Announcement

Collapse
No announcement yet.

Turning for a bearing

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

  • Turning for a bearing

    How much smaller does the hole have to be to tap or press in an outside bearing race? I am working on a bandsaw mill and need to know these things.

  • #2
    start reading the bearing co's manual. They're user angry, but have all the info. Its not an off the cuff answer it will depend on the bearing - i.e size, application, type, even the tolerance and material of the shaft/housing for some bearings. Likely this is a smallish deep groove, standard P4, but we don't know the size and you're better armed knowing the correct tolerances

    Generally the rotating part is interference and the non rotating is a very close slip fit

    page 334 and onward here https://www.skf.com/binaries/pub12/I..._12-463040.pdf
    Last edited by Mcgyver; 12-22-2021, 03:05 PM.
    in Toronto Ontario - where are you?

    Comment


    • #3
      Machinery's handbook has a good section on "Fits" and what constitutes a slip fit vs a press fit for various diameters. I wouldn't recommend a slip fit without using bearing retainer.
      21" Royersford Excelsior CamelBack Drillpress Restoration
      1943 Sidney 16x54 Lathe Restoration

      Comment


      • #4
        Outside race usually a snug slip fit. Same size to .0005 larger. It only needs to have more turning resistance than bearing. Inside race pressed on about .0005 larger on shaft. Take apart an electric motor and check fits. If hole gets a bit oversize loctite will save it. For new people advise is stop short .001 and finish with abrasive cloth. Depends on how fine a finish you get in machining. Inevitably there are microscopic ridges when sanded or ground will change hole size. If you have cnc machining you may be able to get size without finishing operation.

        Comment


        • #5
          The fit varies according to whether the inner or the outer race is rotating. It also varies with the type load. Do what Mcgyver says and read the damn book.

          Comment


          • #6
            Size is important.

            A <1" OD bearing or a >6" OD bearing ?

            Comment


            • #7
              Originally posted by Bented View Post
              Size is important.

              A <1" OD bearing or a >6" OD bearing ?
              +1

              I design high-speed rotating machinery, and probably design & detail 2-3 bearing housings/shaft seats every week, and it will all vary with size and application.
              Generally though, I can tell you that we tend to go for around just under half a thou' interference on housings or shafts, although we use the ISO limits and fits tolerances for a little leeway.

              SKF Bearings have some good online data for tolerancing bearing seats, which is worth looking up, but I can also tell you that even they don't always get it right - like suggesting an H7 tolerance on a housing bore for a self-aligning bearing. Which almost guarantees it's going to spin in the bore...

              Comment


              • #8
                Originally posted by deltap View Post
                If you have cnc machining you may be able to get size without finishing operation.
                I truly wish that people would stop assuming that a NC lathe is more accurate then a manual lathe, it is simply a machine that removes the drudgery of running a manual machine.

                For the most part the setup is the same, choose the work holding, choose the tools, set the tools too size, choose the feeds and spindle speeds, program it to do the repetitive hand motions required on a manual machine.

                If indeed you make a mistake in the manual set up portion of this process you will simply make bad parts faster with less effort.

                I often set up machines for thousand part runs operated by employees that are not skilled enough to do it themselves, I leave them a written list.

                Open door
                Un chuck finished part
                Chuck next part
                Close door
                Press start
                Repeat 1000 times
                Fill the coolant sump when it runs out
                Bring me 1 in 50 parts for measurement so that I may determine if the inserts are failing
                Ask me if you think the parts are going out of dimension and I will adjust the operation as needed.

                Have had people turn the spindle on in reverse, not tighten the chuck, not place the part against the stop.
                This often ends poorly.

                Comment


                • #9
                  Originally posted by Bented View Post

                  I truly wish that people would stop assuming that a NC lathe is more accurate then a manual lathe, it is simply a machine that removes the drudgery of running a manual machine.

                  ...........................
                  The NC lathe, once set up, is probably more accurate than the operator. And they are usually newer, less worn, and, yes, more accurate, than the lathes the typical home shop person, or even many shops, may have.

                  But more importantly, it is more "repeatable" than the operator. Shop owners and manufacturers LOVE that aspect of CNC... set it up once, and you can "qualify the process". After that, it is usually fairly easy to determine the causes of excessive variation.
                  CNC machines only go through the motions

                  Comment


                  • #10
                    Newer and less worn is the most sensible thing that you have ever posted (-:

                    Comment


                    • #11
                      Better read the book, like others have said. McGyver posted a download link for a good one.
                      For example what can go wrong, we lost a bearing at work last week.
                      An expensive one.

                      Rear wheel bearing on a semi trailer, the fit got loose between the inner cone and the spindle.
                      50 miles later, the bearing was friction-welded onto the spindle.
                      Shortly thereafter the entire wheel etc was seized solid.
                      It was incredibly painful and expensive,
                      but you can replace individual trailer spindles as a replacement item
                      (the bearings are ~ 4" dia, the entire axle is north of $20K)

                      I just thank goodness it wasn't my job, but some pf the mechanics were taking pepto-bismol
                      25 miles north of Buffalo NY, USA

                      Comment


                      • #12
                        I would play safe and make the parts with +0.001" clearance and use a little Loctite.

                        Comment


                        • #13
                          How accurately can you measure the bore you are making? And second (just as important), how good is the final surface finish? It's likely easer to make an OD go/nogo guage first, then your bore. Be guided by the bearing manf data. If too tight final bore will affect the bearing internal clearance, probably not an issue in your application but be aware of it.

                          Comment


                          • #14
                            A pretty low standard.....................
                            CNC machines only go through the motions

                            Comment


                            • #15
                              Originally posted by Bented View Post
                              Newer and less worn is the most sensible thing that you have ever posted (-:
                              A pretty low standard.....................‚Äč
                              CNC machines only go through the motions

                              Comment

                              Working...
                              X