Announcement

Collapse
No announcement yet.

Ball bearing fit question

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Ball bearing fit question

    I'm repairing an industrial PTO clutch, on a 6-cyl Volvo Penta diesel engine.
    One issue it that the shaft had been fretting and come loose in the pilot bearing.

    I've turned down the shaft nose and sleeved it:-



    Now need to finish it to be the right fit in a new pilot bearing. The question is what size to aim for?

    The bearing is 30mm bore, and 30mm long. I don't have it yet, but need to have the thing ready to assemble when it does hopefully arrive on Monday., otherwise the good people of Stockport might run short of their daily bread, it drives a pneumatic grain elevator for emptying barges.

    The bearing is pressed into the engine flywheel, but it doesn't sit against a lip, it has to be pressed in to just the right depth. If the shaft is too tight in the bearing, there's the risk of pushing the bearing further in to the flywheel before the shaft is fully entered into the bearing. If it's not tight enough, then it's likely to start fretting again. There's substantial side loading on the bearing in service because of the belt drive from the clutch output. If I assemble the bearing onto the shaft first, then there will probably be a permanent and substantial axial load on it, because it is a pretty good fit in the flywheel.

    Suggestions please as to what finish size to aim for? I do wonder whether a bit of low strength Loctite on the shaft would be a good plan.

    Tim

  • #2
    30.00 mm diameter and you get a good fit.
    Amount of experience is in direct proportion to the value of broken equipment.

    Comment


    • #3
      What kind of bearing Phosphor Bronze, Needle Roller?

      Comment


      • #4
        Originally posted by _Paul_ View Post
        What kind of bearing Phosphor Bronze, Needle Roller?

        Sorry, should have specified, thought I had done - it's a twin row ball race.

        Tim

        Comment


        • #5
          Originally posted by Timleech View Post
          I'm repairing an industrial PTO clutch, on a 6-cyl Volvo Penta diesel engine.
          One issue it that the shaft had been fretting and come loose in the pilot bearing.

          Now need to finish it to be the right fit in a new pilot bearing. The question is what size to aim for?

          The bearing is 30mm bore, and 30mm long. I don't have it yet, but need to have the thing ready to assemble when it does hopefully arrive on Monday., otherwise the good people of Stockport might run short of their daily bread, it drives a pneumatic grain elevator for emptying barges.

          The bearing is pressed into the engine flywheel, but it doesn't sit against a lip, it has to be pressed in to just the right depth. If the shaft is too tight in the bearing, there's the risk of pushing the bearing further in to the flywheel before the shaft is fully entered into the bearing. If it's not tight enough, then it's likely to start fretting again. There's substantial side loading on the bearing in service because of the belt drive from the clutch output. If I assemble the bearing onto the shaft first, then there will probably be a permanent and substantial axial load on it, because it is a pretty good fit in the flywheel.

          Suggestions please as to what finish size to aim for? I do wonder whether a bit of low strength Loctite on the shaft would be a good plan.

          Tim
          I think you are asking for the impossible, mainly due to the design of the output shaft.

          The pilot bearing was never meant to carry the side loads that are being imposed in this configuration. Given the nature of the assembly requirements, it is nearly impossible to achieve fits on both the OD and ID of the pilot bearing to carry continuously the rated load that the bearing is capable of. There should be 2 outboard bearings in addition to the pilot bearing, with the pulley either between these two bearings or on the extreme outboard end. This would also alleviate possible problems with the driven disc wanting destroy it's connection with the output shaft. Usually this is some type of spline, but this is not clear in the photographs.

          Probably none of which helps you get the good citizens of Stockport back in operation on Monday.
          Unless, of course, you could get them to agree to a complete re-design of the port facilities.

          Here is my logic.

          Ignoring the differences that might have been present in the quality of fit, (tightness or looseness) between the ID and OD originally, the reason the ID fit wore out before the OD is because of the difference in surface or contact area.

          Consequently, given that one or the other has to be something other than a tight press fit to allow for assembly, it would be preferable to have the fit on the OD be the loose fit. This would allow the greater area of the OD to provide the longest life possible before needing rebuilding again.

          As you mention, and I agree, an alternative would be to use Loctite to achieve a tight fit. This would allow ease of assembly with a relatively loose fit, that would then magically become a tight fit as the Loctite cures. Again it would be my preference to make the the loose fit on the OD as then the Loctite would have the greatest area on which to work it's magic. This would also allow the use of a tight fit on the pilot shaft to achieve certainty in the axial location of the bearing after assembly. You should be able to take enough out of the hole in the flywheel with a die grinder and flap wheel to achieve this loose fit.

          If you are worried about the axial location causing an ongoing preload condition, give the end of the shaft a good wack after assembly but before the Loctite has had an opportunity to set up. It is unlikely that any axial preload would cause any problems, as this bearing would only be spinning when the clutch was dis-engaged and with the motor running and the load stationary. Even if this happens for long periods, it is likely that thermal forces would shove the bearing sideways on the non Loctite fit enough to alleviate this problem.

          Dave

          Comment


          • #6
            Did the original bearing fail? How long did it last? Was it frozen? How rigid is the clutch disc?

            The pilot bearing sees only a static load after the clutch is engaged, and not much of one if the disc is rigid.

            I'd go for a tight slip fit into the ID of the bearing, and let the side load on the shaft drive the inner race of the bearing while the clutch is disengaged. Making it any tighter risks driving the bearing too far into the flywheel, and that's sure to cause problems. You'd probably have no way to see if that had happened after it's assembled.

