Announcement

Collapse
No announcement yet.

help with a machinist level purchase

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

  • help with a machinist level purchase

    ok, so i've had my sb9 lathe for a couple years, and it sits on the floor where it landed. i want to get it correctly set up so i can tell what is the machine's fault and what is mine. from what i have been able to learn from reading, there are machinist's levels that are far more sensitive than a basic carpenter's level. if i am to find one of these, what am i looking for?

    keep in mind that this is purely a hobby, and not even one that i do very often. i can't bring myself to spend the amounts required for some of the high end precision tooling that many of you use. i do, however, have a local shop that buys and sells a LOT of used tools. they have new stock nearly every weekend, and if you are lucky enough to get there early on a stocking day, you can find some good stuff on occasion. last week i grabbed a starrett edge/center finder for $3.00, and i've dragged home some new / newly resharpened / barely used domestic mills for under a couple dollars - even cobalt and carbide ones.

    i want to know how to identify a machinist's level - what are the characteristics to look for. how big or small should i get? are there any new ones that aren't ridiculously expensive, so i have a base idea on what a used one should cost.

  • #2
    If you just have to have a precision level, you can get the so-called "master precision" type, typically 0.0005" per foot sensitivity for as low as maybe 50 or 60 bucks from several importers in teh US.

    I don't know that I recommend that.

    You probably should look for a 6" size Starrett 98 level. Longer is OK, but I find them less handy.

    http://www.starrett.com/metrology/product-detail?k=98-6

    While this usually draws loads of scorn for "imprecision", the use of this type 0.005" per foot level is practical and useful, because it isn't "crazy-sensitive".... while it IS sensitive enough to do the job.

    The 0.005" per foot shows up on the level as a movement of the bubble by one marked division.... But if you simply make the bubble repeat position, which can be done to a fraction of a division, you can get a leveling within 1 to 2 thou.

    Don't be fooled though..... there are also the # 97 levels from Starrett that look the same but are basically a carpenter's level. The 98 will seem quite sensitive... a minute movement will send the bubble from one end to the other. With the 97, it will take a very noticeable lift of one end.
    1601

    Keep eye on ball.
    Hashim Khan

    Comment


    • #3
      Here is my take.

      Your lathe needs to be straight with no twist. Ideally all the feet equally loaded.

      It has no need to be level that I have discovered yet.....sometimes lathes are on ships.
      "...do you not think you have enough machines?"

      Comment


      • #4
        i understant that principle, but educate me on what tool use to check for twist.

        my thought had been to make two identical blocks that would set on each side of the bed - milled so that they set on the diagonal faces of the bed, and then place a level across the blocks, perpendicular to the length of the bed. i know it wouldn't be practical to use a level directly on the top of the bed, because the top is not the critical point - the diagonal edges of the bed are what count.

        Comment


        • #5
          That's pretty standard technique covered in texts.... should be fine. just don't let the piece ride on the "ridge" often found on some brands of machine after some wear.
          1601

          Keep eye on ball.
          Hashim Khan

          Comment


          • #6
            Originally posted by J Tiers
            That's pretty standard technique covered in texts.... should be fine. just don't let the piece ride on the "ridge" often found on some brands of machine after some wear.
            i had menat to quote the post above mine first. my question is how do i do it without a level, as suggested by the above poster?

            Comment


            • #7
              Originally posted by lost_cause
              i understant that principle, but educate me on what tool use to check for twist.

              my thought had been to make two identical blocks that would set on each side of the bed - milled so that they set on the diagonal faces of the bed, and then place a level across the blocks, perpendicular to the length of the bed. i know it wouldn't be practical to use a level directly on the top of the bed, because the top is not the critical point - the diagonal edges of the bed are what count.

              I'll have to give this some thought and get back to you. The more I consider it the less simple it seems, in the meantime one of those 3 dollar levels might be the way to go.
              "...do you not think you have enough machines?"

              Comment


              • #8
                Originally posted by lost_cause
                my thought had been to make two identical blocks that would set on each side of the bed - milled so that they set on the diagonal faces of the bed, and then place a level across the blocks, perpendicular to the length of the bed. i know it wouldn't be practical to use a level directly on the top of the bed, because the top is not the critical point - the diagonal edges of the bed are what count.
                While you're technically correct the datum from manufacturing should also be preserved in the front and back flats. In practice you're often better with the flats since they usually show less wear than the V ways, and on the flats it's easier to bridge localized wear with a heavy ground parallel.

                As J Tiers said you're better off with a .003"/ft level than a .0005"/yard level - with the latter you can read the oil film thickness and you'll spend hours chasing the bubble. You don't need the higher level of precision unless you're trying to rescrape or quantify error on the lathe bed.

                If the lathe bed is work be careful to try and bridge the local wear when reading the level, and try and keep one end on a unworn section of the bed. It's hard to tell sometimes but if/when you get a screwy reading try moving a little bit one way or the other on the bed and see if the readings change.

