Announcement

Collapse
No announcement yet.

Dividing head rework

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

  • Dividing head rework

    At the peak of "I want to have everything" I got a screaming deal on a B&S clone TB01 dividing head (CL) about 10 years ago. Well... like a lot of acquisitions from that period, I just got around to needing it.

    It was "like new. The prior owner said he got it "years ago" (probably late '80) and only used it once; that's once more than I've used it. lol. I want to cut about 15 gears for my lathes, so now's the time to dust it off.

    Hmmm... feels a really stiff. Worse, it of ratchets a bit at points in the rotation, particularly when changing direction. I figure its got swarf or casting sand caught in the dried out pig fat used for grease. That was a correct assumption, but not the cause and there was far more wrong.

    Here's the DH after I repaired it - doesn't look a lot different to how it was before it started work on it. I didn't intend this to be more than a quick clean/lube, so no pics of the disassembly. Pretty well made - tapered roller ( 32009 Nachi) at the spindle top, thrust bearing at the spindle tail, spindle all ground and nicely finished. Overall fit and finish was good. Hey, this should be a breeze.



    Yep.. it was full of grit inside (teaspoon or more non-magnetic sand), but none of it looked like it came though the roller bearing (from grinding wheels etc). The internal casting was sealed with paint where they could get to it easily with a brush. I suspect the grit came from the 10% they didn't paint. Strike 1. The original B&S design was basically good. No idea if it's an original flaw, but the top roller has an open gap (about a 1/16th) at the top and bad crap can get in where it can damage the roller bearing, Rollers have no ability to easily move junk out of the way. But in this case, after washing out, the roller and cup were like new. Later I'll put a 75mm (id) oil seal around the spindle flange to seal it.

    The spindle bearing cup and holder:



    The ratchet feeling wasn't the grit or hard grease (awful grease had dried rock hard); it was a cracked bearing disc on the rear thrust bearing. Every time the roller dropped into the crack it felt "odd".



    But wait.. How can the roller drop into crack if the plate is pressing on a solid machine surface...? uh oh.... it's not solid. Frigg... the casting was machined to the correct depth, but an occlusion still existed: (47mm plate, 51mm diameter hole):



    But... the occlusion still doesn't fully explain why the plate cracked.
    Last edited by lakeside53; 03-05-2015, 01:52 AM.

  • #2
    The thrust bearing and tapered bearing are pulled together (preloaded) by a threaded collar on the rear of the spindle.. To lock the preload, the collar is split on one side and a screw compresses the thread.



    Ha... the damn screw is too long. and when tightened even carefully puts a high point load directly on the bearing. This load was at 6 o'clock in the casting picture above - worst possible place. You can see the shiny spot on the screw end. It protruded about 50 thou... and I could see the contact point right on the bearing plate crack. Basically it was "locking" by driving a screw into the top bearing plate, not by compressing the threads.



    Ok, mystery solved. I might have got away just shortening the screw and leaving the casting alone, but I wouldn't be able to sleep. Mounted the casting in the BP and machined down until the occlusion was gone, about 80 thou... Easier with cnc to just use an end mill, but I could have used a rotary table or boring head.



    Much better :

    Last edited by lakeside53; 03-26-2015, 12:52 AM.

    Comment


    • #3
      Ready for assembly? - fat chance!

      Looking at the spindle threads, I had noticed they looked a bit wide on the top flats but figured they has just relieved them more to stop damage. They measured about at about 1.475 inches, a bit low but "ok". After all... I has a nose protector nut and back plate that screwed on nicely. On a whim I grabbed a 4-jaw check with a 1.5-8 backing plate. Wouldn't screw on at all! Grabbed two more - same deal. WTF? Check again , maybe some wacky metric/whatever thread - nope, 8tpi...

      Decided to get technical. No thread wires so I checked all my #50 drills shanks until I found three that were exactly 0.0700. Probably not perfectly round but good enough for a quick test.

