Announcement

Collapse
No announcement yet.

Double thread one shaft

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

  • #16
    Two thoughts come to my mind.

    First, are you sure that both "pillar blocks" actually have internal threads? Perhaps only one of them does.

    Second, there is always some slop in threads. Your 9 TPI thread looks highly worn. You did not say how many turns that shaft will turn after it starts to engage the threads. The difference between a 10 TPI and a 9 TPI thread is only 0.0111" per turn. If they only engage for two or three turns, perhaps the slop in the threads is enough to prevent them from binding.

    OK, a third thought. It is also possible for one of those internal threads to have an extra large diameter. That would give it extra clearance.

    A closer examination of the situation will probably reveal one, two, or even some fractional combination of all three of these conditions.
    Paul A.
    SE Texas

    And if you look REAL close at an analog signal,
    You will find that it has discrete steps.

    Comment


    • #17
      Originally posted by winchman View Post
      How would you go about "timing" the threads in the pillow blocks?

      You certainly cant do it with a tap that has both threads on it. Maybe you could do them from outside with a tap on each side, but getting them in sync would still be a trick.

      A special tap could easily be ground at two diameters in the one setting at the same lead.

      It would be good if the OP could re check the thread count on each end to confirm that they are different TPI's, or are both the same, which would answer his own question as to why it does not bind.

      Comment


      • #18
        They are different:



        Over the four threads that are "lined up" in the photo, it looks like just about the right amount of difference for 10 vs. 9 TPI.



        Originally posted by luthor View Post
        A special tap could easily be ground at two diameters in the one setting at the same lead.

        It would be good if the OP could re check the thread count on each end to confirm that they are different TPI's, or are both the same, which would answer his own question as to why it does not bind.
        Paul A.
        SE Texas

        And if you look REAL close at an analog signal,
        You will find that it has discrete steps.

        Comment


        • #19
          One might note the deformation of the thread form to confirm "there was a whole lot of binding going on". ;-))

          Comment


          • #20
            This is a strange design. Apparently the two differential threads are supposed to lock the shaft in place. But over the years(?) they have been over tightened many times and now they are worn past the point of providing that original purpose. It is little wonder that this is not a common feature.

            One can only wonder how they were originally synchronized. Neither the two external threads nor the two internal threads could have been made simultaneously with one tool setup (taps or pair of dies or two single point cutters) so they must have each been cut with two operations. And somehow they were synchronized to both start at the same rotational point. Crazy! Difficult and crazy! There must be at least ten easier ways to accomplish whatever the original designer wanted.
            Last edited by Paul Alciatore; 05-12-2017, 01:17 AM.
            Paul A.
            SE Texas

            And if you look REAL close at an analog signal,
            You will find that it has discrete steps.

            Comment


            • #21
              Hard to say without seeing what they fit into.. One could have had a sliding threaded bushing that was then welded in place.

              Comment


              • #22
                Originally posted by Paul Alciatore View Post

                One can only wonder how they were originally synchronized. Neither the two external threads nor the two internal threads could have been made simultaneously with one tool setup (taps or pair of dies or two single point cutters) so they must have each been cut with two operations.
                It might take a bit of futzing around, (or maybe a whole lot!) but I think I could envision how to do at least the external threads to some specification of offset.

                If they were single pointed in one chucking, using one single point tool, with the compound set parallel to the axis of rotation, just about any amount of angular offset could be achieved within fairly close limits.

                On second thought, even the one chucking caveat could be circumvented by matching the start of the second thread by adjusting the compound to get the single point tool synchronized to the start ( or some other reference ) of the previously cut thread.

                Dave
                Last edited by becksmachine; 05-12-2017, 04:47 AM.

                Comment


                • #23
                  Originally posted by Paul Alciatore View Post
                  This is a strange design. Apparently the two differential threads are supposed to lock the shaft in place. But over the years(?) they have been over tightened many times and now they are worn past the point of providing that original purpose. It is little wonder that this is not a common feature.

                  One can only wonder how they were originally synchronized. Neither the two external threads nor the two internal threads could have been made simultaneously with one tool setup (taps or pair of dies or two single point cutters) so they must have each been cut with two operations. And somehow they were synchronized to both start at the same rotational point. Crazy! Difficult and crazy! There must be at least ten easier ways to accomplish whatever the original designer wanted.
                  No Paul , It is rather simple really .
                  You cut the threads in both parts , then you cut off the threaded end (Sq thrd) of the shaft until it is timed to enter the second block. The "fiitting" is far easier than setup for duplicate starts on 4 thread functions.

