Announcement

Collapse
No announcement yet.

Machining is a constant learning experience.

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

  • Machining is a constant learning experience.

    For me anyway. I need some more learning I guess.

    I posted last weekend about the trials & tribulations of grinding the ER32 spindle taper on my little CNC lathe.

    After a bit of g-code struggling, I got the grinding process going well but never got the entire surface of the taper smooth & accurate enough to suit me. I think the lathe is accurate & rigid enough but the HF pencil grinder & the 1/8" shaft of the stone were too flimsy. Some areas were smooth with good runout but other areas were rough & out of round.

    This morning I mounted up a homemade boring bar with a Tungaloy TCMT insert tilted a bit for clearance. I set the rpm at 655 (250 sfm) and the feed @ 2 ipm/.003" ipr. I made 2 roughing cuts @ .003" dia and a finish cut @ .001" dia and it came out as smooth as my grandaughter's bootie!



    The runout is a very slight vibration of a .030" travel DTI checked at several spots on the taper. I checked the taper with some engineers blue and a collet and it looks great.

    Now the puzzler. When I mount up a 1/4" dowel and a 1/2" endmill, they both measure around .001" runout. Ahh, you say, the collets are off. I then moved the collet/nut/dowel-endmill assys undisturbed over to the ER32 chuck in my other lathe and they both measure less than .0005" runout.

    I guess I can live with the .001" runout but it's disappointing to do all that work & still have so much runout. Any ideas??
    Milton

    "Accuracy is the sum total of your compensating mistakes."

    "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

  • #2
    Spindle Taper

    I would tilt the insert a little bit more for more of a scraping action and make 6 to 10 passes on the same setting. These are sometimes called "spring passes" because of the flex or spring in all the components. Color the taper first with a blue or red magic marker so you can make sure the entire surface was done. If there is marker left on part of the taper, move over .001 and repeat. Run a slow RPM, like 200 or less and the same feed.

    Comment


    • #3
      One possibly fruitful area or search might be the taper. In my experience, a small misadjustment of the compound, and I'm talking a couple tenths, can show up as that kind of behavior. Collet contact will be just a little light either at the nose or the rear and it will let the stock wander.

      I'm not sure what the best foolproof way is to check the taper. I've been a little suspicious of collets since they can flex. I use a collet with a dowel in it to try to firm it up but I'm not sure that's completely sufficient. A theoretically better solution would be to make a solid plug gauge, particularly when you have the compound already set but that raises the next question of what you check your gauge against. A sine bar and gage blocks should be definitive but that's hard to test, adjust and re-cut as necessary.
      .
      "People will occasionally stumble over the truth, but most of the time they will pick themselves up and carry on" : Winston Churchill

      Comment


      • #4
        A boring bar cuts its own path rather than following a hole so your setup may have been off. In addition, one thou is a lot of runout on a spindle - should be zero to no more than .0001 if you can get it that close.

        Comment


        • #5
          Thanks for the tips ya'll. I'm using a collet with an endmill in it to do the blue transfer check of the taper and it looks OK. I blue checked the same collet on the chuck on the other lathe and it looks about the same but the process does seem a bit vague. I think I'm using too much blue which could give a false reading. I'm going to try some magic marker and see what the ruboff pattern looks like after twisting the collet back & forth a bit.
          Milton

          "Accuracy is the sum total of your compensating mistakes."

          "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

          Comment


          • #6
            Well, that's interesting! I cleaned everything with acetone, marked up the taper with magic marker then twisted the collet back and forth for a while. The 1st 3/16" or so of the taper has the ink wiped off clean & shiny with minor scratches on the rest.

            Either my 8 deg angle g-code/math is wrong, the machine is out of square or most likely there was some springback as the boring tool entered the work. As suggested above I'll run some spring passes. If that doesn't work I'll have to do some trial & error cutting.

            Apparently my blueing transfer juju isn't good enough for this kind of stuff!
            Milton

            "Accuracy is the sum total of your compensating mistakes."

            "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

            Comment


            • #7
              Probably your CNC machines screws have pitch error (usually there is quite much) and your steppers have positional error.
              Amount of experience is in direct proportion to the value of broken equipment.

              Comment


              • #8
                I think it's reasonably close in those areas Jaakko. I spent a lot of time during the setup process making sure that a 1" move command was moving the axes exactly 1" as measured by a dial indicator. It's no Haas but should be good enough for my mickey mouse work.

                I've got the taper close enough now to call it good. I ran a bunch of spring passes and sure enough, a teeny bit of material was removed for the 1st 3 or 4 passes. I rechecked it with the magic marker/rub method and the taper was definitely closer plus the runout measured on a shaft was down from .001" to .0006".