      For a class 2A 1 1/2-8 using 0.0700 wires, I'd expect 1.5184 to 1.5111. My measurement was 1.529. Now it was obvious... it wasn't that the thread tops were cut down - they weren't finished to the correct depth!. Argghh.. worse, they supplied a nut and faceplate that fitted nicely, but not standard. I bet they made a run of beautiful but messed up spindles and decided to "fix" it by supplying two matching parts. lol

      Thread calc I used. You can enter any wire sizes or select the class. http://theoreticalmachinist.com/Thre...dImperial.aspx

      File test shows it's not real hard (probably RC28-32 if 4140ph) so I'm going to try to recut them. I've wrapped the roller bearing with blue masking tape to keep the junk out.

      First, dial in the set-tru for zero run out (I could have used soft jaws also).




      Then a bunch of mucking around getting the existing thread synched to the feed. A few trial passes and I'm ready. Damn, things go by real fast at 80rpm on 8tpi. Sure glad I had my carriage limit switch to cut the spindle and apply the brake. I used a 16ER8UNJ full profile insert. I'll end up with a nice radiused root also. Ok.. adjust depth to skim the old thread.. looking good.. 3 thou more... nice... BANG. Busted off the insert on a hard spot near the shoulder end of thread. Reset the set-tru again, rotate the insert, sync to the old thread, dig out parts of the old insert from the spindle and skim again. Sneak up a few thou at a time and all is well. the ER insert are razor sharp, and really strong in the larger profiles so no issues with shaving even at low sfm.



      Recheck with wires; spot on (my target) - 1.517. Grab my other chucks and they are thread on beautifully. Whew... Oh.. note the bed cloth to catch the thread wires when they try to hide in the swarf tray...


      The endplay adjustment part has a setscrew to lock it. That screw misses the smooth zone and mangles the thread. Need to fix that.



      Done. Yes, didn't have much of a hold on the part!



      Ready for reassembly.
      Last edited by lakeside53; 03-05-2015, 04:10 PM.

      Comment


      • #4
        Main castings back together (easier to hold and work on if the rotating casting is in the cradle). Spindle greased, gear mounted and ready for insertion.



        Rear thrust bearing and locking nut. Because I took about 80 thou out of the casting, I need to build that up again. Turns out the bearing thrust plates are 1mm thick, so the two old plates (including the cracked one!) went into the hole before the new thrust bearing. Perfect!



        Now the worm assembly and endplay/backlash adjuster.



        Inserted. This is a good time to adjust the spindle preload - you can't do it with the worm gear engaged. I have a nice Famco tool to adjust the nut, but you can make something really easy with two bolts and a disc. I've moved the eccentric worm mount to the disengaged position for the next picture to enable spindle preload adjustment. Tighten until you feel the spindle just start to bind, then back off a tiny amount and tighten the lock screw,



        Move the worm eccentric such that the worm fully meshes with the gear then back a tad. Tighten the backlash/endplay while rotating the worm so it all moves freely with minimal backlash. By juggling the worm mesh position and shaft endplay, you can pretty much get rid of all backlash if everything is in good shape. Over do it and you'll end up prematurely wearing the bronze gear

        That's it - all done... I'm happy with the outcome. A zillion things could have gone wrong when I recut the spindle threads, but they didn't. lol.

        Next. CNC Stepper control of the handle so I don't have to mess with counting rotations and remainders, but that will be another thread.
        Last edited by lakeside53; 03-05-2015, 02:22 AM.

        Comment


        • #5
          Great job and write-up! I too have one of these clone dividing heads, mine made by Vertex. It had almost the identical maladies that your's had. The only additional imperfection that mine has is at one point in the main gear it's just a tiny bit snugger than the rest of the rotation. Oh yes, I had to file the bronze sector arms because the didn't sit flat against each other.
          Isn't it curious how they can take a good tool apart, reverse engineer it and then make it crappy!
          gbritnell

          Comment


          • #6
            Originally posted by gbritnell View Post
            Isn't it curious how they can take a good tool apart, reverse engineer it and then make it crappy!
            Welcome to my world, you'd likely laugh until you cried if you spent a day in my office. The biggest problem with reverse engineering is that folks not only feel the need to copy, but also to "improve." Not that the improvements in this case necessarily caused the manufacturing issues, but they sure didnt help simplify things....KISS is my motto.
            "I am, and ever will be, a white-socks, pocket-protector, nerdy engineer -- born under the second law of thermodynamics, steeped in the steam tables, in love with free-body diagrams, transformed by Laplace, and propelled by compressible flow."