                  Again however this assembly makes no sense . yes, it would self tighten as you mentioned, but for what purpose?
                  It is a total lack of axial control ...that is, each tightening means more shaft entering the block and a loss of the previous location or position .

                  Rich

                  So to answer the OP's question, make the shaft longer and then grind/cut/chamfer the end thread to fit
                  Green Bay, WI

                  Comment


                  • #24
                    Big opoligies for the time taken to respond to some wonderfull ideas i have been in hospital getting my ball joint serviced so now home again and can have a little time on the PC IMG][/IMG][IMG][/IMG]

                    Comment


                    • #25
                      So, The application is some sort of self securing pivot shaft.

                      Originally posted by Kiwi View Post
                      Big opoligies for the time taken to respond to some wonderfull ideas i have been in hospital getting my ball joint serviced so now home again and can have a little time on the PC IMG][/IMG][IMG][/IMG]
                      I had thought as much by your original description. The competing threads serving to jamb the shaft between the trunions.

                      Good enough, Just remake the part with a somewhat sloppy thread fit and you are good to go!
                      Accommodate any wear on the bore of the "other" member.

                      Comment


                      • #26
                        Ok what i did (well don't ever do that) spun the new shaft up set the lathe to cut the thread started with the 3/4" thread locked the lead screw in for the entire job (just backed off and reverse)to keep the threads timed so machined the 3/4" thread then changed the tpi from 9 to 10 tpi without disengaging the lead screw hopping this will keep the threads timed it did I think but however it all didn't work the 3/4" binds up by itself so i'm thinking it's not whitworth threads after all if metric say M18x2 and M20x2 but i have to return to the workshop in the next few days there has been some question in regards to my terminology its not the best i admit pretty much self taught after the old man passed away forty five years ago Pillar blocks to me is pillar that supports a pillow block is bearings but i think that was worked out

                        Comment


                        • #27
                          Originally posted by Kiwi View Post
                          Ok what i did (well don't ever do that) spun the new shaft up set the lathe to cut the thread started with the 3/4" thread locked the lead screw in for the entire job (just backed off and reverse)to keep the threads timed so machined the 3/4" thread then changed the tpi from 9 to 10 tpi without disengaging the lead screw hopping this will keep the threads timed it did I think but however it all didn't work the 3/4" binds up by itself so i'm thinking it's not whitworth threads after all if metric say M18x2 and M20x2 but i have to return to the workshop in the next few days there has been some question in regards to my terminology its not the best i admit pretty much self taught after the old man passed away forty five years ago Pillar blocks to me is pillar that supports a pillow block is bearings but i think that was worked out
                          Oh! As soon as the norton box was changed, all timing went to hell in a hand bag. You could re-clock by resetting the norton box for the "other" thread pitch and then using the compound to "catch" the freshly cut thread before moving on to the alternate thread pitch yet to be cut. Mind the threading dial position!

                          Test those threads in the trunnions with whatever bolts you may have laying about, or cut a test thread on a piece of 3/4" wire.

                          The discovery process is the enjoyment of any restoration.

                          Comment


                          • #28
                            Originally posted by Kiwi View Post
                            Ok what i did (well don't ever do that) spun the new shaft up set the lathe to cut the thread started with the 3/4" thread locked the lead screw in for the entire job (just backed off and reverse)to keep the threads timed so machined the 3/4" thread then changed the tpi from 9 to 10 tpi without disengaging the lead screw hopping this will keep the threads timed it did I think but however it all didn't work the 3/4" binds up by itself so i'm thinking it's not whitworth threads after all if metric say M18x2 and M20x2 but i have to return to the workshop in the next few days there has been some question in regards to my terminology its not the best i admit pretty much self taught after the old man passed away forty five years ago Pillar blocks to me is pillar that supports a pillow block is bearings but i think that was worked out

                            Kiwi has realised that the 2 threads have to have the same pitch to be able to screw in. All the talk about differential threads locking the shaft in place was just ridiculous given that there are obvious drilled holes for locking pins or screws in the shaft and housing.

                            Comment

                            Working...
                            X