Announcement

Collapse
No announcement yet.

Small mill project (SMP)

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

  • Getting closer and closer...

    Comment


    • Thanks for the update Sam....it's looking great. The (almost) end is near. They're never completely done are they!
      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


      • Hey Sam, I have that same probe, also from auction, and experienced the same issues you are seeing. On the LCNC forum there are a couple threads about renishaw probes and their calibration/setup. The problems you are seeing are common with ALL probe setups and handled by calibration software from the probe manufacturers. Software is the answer ! Here is a link to one of the threads on the LCNC forum on the subject.

        https://forum.linuxcnc.org/40-subrou...-renishaw-mp12

        I have done more reading on the subject but not got around to doing any experimenting. Keep the reports coming.

        Comment


        • Sure - I could calibrate the probe using tool offset. I don't want to. I don't need it to be perfect - if it is within .001ish when rotated 180 deg I would be happy. The problem I see currently with it is the tool I have the probe in isn't coaxial with the spindle. - so - if I put a longer stylus in the probe - it will be way off again. So - if I get the probe lined up with the spindle as good as I can - I think then I don't have to worry about 'calibrating' it.

          That is how the probe in the K&T is setup. It is just aligned to the spindle and has been great ever since..

          sam

          Originally posted by Sparky_NY View Post
          Hey Sam, I have that same probe, also from auction, and experienced the same issues you are seeing. On the LCNC forum there are a couple threads about renishaw probes and their calibration/setup. The problems you are seeing are common with ALL probe setups and handled by calibration software from the probe manufacturers. Software is the answer ! Here is a link to one of the threads on the LCNC forum on the subject.

          https://forum.linuxcnc.org/40-subrou...-renishaw-mp12

          I have done more reading on the subject but not got around to doing any experimenting. Keep the reports coming.

          Comment


          • Awsome work on that machine, Sam.

            The problem with the probes is that the force required to trip them depends on the direction of the probing. You can get trilobular errors way high،er than thou you are talking. Where is the center of the ball does not really matter that much. And if the change the stylus, the amount of error will change too

            There is a couple of ways to solve this: orient the spindle so you always face the probing with the same side of the probe and calculate the lobular diameter for that side, or calibrate the probe in software on each direction of probing, locking the spindle in the same position always. The calibration will also compensate the centering error of the stylus. The Sparky_NY post is very usefull on that matter.

            I have a couple of Renishaw probes (MP10 and MP12, old retired from my work) and a homemade TTS Wireless one, as the Renishaws are too big for my mill. The only reason I don't use them yet is that I'm still installing the spindle encoder to get the spindle orient.

            Comment


            • Originally posted by skunkworks View Post
              Sure - I could calibrate the probe using tool offset. I don't want to. I don't need it to be perfect - if it is within .001ish when rotated 180 deg I would be happy. The problem I see currently with it is the tool I have the probe in isn't coaxial with the spindle. - so - if I put a longer stylus in the probe - it will be way off again. So - if I get the probe lined up with the spindle as good as I can - I think then I don't have to worry about 'calibrating' it.

              That is how the probe in the K&T is setup. It is just aligned to the spindle and has been great ever since..

              sam
              Read the linked thread Sam.... proper calibration routines compensate for not being coaxial in the process.... also calibrate for trip point variations in different vectors which is necessary for the accuracy a Renishaw is capable of. You will find the thread enlightening.

              Comment


              • Originally posted by DEVILHUNTER View Post
                Awsome work on that machine, Sam.

                The problem with the probes is that the force required to trip them depends on the direction of the probing. You can get trilobular errors way high،er than thou you are talking. Where is the center of the ball does not really matter that much. And if the change the stylus, the amount of error will change too

                There is a couple of ways to solve this: orient the spindle so you always face the probing with the same side of the probe and calculate the lobular diameter for that side, or calibrate the probe in software on each direction of probing, locking the spindle in the same position always. The calibration will also compensate the centering error of the stylus. The Sparky_NY post is very usefull on that matter.

                I have a couple of Renishaw probes (MP10 and MP12, old retired from my work) and a homemade TTS Wireless one, as the Renishaws are too big for my mill. The only reason I don't use them yet is that I'm still installing the spindle encoder to get the spindle orient.
                That thread was very educational, unfortunately it just died and the fellow has not reappeared. There is one other fellow that is mentioned in the thread that did similar homework on proper calibration routines. I only wish there was more follow up in the thread. I need to revisit my probe with that new information.

                Comment


                • Originally posted by Sparky_NY View Post
                  Read the linked thread Sam.... proper calibration routines compensate for not being coaxial in the process.... also calibrate for trip point variations in different vectors which is necessary for the accuracy a Renishaw is capable of. You will find the thread enlightening.
                  You mean I just can't gloss over it and assume I know what the thread is about? ;
                  Last edited by skunkworks; 07-02-2019, 02:24 PM.

                  Comment


                  • This probe is adjustable, and allows you to set it to a perfect zero without software.
                    http://www.wildhorse-innovations.com...d&productId=80

                    Comment


                    • Originally posted by skunkworks View Post
                      You mean I just can't gloss over it and assume I know what the thread is about? ;
                      Ha! I read the thread several times, ended up with a headache and very confused.

                      Comment


                      • Originally posted by RB211 View Post
                        This probe is adjustable, and allows you to set it to a perfect zero without software.
                        http://www.wildhorse-innovations.com...d&productId=80

                        That is a common hobby probe, basically a cheap copy of a Renishaw which is the high end industrial version. The same calibration issues exist for both. The wildhorse is a simple 3 screw adjustment to set the tip coaxial, that is all, it does nothing for compensating trip errors for different vector approaches or other variables like the higher end probe calibration routines do. Big differences in the expected accuracy between a $100 probe and one north of $1000 Renishaw probes are used on many of the CMM machines.

                        A friend has one of the wildhorse probes, its pretty decent for a hobby class probe.
                        Last edited by Sparky_NY; 07-02-2019, 02:54 PM.

                        Comment


                        • I had almost made a kinematic probe... Then found a few 'real' ones on ebay. Very happy with the performance. Can't imaging life without a touch probe

                          Comment


                          • One way I want to try to avoid the calibration is to make an electrical contact probe, instead of a mechanical one. This way centering the ball should work, or you can just use tool offsets.

                            Comment


                            • Yah - kinda a tangent.... Maybe

                              Comment


                              • Here we go...

                                Comment

                                Working...
                                X