Announcement

Collapse
No announcement yet.

Z axis opto probe for a CNC lathe.

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

  • Z axis opto probe for a CNC lathe.

    I’m finally feeling much better after getting my rotten gall bladder jerked out and made it out to the shop this morning to finish up this project that was started before getting sick.

    It’s a Z-axis probe that’s semi permanently mounted to the back side of the carriage of my CNC lathe. It has a little opto sensor/board I got off ebay from Holland. A spring loaded plunger with a flat dial indicator tip on it, some aluminum scraps & acetal bushing is all it needed to work very well. It’s probably bigger than needed but you never know what widget may come along down the road that needs a place to live.

    I previously had a Z-home sensor mounted near the tailstock but never used it. Since most all my work is modifying or repairing existing parts that don’t get faced off, I’m always having to touch off the tool on the face or a feature of the part. I’ve used all the methods including a USB microscope to set Z-zero but it’s a real pain to accurately do this on 15 or 20 parts in a row. So, I decided to build this device. It works amazingly well and is dead-nuts repeatable & accurate as I can measure with a tenths D.T.I. I use the “Set Home Z” button on Tool 0 in the standard Mach3 Turn screenset. I’m in the process now of setting up all of my tools’ Z-axis offsets in the tooltable. Once that’s done, I should be able to home the Z to a feature of the part, slap on which ever tool is needed and cut away with confidence that it’ll come out on the money.







    Here’s a pic of a typical setup

    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
    Very interesting. Was just thinking about a very similar mechanism to dust/chip proof optointerrupters for homing switches on my mill. A round shaft with a hefty spring preload, and a few o-rings as wipers.

    What aperture size is the optointerrupter you're using there? I only really have 1mm aperture ones free available to me but I could probably dig up something smaller. However the way I see it, there will be a fixed gating point regardless of the aperture size, at which point the logic triggers. Whether that's 99% of a 1mm aperture or 99% of a 0.1mm aperture, the accuracy is probably going to be similar I figure as more than likely the electro components will be identical across the range, regardless of aperture size, and will therefore have similar cutoff thresholds.

    Either way great work, looks like a massive timesaver.

    Comment


    • #3
      Thanks S/G. By aperture size were you asking about the width of the sensor gap? If so it's 3/16" or a bit less than 5mm. The diameter of my probe is 1/8". I was going to use a keyed shaft and a flat blade on the end of the probe but tests showed that the round shaft is more than accurate enough so the extra work & friction of a key wasn't needed.

      I have another one of the little Dutch opto sensor/boards on the X-axis and it has worked flawlessly for over a year. It's triggered by a slotted disc on the stepper motor (and a mechanical switch on the carriage) to gain high homing accuracy similar to the indexing pulse signal on servo systems. After seeing how repeatable & accurate the sensor is by itself, I'm not sure the extra work was called for.

      Timesaver? Yes, after I get all the tool Z-axis offsets loaded. Right now I'm burning up shop time trying to get accurate results which on my little ORAC lathe requires measuring/recording the offsets, making a test cut, then recording the tweaks needed to compensate for flex. My lathe has decent ballscrews but still moves around a little bit under cutting loads.

      The other time-eater is fixing the tool height-setting knurled nuts & locknuts on the cheapo QCTP toolholders I bought a while back. I noticed wild variances in trying to set the height & finally traced it to the the tapped holes in both nuts being crooked as a dog's leg!! Fortunately there was enough metal there to straighten them out with some lathe work.
      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


      • #4
        By aperture I mean not the width of the slot between opto sender and detector, but rather, the width of the slot through which the light passes in the optointerruptor assembly. i.e. the width of the slot that covers the sender.

        Comment


        • #5
          As you can tell, I'm not much of an electronics expert. There is no p/n on the sensor itself (that I can see) to get specs online and the seller didn't list one in the ebay ad but here's a closeup. The gap shown on the "E" & "S" is 2mm wide & 3mm high. Is that what you're looking for?



