Announcement

Collapse
No announcement yet.

More Manual Machining: sine bar!

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

  • More Manual Machining: sine bar!

    Let me state up front that I have absolutely no use for a sine bar, I'm just doing this for the fun and exercise of it. Maybe a use will come to me in the future. Being quite limited in my shop capabilities at the moment, I like to challenge myself to find alternative ways of doing things. In this case, I have no mill, but I want to do something that is not a lathe job. How can I make a sine bar using only manual methods?

    I think I came up with a way. Mainly, make a piece of steel exactly 5 inches (125 mm) long and bolt some knife-edged end pieces perpendicular to this. Set them at the same height relative to the bottom/top of the bar, so that you have exactly 5 inches between the knife edges.

    I have mostly cheap structural A36 steel (welding construction steel) to work with. Hot-rolled and nasty, but free. I chopped off a chunk of 3/8 x 2 flat bar (10mm x 50mm) a bit over length, to clean up with manual labor. Everything on this project can be measured with mics and simple tools such as the mic standards, and manual labor.

    I cleaned up all surfaces with the cheap Harbor Freight diamond hone and some 220 wet/dry paper with oil on glass. I burned some calories over a few hours. The top and bottom are within .0005 of parallel over the length of it. I'll rest up and do some more honing tomorrow.

    BTW that small wood box on the left houses the cylinder square from last time.

    Click image for larger version

Name:	sinebar-1.jpg
Views:	313
Size:	70.0 KB
ID:	1864709

    25 miles north of Buffalo NY, USA

  • #2
    NO FAIR! ! ! ! You got a box for your cylinder square ALREADY ! ? ! ? ! ?

    Apparently I need to get cracking! ! !

    The knife edge is a slick idea but for the wear that will occur and the loss of absolute accuracy as it wears or distorts....

    But can you NOT do the job with round stock as the supports? Your bar is 3/8 thick. So if you used a couple of two inch pieces of 1/2" round stock set against a piece of bar that was 4.5" long then the centers would be at 5.00000....000000, right? And if you measure the round bar and it's not quite spot on .5000000000 then you could fudge the length of the spacer bar. As in if the drill rod was .4985 then make the sized bar you're doing 4.5030 so the centers are spot on 5.0000.

    Then to line it all up you lay on a top plate to use to align and hold the spacer bar and round bars. Screws not shown.... Notch added to allow for a bit of clearance for the gauge block stack.

    Click image for larger version  Name:	Milless sine block.jpg Views:	0 Size:	44.2 KB ID:	1864725
    Last edited by BCRider; 03-29-2020, 03:01 AM.
    Chilliwack BC, Canada

    Comment


    • #3
      Without a surface grinder and a cylindrical grinder, making a sine bar really is not on. It is very different from a cylinder square. Incidentally, I have been unable to finish the cylinder square you gave me as the museum is closed for the duration. The part finished bar is oiled up and safe inside one of those large tool tubes. I was unable to get into the museum at the last moment to oil up the machines as the lockdown prevented me. Hopefully, they will be ok when I get back in a couple of months time.

      Comment


      • #4
        I'm thinking I could get away with it by using some pieces of an old file for the knife edges, and plenty of patience. Grinding the bevel on only one side, so that the inside is exactly 5.0000 between the edges. No surface grinder or cylindrical grinding required, since there aren't any round pins
        25 miles north of Buffalo NY, USA

        Comment


        • #5
          Things like cylinder squares, sine bars, and other high accuracy tools, I have to wonder sometimes.
          How accurate is accurate?
          How accurate are we actually in the typical HSM enviroment?
          It is easy for people to talk about accurate grinding, tolerances out to .000x",
          But, how many of us have a machine that will really hold .000x consistantly?
          Given such a machine can the typical among us operate said machine out to .000x?
          Many of us have imported inexpensive machines, or older iron machines with wear in them, they are cheaper and that is why we bought them.
          A benchtop round column mill? A benchtop tilt column mill? A old lathe with backlash in the cross slide??
          Is it really gonna matter if our sine bar is accurate to tenths?
          All those that have a metrology lab in their garage raise your hand and say "AYE".
          I like the direction N-C-F is going, and I believe the results will satisfy the vast majority of HSM guys.
          The post by BCRider looks so simple maybe even I will try it.

          Comment


          • #6
            Originally posted by Ringo View Post
            Things like cylinder squares, sine bars, and other high accuracy tools, I have to wonder sometimes.
            How accurate is accurate?
            How accurate are we actually in the typical HSM enviroment?
            .........
            Is it really gonna matter if our sine bar is accurate to tenths?
            ..........
            I like the direction N-C-F is going, and I believe the results will satisfy the vast majority of HSM guys.
            LOL believe me if I actually *had* machines I would not be doing it this way.
            Waiting on tax return and corona stimulus to kick start my shop.
            This is a lot of work. One nice thing about doing it this way,
            it's easy to "sneak up" on tenths and hit your mark.