                And to the comment that you don't need the bed 'level' - it's correct but missing the point: a level is the cheapest way to be sure that the bed lies in a flat plane. That being said a lathe with an oil sump is going to want to be grossly level and not mounted on the wall.

                Comment


                • #9
                  Not the ship issue again. The issue is to use a level to check for twist in the bed not to level the bed. The bed can be at 30 degrees, just as long as both ends are at 30 degrees, + or - 0.0004" per foot

                  Phil

                  Originally posted by Davidhcnc
                  Here is my take.

                  Your lathe needs to be straight with no twist. Ideally all the feet equally loaded.

                  It has no need to be level that I have discovered yet.....sometimes lathes are on ships.

                  Comment


                  • #10
                    Originally posted by Davidhcnc
                    I'll have to give this some thought and get back to you. The more I consider it the less simple it seems, in the meantime one of those 3 dollar levels might be the way to go.

                    I think this works

                    1. put a long bar (full lathe length) between centers and make a small cut at each end, measure diameters and adjust your tail to ensure it is in line. Repeat the cut to validate. That is the same as not cutting tapers.

                    2. put a very short bar between centers and take a small cut at each end ( same dial setting )

                    3. measure your results, the difference, if any, will show you the twist and direction of the twist.


                    Note bed wear is not considered although I would expect bed wear to be in the downward direction therefore having lesser effect on the test result.
                    "...do you not think you have enough machines?"

                    Comment


                    • #11
                      Buy a 6" or 8" and put it on the cross-slide and check both ends of the bed by running the carriage along the bed. Do it with the level orientated both longitudinal and transverse. After all it is about how the carriage moves not about the bed as such. This is how manufacturers will do the final tests on a new lathe after assembly.

                      Phil

                      PS: the level doesn't even need to read zero, just as long as you have the same reading at both ends. Shim the level to get the reading on the scale if necessary.

                      Originally posted by lost_cause
                      i understant that principle, but educate me on what tool use to check for twist.

                      my thought had been to make two identical blocks that would set on each side of the bed - milled so that they set on the diagonal faces of the bed, and then place a level across the blocks, perpendicular to the length of the bed. i know it wouldn't be practical to use a level directly on the top of the bed, because the top is not the critical point - the diagonal edges of the bed are what count.
                      Last edited by philbur; 12-15-2011, 09:39 AM.

                      Comment


                      • #12
                        Here is an example of the Starrett Master Precision Level #199. They can be found for much less than what's quoted on this page. That level is a sweet one to have and makes setting up a lathe a breeze in most cases:

                        http://www.mytoolstore.com/starrett/199.html

                        There's also a 6" version (Machinists level) for smaller work. Hunt around eBay or craigslist and you'll probably find one for reasonable cost.
                        "To invent you need a good imagination and a pile of junk" Thomas Edison

                        Better to have tools you don't need than to need tools you don't have

                        73's KB3BFR

                        Comment


                        • #13
                          Also remember when moving your level around try not to touch the vial or the surrounding area as the warmth from your fingers and the static build up (if you rub your finger across the vial or near by arrea) will pull the bubble and result in an erroneous reading.

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

                          Comment


                          • #14
                            plenty of good suggestions on leveling. but back to the initial question...

                            if i am looking for a level, how do i know if it is a suitable machinist's level. so far the best suggestion is to look specifically for the starrett 98, since it can be bought for under $90 new, a used one would be even cheaper.

                            while that #199 may be a great tool, it's out of the range i'd even consider at this point in time.

                            are there any tell-tales, or other characteristics i can look for to determine whether a level is what i want? short of it specifically telling the accuracy on the level?

                            Comment


                            • #15
                              Originally posted by lost_cause
                              are there any tell-tales, or other characteristics i can look for to determine whether a level is what i want? short of it specifically telling the accuracy on the level?
                              If you don't have your hands on the level all you can depend on for accuracy is the manufacturer's statements. If you have one you can get it level then shim up one end for a rough number, but really all you need is a rough number. Pretty much anything considered a 'machinist's level' in an 8" to 12" would be fine, even a 5" would be OK. Just requalify it when you get it (zero against a rest on one side, swap ends and make sure it's still zero. If not clean like heck and try again, make sure you have good contact and if everything else looks good play with the adjuster. If there is no adjuster you bought the wrong level.)

                              I'd also suggest looking for a level with a flat base and not one with a 'v', in my experience the V interferes with placing the level on a narrow lathe way. When you do use the lathe way be sure to run a good stone over the way first and take off the high spots. Just 'float' it on some light oil and you'll feel it catch the high spots, knock them down and then run the level.

                              *After* the late is reasonably level and bed untwisted you can take some tests involving cutting. Doing those before knowing the bed straightness just means that you'll twist things further in the process of trying to make the tool follow a straight path on the twisted bed.
                              Last edited by rkepler; 12-15-2011, 10:22 AM.

                              Comment

                              Working...
                              X