Announcement

Collapse
No announcement yet.

Acceptable backlash ?

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

  • Acceptable backlash ?

    What do you people think acceptable backlash on machine tool feed screws would be ? I have a near new Super seven, the cross slide has a total 5 thou of slack an ML10 the same and an ancient Mikron mill (even though old little used and loved) with 7 thou play in all screws. All are adjusted correctly ie the play is in the thread, not the shaft bearing.
    Would this be about the norm and within limits ? David

  • #2
    Seven thou. is good. If you get it to close it may be tight in some places but being new you could take the back lash out.
    It's only ink and paper

    Comment


    • #3
      First of all on plain lead screws and nuts you can't have zero backlash. They have to have some clearance to operate. The minmum clearance is about 0.003 for most purposes. Many machine tool have adjustable backlash features allowing 0.001" or even less. As the screw wears it gets smaller in PD usually near the center where most of the wear takes place. This limits the adjustment if the screw is to be used over its full range.

      0.005" B/L aint bad. Neither is 0.007". OK now rejoice and don't fix it. It ant broke. The only way you get to zero backlash is in pre-loaded ball screws and they aren't trouble free either.

      Comment


      • #4
        Originally posted by Carld
        Seven thou. is good. If you get it to close it may be tight in some places but being new you could take the back lash out.
        But would a machine such as a Swiss Mikron come out of the factory with 7 thou backlash ? They were hand built and would think VERY expensive. In saying this probably so as I have had this machine from near new and it has had little and careful use, and always well maintained. David

        Comment


        • #5
          David: You do know that backlash isent the end of the world right? Just because theres 7thou backlash does not mean half your cuts are gonna be 7 thou off, it means you need to compensate for 7thou backlash.

          Now what would be a really good useful messurement of wear, is how much the backlash varys over several sampled spots along the travel, Inconsistant backlash is slightly more annoying to compensate for, and indicates wear has occured.

          As for compensateing, one of the simplest ways is to make sure you allways approch cuts from the same direction. Ie want to make a block 1" wide exactly?

          move table to right till 0, cut, move table 10 turns right, cut. If you accidently pass the 10th turn, you must move the table to the left *past* the 10th turn by at least the backlash amount, idealy much more to be safe, and then approch the 10th turn '0' mark by going right again.

          Alternately, you can messure the backlash at that location (or just use your avg backlash if a few thou off is fine), and approch from the left but adding the backlash amount to all dial messurements, But this tends to be less accurate.

          Or you can just buy a DRO that reads the actual table travel, and the backlash becomes much more a non issue.

          Basicly, unless your making a CNC, Backlash isent a big deal and should not limit your accuracy. Even if you are making a CNC there are ways to compensate for backlash... And the only way to get rid of backlash, aka preloaded ballscrews, transfer motion from the cutting force back into the handles, meaning you can't easily use them for manual mills (You have to lock all axises you arnt holding onto!)
          Play Brutal Nature, Black Moons free to play highly realistic voxel sandbox game.

          Comment


          • #6
            [QUOTE=Black_Moons]David: You do know that backlash isent the end of the world right? Just because theres 7thou backlash does not mean half your cuts are gonna be 7 thou off, it means you need to compensate for 7thou backlash.

            Now what would be a really good useful messurement of wear, is how much the backlash varys over several sampled spots along the travel, Inconsistant backlash is slightly more annoying to compensate for, and indicates wear has occured.

            As for compensateing, one of the simplest ways is to make sure you allways approch cuts from the same direction. Ie want to make a block 1" wide exactly?

            move table to right till 0, cut, move table 10 turns right, cut. If you accidently pass the 10th turn, you must move the table to the left *past* the 10th turn by at least the backlash amount, idealy much more to be safe, and then approch the 10th turn '0' mark by going right again.

            Alternately, you can messure the backlash at that location (or just use your avg backlash if a few thou off is fine), and approch from the left but adding the backlash amount to all dial messurements, But this tends to be less accurate.

            Or you can just buy a DRO that reads the actual table travel, and the backlash becomes much more a non issue.

            Basicly, unless your making a CNC, Backlash isent a big deal and should not limit your accuracy. Even if you are making a CNC there are ways to compensate for backlash... And the only way to get rid of backlash, aka preloaded ballscrews, transfer motion from the cutting force back into the handles, meaning you can't easily use them for manual mills (You have to lock all axises you arnt holding onto!)[/QUOTE

            Just checked it (the Mikron) and the 7 thou is right through its travel, so probably was this from new ? David

            Comment


            • #7
              Likey. also means you now know an easy way to compensate for it. (By just subtracting or adding 0.007" when moving the other way)
              Some nuts are actualy 'split' and have little setscrews to adjust the split to pinch the acme to reduce backlash. but they are not designed at all to eliminate it, as they would induce large wear on the nut and some wear on the leadscrew.
              No backlash = no clearance = high forces.
              Play Brutal Nature, Black Moons free to play highly realistic voxel sandbox game.

