Announcement

Collapse
No announcement yet.

how to precisley align 2 perpendicular linear stages

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

  • how to precisley align 2 perpendicular linear stages

    I'm working on this grinding fixture, its nearly done & for the life of me I cant figure out of the precisely align (or measure) one stage perpendicular to the other.

    the stages: note the post on the left stage
    Click image for larger version

Name:	20220419_033220 (Custom).jpg
Views:	495
Size:	1.48 MB
ID:	1997368


    they need to sit precisely perpendicular, something like this:
    Click image for larger version

Name:	20220419_033238 (Custom).jpg
Views:	479
Size:	1.46 MB
ID:	1997369


    right now its free to pivot on that front pin:

    Click image for larger version

Name:	20220419_033243 (Custom).jpg
Views:	480
Size:	1.34 MB
ID:	1997370

    I had hoped to set up on a surface plate, or something with a dial...establish / prove it perpendicular then clamp it, drill another pin & mounting holes. Its pretty small, long table is 4" x 2", short one is 3x2. All edges where milled square & there are two pin holes in each table, drilled @ same time as milling dovetails, that establish a reference axis...if thats useful. **

    Its kinda rare that I'm completely stumped like this, am I over thinking this or is it not quite as simple as I thought?

    I don't have a ton of measuring kit for the surface plate, but I have some stuff. Let me know how this needs to be set up & i'll see what I got.

    thanks!


    **I'm really looking for more "surface plate" accuracy than "milled" accuracy. To quantify it, I'd like to be measuring it in ten-thousandths of an inch.
    "it is no measure of mental health to be well adjusted to a profoundly sick society." -- krishnamurti
    "look deep into nature, and then you will understand everything better." -- albert einstien
    "any intelligent fool can make things bigger and more complex...It takes a touch of genius - and a lot of courage to move in the opposite direction."

  • #2
    You need a perpendicular reference surface. A granite square comes to mind.

    Comment


    • #3
      I see a few poor souls applying these techniques to their milling machines and entering deep depressing episodes.

      Comment


      • #4
        Small rotary table, fix one, plonk 2nd on top clock, rotate 90, the axis of one is then perpendicular
        mark

        Comment


        • #5
          Originally posted by boslab View Post
          Small rotary table, fix one, plonk 2nd on top clock, rotate 90, the axis of one is then perpendicular
          mark
          ........mmmmmmmmmmmmmm No.

          -D
          DZER

          Comment


          • #6
            Originally posted by metalmagpie View Post
            You need a perpendicular reference surface. A granite square comes to mind.
            I got a Chinese granite square, the quality was so-so but the test sheet did agree with the checks I made on it 1.5um over 150mm so pretty good.

            mtraven do you have a surface grinder?
            Peter - novice home machinist, modern motorcycle enthusiast.

            Denford Viceroy 280 Synchro (11 x 24)
            Herbert 0V adapted to R8 by 'Sir John'.
            Monarch 10EE 1942

            Comment


            • #7
              I'm not seeing the challenge so might be missing something. Put the near edge of the bottom piece on the plate, so its like a book standing up on its spline. clamp to something heavy, angle plate, tool maker block etc so its vertical and doesn't fall over. If the dataum surfaces are indented or its a dovetail, build up a stack blocks and/or pins in the dovetail so that datum surface is parallel to the plate. Put the second piece where it goes and get it square, easiest is with a precision square, something good to a tenth of 6".

              Other than the plate, the only other really high precision item needed is a square which is kind of a necessary bit of kit for precision plate work.
              in Toronto Ontario - where are you?