            Comment


            • #7
              Great work and a nice upgrade to the dividing head! I probably wouldn't have got that far into it, or at least that's what I'm telling myself.

              One thing I noticed right away, which has nothing to do with the work, is that I have the same Mitutoyo 1-2" micrometer and just love it. Mine also was reading without the "1" in front of the decimal but after I played with the buttons and battery and eventually did get the full reading back.

              Comment


              • #8
                Nice write up and a piece of Garbage turned into Gold. I have two small dividing heads, both bought used and abused. The first one had not been lubricated for so long that the grease was some powdery clay like substance that was a pain to remove. After a complete tear-down it is a nice tool. The second one was owned by someone with a 15 inch Crescent Wrench and a 3 pound Drilling hammer. After a complete tear-down and some home made parts it too is sweet.

                In both cases they had been on the shelf for a while and the rebuild was because of a Need To Use situation.

                Comment


                • #9
                  Originally posted by PixMan View Post
                  Great work and a nice upgrade to the dividing head! I probably wouldn't have got that far into it, or at least that's what I'm telling myself.

                  One thing I noticed right away, which has nothing to do with the work, is that I have the same Mitutoyo 1-2" micrometer and just love it. Mine also was reading without the "1" in front of the decimal but after I played with the buttons and battery and eventually did get the full reading back.
                  I hadn't changed the battery for years (ever?) and the display was a bit dull. After changing it and putting in the reference measure, I pressed something wrong and it up and didn't have time to figure out how to get the offset back. Guess I should read the instructions

                  Comment


                  • #10
                    Originally posted by gbritnell View Post
                    Great job and write-up! I too have one of these clone dividing heads, mine made by Vertex. It had almost the identical maladies that your's had. The only additional imperfection that mine has is at one point in the main gear it's just a tiny bit snugger than the rest of the rotation. Oh yes, I had to file the bronze sector arms because the didn't sit flat against each other.
                    Isn't it curious how they can take a good tool apart, reverse engineer it and then make it crappy!
                    gbritnell

                    yep.. I found one more thing that need attention : the o-ring that holds the bronze arms from slipping can't go into the groove because the arms are too thick. I'll need to cut a recess grove into the outer arm.

                    Oh, my main gear/worm is a bit tighter at one point also! Some tiny run out somewhere that make it hard to have zero backlash at all points.

                    Comment


                    • #11
                      Lakeside I just have to ask how do you measure the threads with .700" wires (or just a typo).

                      Comment


                      • #12
                        You can use any wire size with a certain range for a given thread. In the my case of 1 1/2-8UN, that's 0.0700 to 0.1125. You either do a bunch of math or... use this online calc.

                        http://theoreticalmachinist.com/Thre...edImperial.asp

                        This calculator is really good - you can use the drop-down box for wire sizes (and it will give you the min-max useable), or you can type in any arbitrary size. I did not have any wires at that time, so I improvised with #50 drill shanks. I now have wires, until I lose the one I need

                        You can buy the apps(s) for you phone too.


                        For those that haven't used wires before, if you weren't born with 5 hands the the "secret" is to simply put sticky grease on the threads to hold them. But do use a catch cloth - they are almost invisible in you lathe chip tray, especially if it's 6 inches deep of crap like mine.
                        Last edited by lakeside53; 03-05-2015, 04:09 PM.

                        Comment


                        • #13
                          Actually my sector arms are held in place by a c-clip made from spring steel. It has a bow to it so it keeps pressure on the arms to keep them from turning when you rotate the handle. We had a B&S in our shop and it had the same type of spring clip. I can take a picture If you like.
                          gbritnell

                          Comment


                          • #14
                            I just poke them into a piece of closed cell foam, guided by the threads they are going across. No mess, no juggling act.
                            Kansas City area

                            Comment


                            • #15
                              Originally posted by lakeside53 View Post
                              You can use any wire size with a certain range for a given thread. In the my case of 1 1/2-8UN, that's 0.7000 to 0.1125. You either do a bunch of math or... use this online calc.
                              I think you've got the same typo TWICE now.

                              Shouldn't that be a 0.0700" wire, not .7000?

                              Comment

                              Working...
                              X