Announcement

Collapse
No announcement yet.

Project: Squareness Comparator

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

  • Project: Squareness Comparator

    For some time now I had wanted to make a squareness comparator. You know, the conventional kind with a large radius on the base, which rides on a surface plate and carries an indicator. This was partly my motivation behind lapping in my own cylinder square (I will try to clean up and merge those threads later).

    On question bothered me for months: how to check squareness when only one surface is accessible? Most comparators depend on the principle of reversal: test the backside of the part and note the difference to the front. But what if your part is such that you cannot use the reversal technique? What if part geometry only allows testing *one* plane?

    I concluded that the only way, is to have a disc-shaped base with the indicator held centrally, and be able to vary its height with no tilt in any direction. My project isn't quite that perfect yet but it's getting there.

    Next question was what to use? The only steel pieces I had were way too large. I do have a pair of indicators that mount on the back with a threaded hole. And I have the pistons left over from when I rebuilt my welder -- nice and round already, and just the right radius for my rear-mount indicators! All I had to do was cut off the piston top and flatten it on the plate. drill and tap, and make up a selection of "height rods".

    Pics:

    Click image for larger version

Name:	P3040001.JPG
Views:	376
Size:	86.1 KB
ID:	1859695Click image for larger version

Name:	P3040006.JPG
Views:	380
Size:	81.8 KB
ID:	1859696Click image for larger version

Name:	P3040005.JPG
Views:	392
Size:	90.0 KB
ID:	1859694
    25 miles north of Buffalo NY, USA

  • #2
    It was nice to be able to use the original center hole from when they made the piston.
    These are the old school split-skirt pistons that are ground round instead of barrel-shaped or cam ground. I miked them and they are indeed round (and worn out).

    Next I had to cut off the piston top. In case anyone is wondering, its out of a 1950 Lincoln welder with a Wisconsin engine. I used a fret saw with a fine blade (50 TPI) and held it against the bottom of the oil ring groove while it spun slowly in the 7x10 lathe:

    Click image for larger version  Name:	P3040007.jpg Views:	0 Size:	108.4 KB ID:	1859698Click image for larger version  Name:	P3040009.JPG Views:	0 Size:	73.4 KB ID:	1859699
    Last edited by nickel-city-fab; 03-05-2020, 02:01 PM.
    25 miles north of Buffalo NY, USA

    Comment


    • #3
      I flattened out the bottom a bit on a piece of glass with sandpaper, then cut a relief in the underside so it only contacts the plate around the outer rim where the original oil groove was. Sorry for the bad pics this time, getting a different camera dialed in:

      Click image for larger version

Name:	P3040011.JPG
Views:	385
Size:	83.2 KB
ID:	1859703Click image for larger version

Name:	P3040013.JPG
Views:	381
Size:	73.5 KB
ID:	1859704
      25 miles north of Buffalo NY, USA

      Comment


      • #4
        Nice project. Tom Lipton built something similar, you may get some ideas from his videos. There are five of them, here is a link to the first.

        In this video we work the lathe work on the 8620 alloy steel base for the traveling squareness comparator. This instrument allows the metrologist to inspect ...

        Comment


        • #5
          Next up, I had to create some "height rods" to put the indicator at various heights above the surface plate. They're pretty simple: the tapped hole in the piston top is 5/16-24, and the tapped hole in the indicator back is 1/4-20. Evidently Federal made them this way as part of an interchangeable mounting system. Eventually I would like to single-point them but this lathe only allows metric threading -- the *ONE* gear you need for Imperial threading is not included nor is it standard. This will have to do for now:


          Click image for larger version  Name:	P3040014.JPG Views:	0 Size:	75.8 KB ID:	1859711Click image for larger version  Name:	P3040017.JPG Views:	0 Size:	72.7 KB ID:	1859712Click image for larger version  Name:	P3040018.JPG Views:	0 Size:	77.1 KB ID:	1859713Click image for larger version  Name:	P3040023.JPG Views:	0 Size:	97.2 KB ID:	1859714
          25 miles north of Buffalo NY, USA

          Comment


          • #6
            Originally posted by pinstripe View Post
            Nice project. Tom Lipton built something similar, you may get some ideas from his videos. There are five of them, here is a link to the first.

            In this video we work the lathe work on the 8620 alloy steel base for the traveling squareness comparator. This instrument allows the metrologist to inspect ...
            Yeah Tom is pretty cool, I've been a subscriber for years. I just drool over that lathe of his...
            25 miles north of Buffalo NY, USA

            Comment


            • #7
              Some shots of it in use:


              Click image for larger version

Name:	P3040025.JPG
Views:	369
Size:	79.9 KB
ID:	1859721Click image for larger version

Name:	P3040027.JPG
Views:	383
Size:	109.1 KB
ID:	1859722

              ...(you would not believe what a PITA it was to get the camera to focus.... decided to keep the washed-out shot because it shows the geometry)
              25 miles north of Buffalo NY, USA

              Comment


              • #8
                You are running on in a direction you may not need to go. The whole point of doing a CS and then having a comparator is that you can calibrate the comparator and then use that as an "indicating square" on other surfaces. So no need for the reversal trick.

                Besides, the reversal trick only works if you rely on TWO basic principles. First is that the item must be the same width between the two faces used for the reversal trick. That's what you did with your piston pin CS in the first phases. You confirmed that it was a constant diameter the whole length and that the center axis was actually straight. If either of those conditions were not true to within a very small degree then the pin would not have been a good CS candidate.

