods discussed there will give you a much better idea of what prop would be suitable and how much thrust it will produce.)

    As you can see, though, your drivetrain is probably churning out something on the order of 17 kN, or about a ton and a half of force. You can see how this thrust bearing is not going to have an easy life- and you can also see why you have a two-inch solid shaft.

    I share your shuddering at the thought of paying through the nose for something fancy like an Aquadrive CV axle. But you do need a thrust bearing that can handle quite a bit of force, plus a good safety factor. At least a few of your local bearing shops should be able to source something suitable that can handle three tons or so axially, as well as supporting the shaft radially. All sorts of industrial equipment uses these bearings.
     
  3. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Many thanks Mat, that clarifies my thinking. Yes, your figures are pretty well as near as can be. On that basis I contacted the local SKF bearing supplier but sofar seem to be going around in circles! They are asking for radial force calculations as well but my gut feeling is that they would be nominal. The majorty of the forces will be axial (ie along the shaft). I wonder if anyone out there can give me any vague direction in terms of bearing numbers? Do I need to install a roller type bearing with an axial one to centre the shaft or does someone make the complete thing? This issue must surely have been experienced by many before me and there must be a simple solution! Even some bearing numbers of another functioning setup would be helpful...
    Regards Owen
     
  4. anthony goodson
    Joined: Mar 2007
    Posts: 451
    Likes: 19, Points: 18, Legacy Rep: 256
    Location: Dorset UK & Murcia Spain

    anthony goodson Senior Member

    From a very old and much coveted SKF catalogue ,have a look at Double row angular contact bearing 3310D. id 50mm od110mm dynamic load 93600N limiting speed in grease 4000 /oil 5300.
    I am not saying that this is your correct spec ,but it is a starting point,I use its smaller brother in a jet unit, doing a similar job, be aware however it is intolerant of misalignment so make sure that the housing you use is sufficiently strongly mounted and properly lined up.
     
  5. hartley
    Joined: Feb 2006
    Posts: 90
    Likes: 2, Points: 0, Legacy Rep: 36
    Location: australia

    hartley Junior Member

    thrust bearing

    Bit late with this but here goes ...The type of bearing that was commonly used with your setup many years ago was a double row self aligning bearing with extended inner ring and eccentric locking ring, these would take thrust both in fwd and reverse and were trouble free .there was also a cast iron housing for these bearings ,the method of mounting the bearing in the housing was very ingenious ,hard to explain.any decent bearing supply should be able to show you ,get to see a old bloke, young blokes don't know .
    and in case you think the locking ring wont hold (it will) put a split collar on each side of the bearing,the shaft will never move ,except rotate of course
    the setup is very similar to a plummer block,but different,I have just come up with a number out of an old exercise book filled with useless information this number i think is for a inch and a half shaft.......UEL208-108D1W3
    there was also another setup a bit more sophisiticated using a bearing with tapered bore and adapter sleeve .I have a ''Steyr'' bearing book with listings ,i would have to hunt it up .BUT a word of warning these systems were in common usage 30 or more years ago using engines such as perkins ,ford,bmc etc around 120 hp or thereabouts ,your engine is considerably more powerfull and higher revving .......cheers Hartley
     
  6. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Thanks Hartley, Never too late!
    Since first raising the issue I now have a better understanding of the issues so for the record will note the various points here. I've also put a schematic drawing at www.medistat.net/drivetrain.jpg

    1. The axial rating is seldom given for bearings but in the case of Shperical Roller Bearings (SRB's) is about 30% of the radial rating.
    2. Cylindrical Roller Thrust Bearings (CRTB's) appear to be able to stand vastly more axial loads than any others. However they are uni-directional so two are required. They also have no radial rating which means that a simple roller bearing should be sandwiched in between the CRTB's to maintain shaft axial positioning.
    3. CRTB's can't tolerate any radial misalignment so I'm hoping to find a self-centering type of housing. Although the thrust will be driven into a rigid bracket bolted between the engine bearers (each are 150x150x3000), I ponder whether the minimal hull flex might be too much to fix the bearings rigidly.
    4. Taper type locking assemblies are a no-no in this application because I'm told that a relatively small axial shock load could drive the taper into the bearing and expand it's dimensions.
    5. As it's clear that a shoulder of some sort is required on the shaft, I'm also exploring the 'grip' figures of split collars. It may be that these are adequate to hold the sorts of thrusts we are talking about here.