              Comment


              • #8
                I'd set the bottom one up on the surface plate like Mcgyver described, then stand the other into position and loosely bolt it to the first. Get it set close with a square, then snug the bolts just enough so that it holds its position. Check the vertical part to make sure its outside surfaces are parallel. Take a surface gage or squaring fixture and mount a test indicator, then butt the surface gage squarely against the edge of your part. While holding the surface gage in good registration with the part, set the indicator to register against the top of the same edge and zero it. Make sure it repeats when you remove the surface gage and put it back. Then check the other side. If the readings on the indicator aren't the same on both sides, adjust the part until they are. You're done. If the part was not parallel on the step that you measured it, you will need to determine if one of the sides is parallel to what you want to square up or the other, or whether both are slightly out, etc. before you decide how to remedy that situation. Hopefully you have it nice and straight since I see the surfaces are scraped.

                Comment


                • #9
                  Um no doesn’t help so try somthing else, how about
                  https://www.vermontphotonics.com/VT_...plications.pdf
                  um no?
                  Um yes
                  mark

                  Comment


                  • #10
                    If I'm reading this right Mcgyver and EKretz are assuming that the lower table itself is in fact square. And I'm sure it is in fact as square as the tool it was made with. But if he's looking to use this on a surface grinder he might be looking for that next higher confirmed degree of squareness confirmation.

                    To be fair and in line with this idea it would be easy to make the two slides a match to the squareness of your milling machine. Just set up one so it's aligned with one axis of travel using a DTI or dial gauge held in the spindle. Sweep the table and line up the base. Then lay the second slideway on the first and zero the runout to the other axis of travel. It'll be as square as the mill table travel axes.

                    But is that good enough for you? What about a way to test your mill for square before you use it that way with confidence?

                    Using the idea of a 123 block as a square reference and checking with a squareness comparator I came up with an idea for checking your mill for squareness between the two axes. And in the end you end up with a handy tool to check your shop squares or for other uses.

                    Using a piece of plate of a suitable size drill holes in the middle area that permit you to clamp it down over spacers so you can mill all four edges in one go. If you trust your mill to produce straight edges then just mill all four edges to form a rectangle or square. If you're not sure it makes truly straight edges then relieve the middles and make something like the sketch below.

                    Now the testing and any fine tuning starts. First off confirm that the width between any two sides is the same at both ends to within a fine degree. If it isn't then pick one side you want to make truly square and figure on the other side being plus or minus by the width difference. Now stand this plate up on your surface plate so one end is seated against the plate and use a squareness comparator to compare the "lean" of the two sides between the base and the top of the plate. If all is good both vertical edges will have the same lean from base to top as indicated by identical readings of the squareness comparator. Or the off side will be plus or minus by the error in parallel.

                    More likely the readings will be out by some small amount indicating that the base isn't square to the primary vertical edge. This is due to some error in squareness between your two axes of travel.

                    But wait, if you bolt the plate back down on the spacers and use an indicator to set up the long edge but with a suitable offset from one end to the other you can re-cut the base edge and re-test it with the squareness comparator. It shouldn't take long to end up with a truly confirmed square rectangle (or a square primary edge).

                    Now you can use that to set up your tables. And if it's the right sort of size or we make a new one it can be handy for double checking our general use shop squares.

                    Don't have a squareness comparator? There's lots of ways to fudge something to make one. In a pinch you could even do up a curved edge bumper plate out of steel sheet and stick it to a mag base. It's not great since those arms are often a bit flexible but with some care it can do the job just fine for occasional use. Or perhaps it's time to cobble one together. They can be pretty fancy or they can be dead simple.

                    If you like this I'd suggest using something like 1/4" thick plate so there's a reasonable edge thickness to work with. Trying to use a squareness comparator on something like 1/8 thick edges can be rather fussy. Doable for sure but still fussy. And with any of these in fairly thin materials it would be nice to have some sort of base to hold it upright without too much of a lean.

                    Click image for larger version  Name:	Shop Made Square.jpg Views:	2 Size:	176.1 KB ID:	1997401
                    Last edited by BCRider; 04-19-2022, 03:33 PM.
                    Chilliwack BC, Canada

                    Comment


                    • #11
                      It is primarily the movements you need perpendicular. So you need a reference surface on each that is aligned perfectly with the movement of the slide of that part.