          I've set up quite a few tools with it now & am very happy with the way the thing works. It's VERY accurate, quick & stupid-simple to use. It's so nice when an idea actually works!
          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
            Glad to hear you are feeling better. Shop work, in moderation, is very therapeutic.

            That looks like a real nice idea. You say it is as accurate as a tenths DI. Is that just short term or does it hold for longer periods, like overnight or several days?
            Paul A.

            Make it fit.
            You can't win and there is a penalty for trying!

            Comment


            • #7
              Thanks Paul, I think I've had problems with the effects of the putrid gallbladder for quite a while...possibly 6 months. I haven't felt right and my gumption was all used up just doing my day job. Hobby stuff was a chore and my ability to focus & work through simple design problems was very weak.

              I have a couple days off from work and even though I stayed up late last night watching a movie with SWMBO, I was up & at 'em at 6:00AM, downed some breakfast and was out in the shop doing some CAD work to make some widgets for my buddy's company by 7:00. I puttered around on & off all day & finished making them around 4:00 & now am looking around for the next project. It feels good!

              The opto thing is indeed very repeatable. I didn't try to apply the full scientific method but I did leave the D.T.I. hooked up overnight and rechecked things through 4 or 5 machine shutdowns. It's very repeatable for a machine that's not a zillion dollar pro rig. Obviously it's not perfect but if I come at the zero spot from the same direction, it will stop with the needle dead on the mark most times with an occasional half needle width off either way after a shut down/restart.

              The widgets I made today had features on either end and it was great to be able to swap the part in the collet, zero the probe to a face, throw on the appropriate tool and know it was going to be spot on with no touching off guesswork, cig papers or whatever.

              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


              • #8
                Interesting and thanks for sharing, Dickeybird.

                I'm still using the paper method of touching off my lathes. The problem with putting a home or a touch-off sensor on the carriage is that the toolpost holder can and does get moved around on the carriage. But I suppose a guy could mount the sensor on a QCTP holder?

                Comment


                • #9
                  Originally posted by MTNGUN View Post
                  But I suppose a guy could mount the sensor on a QCTP holder?
                  That's where I started with the idea but worries about cable plug/wear/oil ingress over the long term had me mounting it semi-permanently to the back side of my carriage tool plate. It's mounted with a stud/reamed hole and a close fitting allen screw so it should mount/dismount accurately but I don't plan to move it unless something BIG comes along. Most of my work is small and my QCTP doesn't get moved unless there's a national emergency.
                  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


                  • #10
                    Although you don't seem to be having any issues with the switch there are ways to improve performance. To maximize the repeatability of the optical switch do as much as possible to prevent any other light source from shining into it. Also ensure that the emitter side power supply is very well regulated. When set up that way it will be extremely repeatable.

                    One thing that isn't at all obvious is what appears to be black plastic of the opto switch assembly may not be opaque at IR wavelengths. The simple solution to that possibility is to paint the entire opto-switch frame with flat black paint. I use plastic model acrylic paint as it will not damage any of the plastic.

                    Where did you find that collet chuck?
                    Free software for calculating bolt circles and similar: Click Here

                    Comment


                    • #11
                      The cover over the opto housing is (I think) light proof. I'll double check the 5V source but I think it's OK; I use a re-purposed PC power supply for the machine's 5V & 12V power needs. So far so good but thanks for the input!

                      I got the ER-32 chuck from AM Tools.com in California a few years ago for $25.00 but they soon went on galactic back order. Now I can't seem to find their site at all to post a link; bummer.

                      I bored the back center of the chuck out a bit to allow it to move around on the lathe's spindle flange seat then drilled & tapped 4, 6-32 holes for 4 homemade brass-tipped screws. That allows me to adjust the chuck/workpiece around to get it dead-nuts concentric & final-tighten the flange nuts. Poor man's Set-Tru collet chuck. Us tight wads, er, frugal engineers are always thinking.
                      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