                I then changed the code to increase the angle a teeny bit and recut it, making sure this time to do springback passes. It now prints along the full length of the collet and the runout measured on a gripped shaft is down to .0002". Close enough for me!

                It's amazing how much effect a teeny little a change in angle makes.
                Milton

                "Accuracy is the sum total of your compensating mistakes."

                "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

                Comment


                • #9
                  well in my opnion try again but when you do the finsih pass with the boring bar take a severl extra pass's with no extra adjustments then try and set up again check the run out,

                  when ever i bore holes and i do the same on my lathe when doing out side turnning i take sever fininsh passes and youll be supprised how much extra comes off

                  Comment


                  • #10
                    Spindle Taper

                    Congrats Dickeybird! i knew you could do it!

                    Comment


                    • #11
                      Originally posted by Toolguy
                      Congrats Dickeybird! i knew you could do it!
                      I've been following this thread with interest as I have never done this before and I've enjoyed the ride.

                      Hats off to Toolguy. Nice call with your advice. *Salutes*

                      Comment


                      • #12
                        Originally posted by Toolguy
                        Congrats Dickeybird! i knew you could do it!
                        Thanks for the help T/G, I forget the simple stuff sometimes.

                        And, as my pappy used to say; "Even a blind hog gets an acorn once in a while."
                        Milton

                        "Accuracy is the sum total of your compensating mistakes."

                        "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

                        Comment


                        • #13
                          Originally posted by DATo
                          I've been following this thread with interest as I have never done this before and I've enjoyed the ride.
                          Yup, my threads are usually filled with pearls of wisdom...from other people that know what they're doing!
                          Milton

                          "Accuracy is the sum total of your compensating mistakes."

                          "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

                          Comment


                          • #14
                            Another consideration

                            Originally posted by DICKEYBIRD

                            I then changed the code to increase the angle a teeny bit and recut it, making sure this time to do springback passes. It now prints along the full length of the collet and the runout measured on a gripped shaft is down to .0002". Close enough for me!

                            It's amazing how much effect a teeny little a change in angle makes.
                            Hi Everyone,

                            I know I’m late to this topic, but I hope that someone is still watching

                            As it was explained to me 40+ years ago, never try to rely on calculations/measurements/setup for tapers that are fussy. This is because tiny changes in tool position above or below center make the taper straighter than intended.

                            This can be easily visualized by picturing what would happen if you tried to turn a taper on the OD of a 1 inch shaft if the tool is ½ inch below center. The cutting edge would move across the bottom surface of the shaft & there would be no cutting at all.

                            Now I realize that this example is an exaggeration w/limitations, but it is only to picture the effect of moving the cutting edge above or below center.

                            I was taught to get as close as you could with your setup, bore or turn to within 15 or 20 thousandths and tweak the compound or taper attachment through tiny amounts of trial & error until your taper fit w/bluing to whatever you were using as a gauge.

                            This also seems to be what TGTool has found, "One possibly fruitful area or search might be the taper. In my experience, a small misadjustment of the compound, and I'm talking a couple tenths, can show up as that kind of behavior."

                            The ‘correction’ that DICKEYBIRD had to apply to the G code corresponds exactly to what we always had to do manually in the shop when turning or boring tapers.

                            Just passing this on because this applies to manual as well as CNC machining.

                            Hope this is of some help.
                            Last edited by jhe.1973; 09-06-2011, 03:24 AM.
                            Best wishes to ya’ll.

                            Sincerely,

                            Jim

                            "To invent you need a good imagination and a pile of junk" - Thomas Edison

                            "I've always wanted to get a job as a procrastinator but I keep putting off going out to find one so I guess I'll never realize my life's dream. Frustrating!" - Me

                            Comment


                            • #15
                              Originally posted by jhe.1973
                              changes in tool position above or below center make the taper straighter than intended.
                              Thanks Jim, there's another one of those "pearls" mentioned above. I knew tool height made a difference cutting tapers but never thought much about the actual effect. I can remember "makes the taper straighter" easily. Thanks, that's my kinda engineer-speak.

                              I went to a lot of trouble ensuring the tool was on center but since all metal moves around like a noodle no matter how rigid we think it is, the effective (or dynamic to use another term) tool height is what counts.

                              The other thing that concerned me using carbide was the sharpness of the insert and depth of cut having another effect. I chose a Tungaloy finishing insert for its keen edge and it did pretty well. It made little clumps of fine steel wool on the springback passes.
                              Milton

                              "Accuracy is the sum total of your compensating mistakes."

                              "The thing I hate about an argument is that it always interrupts a discussion." G. K. Chesterton

                              Comment

                              Working...
                              X