            In a sine bar, tenths actually do matter quite a bit,
            they have a way of multiplying the error in an angle.
            But that is a very important point about being accurate in a home shop.
            I think if you can hit your numbers consistently, that is the most important thing.
            Making things fit properly is the most important thing.
            Being able to consistently hit a thou or two is pretty good IMHO
            maybe even in a commercial shop.

            I'm soaked in sweat from lapping, those Harbor Freight diamond laps are earning their keep LOL best kept secret in machining is a $9 set of 3 diamond laps at HF lol

            Here's a rough sketch of the idea:

            Click image for larger version  Name:	sinebar-plancropped.jpg Views:	0 Size:	254.5 KB ID:	1864840

            The critical dimensions are the 5.000 from end to end of the flat bar, the parallelism of the faces, and setting the end pieces at the same height (easy to do).

            Gonna rest up a bit and do some more honing. Keeping the laps drenched in oil on a glass sheet, fastened to an old drawing board. Occasional cleaning with a nylon brush. I'm all sweaty. Sweat equity!
            25 miles north of Buffalo NY, USA

            Comment


            • #7
              I have 0.3613 thick on both ends of the bar now, with some light scratches remaining from when I removed the scale with the angle grinder. Can't catch a fingernail with them though. You have to really bear down on the diamond plates as the surface area increases contact, and keep the plates clean. There is a bit of roll-off around the edges about 093 wide but I'm not going to worry about that, the main area is flat. Worked up another sweat, taking a break for typing. The finish is a soft sheen like.
              25 miles north of Buffalo NY, USA

              Comment


              • #8
                Whew! More sweat equity in my sine bar today. I chopped the long sides down and cleaned them up, narrowing it to 1.425 (36.2mm). This has the nice side effect of removing the "sloped off" sides and leaving me with a nice "square" bar. Trying to get some decent pics is like pulling teeth. The bar is within a few tenths of thickness all along its length.
                25 miles north of Buffalo NY, USA

                Comment


                • #9
                  Not to burst your bubble BUT how do you really know how near to truth in the measurements you are getting. For instance, how near to 68 degrees F is your shop, when were your mikes last checked by an outside lab, how flat is your surface plate?
                  I thoroughly applaud and appreciate your efforts,in the absence of cash to buy tools, making your own, is the sign of a real enthuiast.
                  In all likelihood your home made tools will meet and exceed the needs of most machine shops.
                  . It is interesting, and sometimes humbling to try to make things to as good a standard as possible, but it is also a little eye opening to have ones best efforts checked on a co ordinate measuring machine run by a well trained operator.
                  Have fun work safe, regards David Powell.

                  Comment


                  • #10
                    Originally posted by David Powell View Post
                    Not to burst your bubble BUT how do you really know how near to truth in the measurements you are getting. For instance, how near to 68 degrees F is your shop, when were your mikes last checked by an outside lab, how flat is your surface plate?
                    I thoroughly applaud and appreciate your efforts,in the absence of cash to buy tools, making your own, is the sign of a real enthuiast.
                    In all likelihood your home made tools will meet and exceed the needs of most machine shops.
                    . It is interesting, and sometimes humbling to try to make things to as good a standard as possible, but it is also a little eye opening to have ones best efforts checked on a co ordinate measuring machine run by a well trained operator.
                    Have fun work safe, regards David Powell.
                    Lol "enthusiasm", more like desperation... ya do what ya gotta do...

                    True, that there is an absolute limit. I checked my mics on a new gauge block set which claims NIST cert. The surface plate is likewise new grade A but non-cert. Altogether, the accuracy of my measuring setup its way beyond what what I can actually do -- which is a good thing! I can't really do much about the temp in a home environment (I'm actually inside the house), but I figure it'll be OK since everything is at the same temp and same level. Ultimately this means I'm trusting the gauge blocks that I setup the mics with.

                    FWIW the big mic set came with its own end standards from Brown & Sharpe, and I tried those against the blocks -- couldn't tell any difference, it must be far less than a tenth.... Which is amazing since I'm sure the mics are as old as me! It really makes you appreciate the old-time mechanics, how much work it takes to get consistent "tenths" with only an oilstone, *and* try to keep the correct geometry!
                    Last edited by nickel-city-fab; 03-30-2020, 02:36 PM.
                    25 miles north of Buffalo NY, USA

                    Comment


                    • #11
                      Making a sine bar an exact integerial length, e.g. 5", only matters if you don't know how to operate a scientific calculator. As long as you can accurately measure the length you're golden. Just plug it into the equation below.

                      I've provided the error equation for the angle error due to errors in the known value of the length. Even a rather large error (0.001) will cause an angle error of only about 24 arc seconds (115 microradians) for a 30 degree angle.