              Comment


              • #8
                David your machine may not have had .007 when new. To check, move as close to the tailstock end of the ways as possible. This area should be unworn for all practical purposes. Measure your backlash here to get an idea of the backlash when new. .007 is not much. Apparently you have never had the pleasure of using the kind of machines many of us make living with. .025 is not uncommon in the real world.
                Last edited by tdmidget; 12-07-2009, 02:23 PM.

                Comment


                • #9
                  In my experience, 0.007" is pretty good and, as long as you always take up the backlash in the same direction as explained by BlackMoons, it doesn't matter a hoot. That technique is standard procedure for operating any machine tool with a regular leadscrew.

                  "Too much" backlash is whenever it gets to the point of being annoying. For me, that tends to be around 0.025" to 0.030" -- the amount you have to crank to back up before you can go forward starts to bug me. In terms of accuracy though, as long as you always take up the backlash in the same direction the amount doesn't matter. After you do it a while, it becomes automatic and you won't even think about it (much, anyway).

                  Compensating for backlash by adding a fudge factor allegedly equal to the backlash is IMO more trouble than it's worth and extremely subject to error depending on how the leadscrew is worn and where. If you don't want to deal with backlash, install a DRO that reads actual table position.
                  ----------
                  Try to make a living, not a killing. -- Utah Phillips
                  Don't believe everything you know. -- Bumper sticker
                  Everybody is ignorant, only on different subjects. -- Will Rogers
                  There are lots of people who mistake their imagination for their memory. - Josh Billings
                  Law of Logical Argument - Anything is possible if you don't know what you are talking about.
                  Don't own anything you have to feed or paint. - Hood River Blackie

                  Comment


                  • #10
                    Backlash of 0.007" and 0.005" is good.
                    Taking the backlash out was one of the first things we were taught many years ago, and it's such a habit now on any machine tool that it's almost an instinctive/automatic action.

                    Peter

                    Comment


                    • #11
                      Re: Backlash

                      David...The only time backlash is an issue is when you are changing directions of travel with the table. It is simple enough to lock down the table, turn the slack out of the handle, and reset the dial. Being a guy who learned on old, well worn machines, this was what was taught. I always used a travel indicator when doing this so that I could be sure that I always had the same result. Now, the mill I have has travadials installed on X and Y so it is not as much of a worry. (yes, I calibrate the travadials as they tell you in their literature)
                      Jim (KB4IVH)

                      Only fools abuse their tools.

                      Comment


                      • #12
                        I've been able with care to get b/l in my leadscrews down to about .003. But then it loosens up fairly quickly. I'd have to say .007 is totally acceptable, and .005 is pretty good.
                        I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

                        Comment


                        • #13
                          Backlash

                          A quick check of backlash in the machines around here:
                          knee mill y = 0.025"
                          knee mill x = 0.060"
                          horizontal mill table = 0.035"
                          horizontal mill saddle = 0.095" (that's 95% of a full turn! )
                          lathe cross = 0.015"
                          lathe compound = 0.005"

                          Knee mill is a 1984 Sharp/Alliant
                          Horizontal is a 1953 B&S #2
                          Lathe is a 1983 Clochester 13" 8000

                          Everything except the knee mill Z (knee & quill), and the lathe compound have DRO's.

                          I'd say that 0.005 to 0.007 are quite good. As has been stated, when approaching your measurement, always approach from the same direction. This takes backlash out of the equation. IE, always come to your setting turning the handle clockwise (for example). If you overshoot, wind the handle the other way far enough to take up any possible backlash, then approach your setting again.
                          Last edited by fasto; 12-07-2009, 07:12 PM.

                          Comment


                          • #14
                            fasto: Woah how on earth can it have a full turn backlash.. doesnt that mean theres uh, no acme thread left? Id hope that thing is geared!
                            Play Brutal Nature, Black Moons free to play highly realistic voxel sandbox game.

                            Comment


                            • #15
                              Originally posted by Black_Moons
                              fasto: Woah how on earth can it have a full turn backlash.. doesnt that mean theres uh, no acme thread left? Id hope that thing is geared!
                              It's not geared...
                              The screw is worn, the nut is worn, the thrust bearing is worn, the nut attachment to the saddle is a little loose, these things all add up. I plan to rework all these parts over the winter.

                              Comment

                              Working...
                              X