                Tom Lipton brought this up in an older video on using 1-2-3 blocks as reference squares to set a comparator. He used a comparator to check opposing faces while one end sat on the surface plate. But he first measured to ensure that the opposite 2" faces were in fact the same distance at each end. Then he was able to use the block to calibrate a comparator using the reversal trick and setting the gauge to "split the difference". The block didn't need to actually be square but it did need to have two very parallel faces. In this case the reversal principle and splitting the difference was used for calibrating the comparator. And at that point he could use the comparator to check other items for square.

                But again you've lapped and confirmed that your CS is true to within a small source of error and now you can use that to calibrate a comparator so you can then use the comparator for some "trusted" time to measure single angles between two faces for square without the need for any reversal.

                It's a nice idea to re-use old items like the piston. But keep in mind that aluminium will wear fairly rapidly against the surface plate. And in fact it'll "pencil mark" the granite pretty badly with aluminium. I'm also thinking that I recall aluminium having a higher coefficient of movement with temperature changes. So just your hand warmth holding the base as you move it around might cause the squareness to shift to some greater degree than if the base was steel. But if you were to paint it with a few coats of paint or give it a thicker textured paint I'd say that would insulate the aluminium well enough for real use. Something like the "stone" paint?

                You'd also want to drill and partially press fit some steel ball bearing feet which you then grind and lap flat to actually run against the granite.
                Chilliwack BC, Canada

                Comment


                • #9
                  I see you were posting more as I was thinking and typing.

                  I love the simplicity of the interchangeable center posts and upward facing indicator. I've got a little Federal indicator with a threaded back hole which I could use for something similar.

                  One more thing. The fancy guys' comparators all used only one "toe". With that in mind you might want to slightly machine back the three upper piston lands and just leave the lowest one as the toe.
                  Chilliwack BC, Canada

                  Comment


                  • #10
                    Originally posted by BCRider View Post
                    You are running on in a direction you may not need to go. The whole point of doing a CS and then having a comparator is that you can calibrate the comparator and then use that as an "indicating square" on other surfaces. So no need for the reversal trick.
                    True, but I thought it all out long before I had the CS. Last year actually. Now that I have both, Thanks for the tip! Now I can use it as an indicating square. The original question I had last year was (for example) how would I know that a large piece of angle iron was square on the plate? You would want to test vertically in the exact same spot on only the one face. In using the squareness checker, I hold it still on the plate and move the part so the aluminum isn't much of a problem.
                    Last edited by nickel-city-fab; 03-05-2020, 03:19 PM.
                    25 miles north of Buffalo NY, USA

                    Comment


                    • #11
                      I have calibrated it against the cylinder square, and now it zeroes beautifully every time. The reason I leave the base round is because that way it doesn't matter where the screw threads end up when I change the rods -- the indicator will still have the same radius to work with regardless.

                      I have both a "tenths" indicator and a .0005 indicator for this tool. Both older Federal Miracle Movement.

                      If I was able to do this the way I had wanted, I would start with a piece of 3.5" dia 12L14 steel and simple turned the base and post all in one setting. And then made a decent indicator clamp. But I have only the 7x10 lathe to work with and a less than zero budget, so I have to get creative.
                      25 miles north of Buffalo NY, USA

                      Comment


                      • #12
                        For some time now I had wanted to make a squareness comparator.

                        But I have only the 7x10 lathe to work with



                        May i ask why?
                        What are you machining or making that requires this sort of accuracy?

                        Comment


                        • #13
                          Originally posted by thaiguzzi View Post
                          For some time now I had wanted to make a squareness comparator.

                          But I have only the 7x10 lathe to work with



                          May i ask why?
                          What are you machining or making that requires this sort of accuracy?
                          Spacecraft, it is not easy to land a craft where you want it 50,000,000 miles from its start.

                          Comment


                          • #14
                            Originally posted by thaiguzzi View Post
                            May i ask why?
                            What are you machining or making that requires this sort of accuracy?
                            Have you never wondered just exactly how square your squares are? Or how square your angle plates are that you use on your mill? Or even how square the jaws of your milling vise really is?

                            When checking down to a thou or two over 4 inches I don't trust my eyes to see light through such a gap. And really are any of the squares we have truly that trustworthy even if they have a "good" name on them?

                            A cylinder square or Tom Lipton's 123 block trick is something that can be tested with a squareness comparator and other common shop tools to a very fine degree. And then in turn the cylinder or block can be used to calibrate the squareness comparator to check other things that we only assumed are square.

                            NCF, I like your center post idea with a set of different height posts. But keep in mind that the thread will not necessarily clock to the same spot each time it is changed. And any specs of dust, grit or any wear in the threads will alter the final resting angle of the post and thus the position of the dial indicator. So each time you change posts you would need to recalibrate the zero dial using the CS. Sorry if that seems obvious but your post above made it sound like you might be thinking that you can just rely on the posts taking their positions consistently.

                            Once set on a post through you should be able to use it for quite a long time with a good degree of trust. Although for things done on a surface plate my own thought is that something like a comparator should be confirmed and calibrated as needed using the CS before the actual work gets started just to be sure that things didn't shift.



                            Chilliwack BC, Canada

                            Comment


                            • #15
                              Originally posted by thaiguzzi View Post
                              May i ask why?
                              What are you machining or making that requires this sort of accuracy?
                              1) Practice hitting my numbers with limited resources
                              2) for the fun of it
                              3) the skill will likely be very useful in the future when I go back to work
                              25 miles north of Buffalo NY, USA

                              Comment

                              Working...
                              X