                      So, you need to align with that. Gage pins in the dovetail, if it has a dovetail, that sort of thing could provide points of reference. Indicating even from a pair of toolmaker's buttons.... you just need something to indicate from to determine perpendicular. Unless you have intentionally made them so, the sides of a slide may or may not be suitable as a reference. Since there is one way to be parallel to motion, and many ways to be out of parallel, the odds are not good even from milling using the mill table movements, unless you make them better intentionally.

                      I think I would prefer to not try to precisely mount it, since that has a reasonable chance of not going to plan. Instead, I would do my best to align it, and also provide some means of adjustment. I would then, after mounting, compare the movement to the best square I have, and adjust accordingly.

                      If you have precise grinding capability, and standards, you can grind the parts to fit and be as precise as you wish, by grinding alignment lips, or other references. You may be able to make it assemble to your desired limits the first time. I do not have that, so I tend to think of other methods.

                      Adjustment can be by some form of screw adjustment. A better provision would be by scraping (or grinding, lapping, etc) a locating surface on each part, based off the results of checking, for locating them positively, butting the ground or scraped flat surfaces together. That will allow disassembly in future, while still allowing precise reassembly.

                      No matter the means, you still need a good square reference. Then you can mount indicators to verify the movements against the square.
                      CNC machines only go through the motions.

                      Ideas expressed may be mine, or from anyone else in the universe.
                      Not responsible for clerical errors. Or those made by lay people either.
                      Number formats and units may be chosen at random depending on what day it is.
                      I reserve the right to use a number system with any integer base without prior notice.
                      Generalizations are understood to be "often" true, but not true in every case.

                      Comment


                      • #12
                        Originally posted by BCRider View Post
                        If I'm reading this right Mcgyver and EKretz are assuming that the lower table itself is in fact square. ]
                        No, you simply make sure that for the two surfaces you want square to each other (the edges, the box way, the dovetail way, whatever), one is supported off the plate (directly or indirectly, but parallel), and the other is in contact with the square (which is also in contact with the plate)

                        easy peasy
                        Last edited by Mcgyver; 04-19-2022, 03:57 PM.
                        in Toronto Ontario - where are you?

                        Comment


                        • #13
                          Originally posted by Mcgyver View Post

                          No, you simply make sure that for the two surfaces you want square to each other (the edges, the box way, the dovetail way, whatever), one is supported off the plate (directly or indirectly, but parallel), and the other is in contact with the square (which is also in contact with the plate)

                          easy peasy
                          Yes it is easy peasy.... provided the square is in fact square within the allowance he's willing to live with. We're still back to using some manner of square. And that raises the question of how square the square really is. And some way to test the square or mill or other X-Y tables or some other shop gear for being truly square.

                          I just got the impression that he doesn't want to just trust his squares but is looking for a more fundamental basic principles method. Or at least some way to confirm one of his other options before trusting them. And so that leads us here.... Perhaps I misread something?

                          Chilliwack BC, Canada

                          Comment


                          • #14
                            All the precision tools we enjoy now can be traced back to clever engineers who worked out to to produce precision with what was available to them.

                            At times like this one should ask "What would Whitworth have done?"

                            Comment


                            • #15
                              Originally posted by BCRider View Post
                              Yes it is easy peasy.... provided the square is in fact square within the allowance he's willing to live with. We're still back to using some manner of square.
                              Its will always come back to that, it is a fundamental requirement. He can make or buy one but achieving squareness of assemblies is comparative process, work is compared to reference. Imo if you are scraping, working with surface plates and worried about creating accurate mechanisms have such a square, that is good enough for your standards, is a prerequisite, as I say, made or bought.

                              If he doesn't trust his, get a better one, or grind or scrape a box square (not hard to do to a 10th over 6") and check and fix his square before proceeding with this project
                              Last edited by Mcgyver; 04-19-2022, 05:25 PM.
                              in Toronto Ontario - where are you?

                              Comment

                              Working...
                              X