    Out of curiosity, I priced an Aquadrive - $4500NZ and given that I have all of the setup in place except the actual thrust assy, well you work it out!!
    Thanks for all comments.
     
  7. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Now got it sorted

    Finally aftermuch ado and misleading information from a range of bearing suppiers, none of whom seem to understand their own products or more likely simply didn't want to help a one-off customer here goes.

    I bought a pair of Timken HR30310DJ bearings 50mmID x 110mmOD x 34mm and mounted the outers back to back about 25mm apart in a sturdy casing turned from a hunk of 200mm OD solid. To that I machined a pair of end plates each holding a 50mm double-liped seal and the whole assy is sandwiched together along with the steel thrust plate that's part of the hull. Thru the middle of that passes the 50mm shaft into which is grooved a pair of circlip grooves. Pre-load yet to be properly determined but currently have about 0.020" of end float which can be shimmed to suit (anyone any ideas?). So the whole unit will stand about 114kN of dynamic axial thrust (I have 17kN) - it should surely last!

    In due course will put some piccies up at www.english-family.net/boatbuild but am still at the hull/woodwork stage.
     
  8. gonzo
    Joined: Aug 2002
    Posts: 16,790
    Likes: 1,714, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    Why are you not using a standard marine gearbox which already has thrust bearings?
     
  9. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Yes I am using a PRM1000D box which will take the thrust. But, as the motor is mounted on rubber, there has to be a flexible coupling between the box and the shaft - in my case a 600mm long sliding splined shaft (same as a truck prop-shaft) with a big UJ on each end. So, the thrust has to be directly into the hull.
     
  10. gonzo
    Joined: Aug 2002
    Posts: 16,790
    Likes: 1,714, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    No, all you need is a standard coupler. Align the faces and torque the four bolts.
     
  11. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    And have the inboard end of the shaft floating around with the motor? How would I take care of the difference in angle between the gearbox output flange and the prop shaft?
     
  12. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    Gonzo, read post #1, Owene uses a sophisticated drive-train.

    Owene, the bearings are good, but the simmer rings have the same OD. Which part takes the axial force? If that is the seal you should make 2 washers to fill the cavity, so the force is transmitted over a larger area than just the rim.
    And make sure the circlip grooves are extra deep so they can't pop out.

    Preload on the bearings isn't necessary, just a shim to reduce the end float so there will be no hammer-blow when you change direction. The 50 mm shaft has a considerable mass!
     
  13. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Thanks CDK, my apologies I forgot to explain that the inner of the housing that holds the two tapered Timken's outers has a 12mm shoulder at each end so the axial thrust in either direction is thru the bearings against that shoulder. Yes, the circlips have a 2mm groove in the shaft although if there is anything I feel unsafe about it's these clips. Does anyone have any experience or knowledge of split collars - ie what sort of axial grip have they got? Perhaps I could machine up a pair of split collars with a matching shoulder to fit into the circlips' groove and forget the circlips....??
     
  14. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    Especially the rear one has a hard life. Look at the way a valve stem is held in any engine, large or small. The valve spring keeps the tapered collars firmly in their very shallow groove.
    You could do something similar at the rear only. Two halves and a washer, recessed on one side to keep the parts seated deep in the groove (which I should have made a bit wider). At the front the circlip will do fine unless you plan to reverse at full throttle regularly.
     

  15. owene
    Joined: Aug 2009
    Posts: 20
    Likes: 0, Points: 1, Legacy Rep: 10
    Location: Nelson, NZ

    owene Mr Owen Charles

    Yes, although if I'm going to the effort of machining one up, I may as well do three (one spare)..... Good food for thought.
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.