                      It's difficult to comprehend just how small 115E-6 radians is. If you make a laser pointing error of that amount at a target 1000 yards away, you'll be off center by only 4+ inches.

                      The equation for a sine bar is:

                      sin(A) = S/L

                      where:

                      A = desired angle
                      S = stack height
                      L = effective length of sine bar (typically distance between roll centers)

                      Differentiating with respect to L, we have:

                      cos(A) dA = S/L^2 dL (minus sign dropped since we're only interested in error magnitude)

                      where:

                      dA = error in desired angle due to...
                      dL = error in L

                      Plugging in some representative values

                      A = 30 degrees
                      L = 5
                      S = 2.5
                      dL = 0.001

                      and solving for dA, we get:

                      dA = 2.5/(25*0.866) * 0.001 = 0.000115 radians = 0.006616 degrees = 23.8... arc seconds
                      Regards, Marv

                      Home Shop Freeware - Tools for People Who Build Things
                      http://www.myvirtualnetwork.com/mklotz

                      Location: LA, CA, USA

                      Comment


                      • #12
                        Originally posted by mklotz View Post
                        Making a sine bar an exact integerial length, e.g. 5", only matters if you don't know how to operate a scientific calculator. As long as you can accurately measure the length you're golden. Just plug it into the equation below.

                        I've provided the error equation for the angle error due to errors in the known value of the length. Even a rather large error (0.001) will cause an angle error of only about 24 arc seconds (115 microradians) for a 30 degree angle.

                        It's difficult to comprehend just how small 115E-6 radians is. If you make a laser pointing error of that amount at a target 1000 yards away, you'll be off center by only 4+ inches.

                        The equation for a sine bar is:

                        sin(A) = S/L

                        where:

                        A = desired angle
                        S = stack height
                        L = effective length of sine bar (typically distance between roll centers)
                        Thanks! I love my TI-30Xa, but I didn't want to assume that it would always be available. Or maybe somebody else has to use my bar. So, I decided to get as close as possible to the "standard" dimensions to hedge against any possibilities. FWIW I have measured a "bulge" of about .0008 in the middle of the working surface that is going to bother the heck out of me until I get it flatter.... I don't actually care about the bottom surface, that can do whatever it wants. But I care a *lot* about the top surface, and the length.
                        25 miles north of Buffalo NY, USA

                        Comment


                        • #13
                          A memorable quote I've always liked: "Measure it with a micrometer, mark it with a piece of chalk, and then cut it with an axe."
                          Lynn (Huntsville, AL)

                          Comment


                          • #14
                            Gah. lynnl might be right. I just checked the profile on the surface plate, and I get a football shape. I actually had a few bucks today, wonder of wonders, so I went on eBay and ordered up a chunk of 3/8x1-1/2 precision flat stock. Gah.
                            25 miles north of Buffalo NY, USA

                            Comment


                            • #15
                              Originally posted by Ringo View Post
                              Things like cylinder squares, sine bars, and other high accuracy tools, I have to wonder sometimes.
                              How accurate is accurate?
                              How accurate are we actually in the typical HSM enviroment?
                              It is easy for people to talk about accurate grinding, tolerances out to .000x",
                              But, how many of us have a machine that will really hold .000x consistantly?
                              Given such a machine can the typical among us operate said machine out to .000x?
                              Many of us have imported inexpensive machines, or older iron machines with wear in them, they are cheaper and that is why we bought them.
                              A benchtop round column mill? A benchtop tilt column mill? A old lathe with backlash in the cross slide??
                              Is it really gonna matter if our sine bar is accurate to tenths?
                              All those that have a metrology lab in their garage raise your hand and say "AYE".
                              I like the direction N-C-F is going, and I believe the results will satisfy the vast majority of HSM guys.
                              The post by BCRider looks so simple maybe even I will try it.
                              I would think quite a few do have that equipment. Start with the surface plate and work your way up and it is easily possible to work to less than .001". Once you get beyond a machine that's in pristine shape it's the man running the machines that provides the precision and accuracy. Learning how to work around the limitations of worn out machines and still produce good, in-tolerance work is what it's all about in a lot of home shops. Experience brings understanding of what is happening and why, along with how to work around it. This costs time compared to working on pristine machines, but in a home shop that's not so important.

                              The better and more relevant question is: is it necessary to work to these levels in a home shop environment? And that can only be answered by the guys who are doing the work and by what they are doing. Sometimes yes, sometimes no. Some do it for the pure enjoyment of doing something as well as they can, and theres nothing wrong with that either.

                              Oh, and do yourself a favor and put the metrology lab in the basement rather than the garage - much more stable humidity and temperature in there in most cases...
                              Last edited by eKretz; 04-11-2020, 03:39 AM.

                              Comment

                              Working...
                              X