Announcement

Collapse
No announcement yet.

Truing Up Starrett 8" Level

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

  • Truing Up Starrett 8" Level

    I have one of these levels that I got at a flea market years ago. I've always noticed that there is some discrepancy between the V groove and the flats.
    It looks like someone ground the flat edges of the bottom at one time and they don't coincide with the V groove.
    I can't grind the V so I was thinking of setting this up on my grinder in my tool makers vise. Setting a gauge pin in the V and indicating it to be the same on each end, then grind the flats in. This should true it up. Any thoughts???


    JL.................

  • #2
    If you can clamp it such that it doesn't distort the bar into an arc like condition I'd say this is a very reasonable plan.

    Before you do that have you checked the level to see if the flat or the V is correct? I'd suggest using a flat machine table or even a shimmed surface plate. Check for level one way then flip end for end and compare the reading. That'll quickly tell you if the flat is truly parallel to the bubble vial. Then check the "V" by using a couple of ball bearings that fit up into the "V" as contact points. Use something like sheet metal or thick card stock with two holes in it to hold bearing balls so they are in contact with the same flat machine table or plate but can't roll around.
    Last edited by BCRider; 10-26-2016, 05:55 PM.
    Chilliwack BC, Canada

    Comment


    • #3
      Yes, the flats coincide with the bubble, the V does not. I could calibrate the bubble to coincide with the V, but I don't want to keep messing around everytime I have to use it.

      JL.............

      Comment


      • #4
        In that case I'd calibrate the bubble to match the V and go ahead with grinding the flat to match. Which brings us back to your original good sounding idea.

        Must be nice to have access to a surface grinder. I'm envious.
        Chilliwack BC, Canada

        Comment


        • #5
          I learned something new today............... I called Starrett, the tech guy told me that the V and the flats never coincide or are parallel.
          Reason...........govt. standards on concavity specs for one. Second the center of the level is not supposed to touch. He said these are sprung or bent during manufacturing for that reason. The flats and V should only touch on the ends. Kind of like the ways of a lathe saddle.
          He also said that the level would have to be calibrated for either the V or the flat depending on which surface you wanted to use.
          What a PIA. I described my idea to him on how I wanted to true this up, he said it would work if I wanted to take the time to do the setup but this kind of setup is not feasible from a cost perspective for the factory to do.
          Learn something new every day.

          JL................

          Comment


          • #6
            As always, money rules!
            Bill
            I cut it off twice and it's still too short!

            Comment


            • #7
              "Setting a gauge pin in the V and indicating it to be the same on each end, then grind the flats in. This should true it up. Any thoughts???"

              You have some nice kit so I believe you have a plan.
              However, testing the error of Vee vs. "flat" must have yielded repeatable error, and if it is not great I would scrape it in.
              Mindful of heat in any event.
              You didn't state which Starrett you have.
              Silk purse analogy.

              Comment


              • #8
                How much of a shim would it take to match them?

                As for wanting the ends to touch that's interesting. But it means that for any other use we are relying on being able to position the level so that there's equal overhang so it sits on matching points of the arc. And of course leveling into a hard edge with only one end of the level in contact isn't any good.

                Makes me wonder how many with such levels have always assumed that the flat is truly flat and used it accordingly. And then trusted the machines in error.
                Chilliwack BC, Canada

                Comment


                • #9
                  It's this one..............



                  JL...........

                  Comment


                  • #10
                    Is that .005"/10"?

                    Comment


                    • #11
                      Originally posted by Carm View Post
                      Is that .005"/10"?
                      I always thought it was .005 in 12".

                      JL...............

                      Comment


                      • #12
                        Originally posted by BCRider View Post
                        How much of a shim would it take to match them?

                        As for wanting the ends to touch that's interesting. But it means that for any other use we are relying on being able to position the level so that there's equal overhang so it sits on matching points of the arc. And of course leveling into a hard edge with only one end of the level in contact isn't any good.

                        Makes me wonder how many with such levels have always assumed that the flat is truly flat and used it accordingly. And then trusted the machines in error.
                        That's what shocked me........ I always thought they were dead flat across the bottom. I guess if the level overhangs what ever your setting it on go to a shorter level !!!! With a level that reads with this accuracy I don't see how you could possibly position it so there is equal overhang on each end. Just a couple thou difference in one side overhanging more than the other could cause a pretty big error.
                        I think these levels come in sizes of 2" increments.

                        JL................

                        Comment


                        • #13
                          I suppose it depends on just how much of an arch we are considering. Is it a case of a couple of thou or a couple of tenths? But it does open up the door to a possible error.

                          Given this new knowledge I'd have to agree about why they come in so many lengths to suit each job.

                          Makes me think that there's reason for looking at our own home shop made levels as this enthusiastic member has built and presented in the "Shop Made Tools" thread.....

                          http://bbs.homeshopmachinist.net/thr...28#post1076328
                          Chilliwack BC, Canada

                          Comment


                          • #14
                            Checked my 8" Starrett 98. The center is at most 10 micon (well under a tenth) high on the flat. At this precision, there is no measurable variation across the flat.

                            By my reading of the standards that I have access to (company standards to meet assorted MILspecs), it is not to be convex. The manufacturer will then insure that any error is in the direction on concavity to meet this, but (in what I have access to), there is no requirement about how concave. Only that it not be convex.

                            It is also specified that the level used should not be longer than needed to span the surface being leveled, which would preclude using an 18" level to cross level a lathe with 10" between the ways, but the 18" would be fine to level it along the ways, presuming they are longer than that.

                            Comment


                            • #15
                              Lufkin, for one, made levels that ONLY had pads at the ends. So no question as to how it is to be used.

                              There is no spec on the "non contact", so you have no justification for any "equal hang-over" scheme. The relief could be of any shape.
                              CNC machines only go through the motions

                              Comment

                              Working...
                              X