            The fact that they used a double-row bearing suggests the side load is pretty high when the clutch is disengaged, so there's not much chance of it slipping/fretting as long as the bearing is in good condition.
            Last edited by winchman; 11-09-2013, 12:17 PM.
            Any products mentioned in my posts have been endorsed by their manufacturer.

            Comment


            • #7
              How about a spacer to prevent the bearing being driven too far?

              Comment


              • #8
                Originally posted by winchman View Post
                Did the original bearing fail? How long did it last? Was it frozen? How rigid is the clutch disc?

                The pilot bearing sees only a static load after the clutch is engaged, and not much of one if the disc is rigid.

                I'd go for a tight slip fit into the ID of the bearing, and let the side load on the shaft drive the inner race of the bearing while the clutch is disengaged. Making it any tighter risks driving the bearing too far into the flywheel, and that's sure to cause problems. You'd probably have no way to see if that had happened after it's assembled.

                The fact that they used a double-row bearing suggests the side load is pretty high when the clutch is disengaged, so there's not much chance of it slipping/fretting as long as the bearing is in good condition.
                The original bearing is not in bad order, main reason for replacing it wear to the ID.
                I don't really understand the second part of your question, about rigidity?
                It's a 2-plate fibre overcentre clutch, the fibre plates engage in a splined ring on the flywheel, the iron plates are on splines on the hub.

                It (the spigot) has slipped and fretted quite severely, with a pilot bearing in decent order.

                In response to becksmachine, these clutches have a specified allowable side loading on the output shaft, I haven't checked but the machine designers probably did their (theoretical) homework on that one. However, the wedge belt drive is working near its design limit (I have checked that), so the belt tension does have to be cranked up fairly high to avoid the belts slipping and throwing off.

                Thanks

                Tim

                Comment


                • #9
                  Measured size is one thing, but surface finish is equally important. I'd polish (or grind) the surface to fit at 30.00mm, and if it's a mirror finish, even a very small undersize will slip on easily, so go for nominal and warm the bearing.

                  If you do need to use loctite, 620 is specifically designed for securing in such a way, but you will need a good amount of heat if you ever want to get it off.

                  Comment


                  • #10
                    Originally posted by winchman View Post
                    The pilot bearing sees only a static load after the clutch is engaged, and not much of one if the disc is rigid.
                    In a perfect world, yes.

                    In the real world, maybe.

                    Even if the driven disc is a solid center type, there has to be clearance between the disc and whatever connection it has with the shaft to allow the driven disc to move away from the flywheel when disengaged. It isn't clear from the photographs what this connection is on this unit, usually it is a spline of some type.

                    If this clearance is great enough, it can mean that any deflection of the shaft caused by the cantilevered forces of the overhanging load, will not be carried by the disc itself, but instead these forces will be carried by the pilot bearing.

                    If the driven disc is a spring center type, it is almost guaranteed that there will be sufficient freedom for the shaft and/or disc hub to deflect, thus the disc will carry none of the radial load.

                    I agree, this is a static load as far as the interior components of the bearing are concerned, as there is no relative rotation of the inner race with respect to the outer race, except when the motor is running and the clutch is disengaged.

                    However as far as the fits of the inner and outer races of the bearing are concerned, in respect to the driven shaft and hole in the flywheel , it is not a unidirectional or "static" load. These fits see reversal of load for every revolution of the crankshaft, which are exactly the same as if the races of the bearing where rotating in respect to each other.

                    This type of application is actually more critical as far as fits of the races of the bearings on the shaft or in the hole. This is because with the entire assembly rotating, both the inner and outer races undergo reversal of loading as opposed to an application where the load is in a constant direction on at least one of the races.

                    Dave

                    Comment


                    • #11
                      This is the type of clutch:-

                      http://www.icpltd.co.uk/icp_pdf/AP%2...0724105741.pdf

                      Tim

                      Comment


                      • #12
                        It looks like the shaft is tapered, not splined. I was thinking it would have a disc with a tapered hub, and that would have more rigidity than the usual automotive-type disc.

                        Maybe the other bearing(s) needs to be replaced, too.
                        Any products mentioned in my posts have been endorsed by their manufacturer.

                        Comment


                        • #13
                          Have you considered using a hardened and ground liner that would be pressed on the shaft? McMaster-Carr has them on page 1145 of their catalog in the size you need.

                          Any products mentioned in my posts have been endorsed by their manufacturer.

                          Comment


                          • #14
                            Since no one answered your question, I'll take a stab at it. The ID will most likely need an easy slip fit into the pilot bearing(s) as most do. I'd go with about .002-.0025" clearance. And I agree with Juno about using a stop-sleeve behind the bearings. Use the OD so the bearings can be removed again by the ID.

                            Comment


                            • #15
                              Tim,I have done probably 50 or so of those sleeve up jobs over the years on Twin Disc and Rockford clutches.Winchman has the solution.

                              Problem here is the roller bearings have the shaft ridgidly located in the housing,but the shaft fit needs to compensate for expansion between the roller bearing and crankshaft.

                              Our solution was to use a hardened sleeve just like the needle bearing race Winchman posted.Get a 30mm OD sleeve and use bearing locker to fasten it to the shaft leaving the sleeve/bearing ID dry for a slip fit.

                              It looks like you already have a sleeve shrunk on,so you could also turn and polish it to 30mm tapering up to 30.02-30.03mm.Then it would start in and tighten up as it's drawn home by the housing bolts.I would use the bearing itself as a ring gauge to check the fit,probably make it so the first 20mm goes in by hand and the rest slides in on pull up.
                              I just need one more tool,just one!

                              Comment

                              Working...
                              X