Announcement

Collapse
No announcement yet.

State of the art accuracy 60 years ago?

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • State of the art accuracy 60 years ago?

    Ok,
    This is in no way a tool gloat or a "nice score" type of thread. That would be highly disrespectfu, And to be honest. I really can't think of something like this as a tool I now own. I'm more of a caretaker of it for now I guess.

    Last month there was a post made by a gentleman in his early 80's showing a Sine Bar built by his long deceased friend on the Chaski forum. If my crap computer skills are correct? I'll attempt to put a link for it here. http://www,chaski.org/homemachinist/...php?=4&t=92286 This link shows the pictures.

    For some unknown to me reason, He thought my comments were enough to send it to me. He refused my buying it from him, A donation to any charity of his choice in his or his friend and builders name, Or even the mailing costs from Arizona to B.C.. He simply wanted it to be passed on to someone who would appreciate it's obvious craftsmanship, And the man who was skilled enough to build it to these accuracy levels.

    I've read just enough about the Los Alamos site to have some ideas about what kind of accuracys were needed, And just how good you would have to be to even work there as a machinist or tool maker. Given that this was a fully funded and controlled government site. And the precision required to even do the work. Then logicly the equipment and methods used would have been to the highest accuracy standards, Actual costs would have been immaterial. The 60 year old hand drawn sketch done by the original builder was included with this sine bar. The inspectors exact measurements were also marked down in five decimal places. The center - center distance for the round rolls are given as 5.00034. There's a lot of members here with far more knowledge than mine. So I'm wondering if anyone here could give an example or at least a few thoughts about what would have been used to measure to these accuracy levels 60 years ago? Gage blocks accurate to within millionths have been around since the 1920s. I've yet to find out how those exact measurements were obtained.

    I've also read that metal will change it's exact dimensions over time. Can anyone confirm that? And if so what direction it goes. Shrink, expand?

    Given how pristine this tool is and it's age. I'd suspect it was built from hardened stainless and then precision ground. There's not a single tool mark left anywhere on this tool. In fact it makes my $300 Suburban built sine bar look a bit crude.

    This has been one of the oddest, Yet nice at the same time things that have ever happened to me.

    Oh yeah, Exactly what Los Alamos is or was, And why that site was needed or not isn't a issue that needs to be debated here. I'd like it if this didn't become a locked thread.

    Pete

    Crap, My Chaski link doesn't work. As of today it's in the general forum, Page 2 and towards the middle of the list. Thread title is "What is this".
    Last edited by uncle pete; 03-21-2012, 06:05 PM.

  • #2
    Your link has a typo. Here is the corrected link.

    http://www.chaski.org/homemachinist/...php?=4&t=92286

    A very nice tool. I too am looking forward to the discussion here on this tool and the accuracy to which it was built all those years ago.
    Cheers,
    Gary

    Comment


    • #3
      Very good post and questions Pete, and congratulations on being the lucky recipient of this tool !!

      Comment


      • #4
        The inspectors exact measurements were also marked down in five decimal places. The center - center distance for the round rolls are given as 5.00034.
        I don't think that 5.00034" for a nominal 5.0000" is any big deal and is an error of 0.00034" which is about 3 1/3 "tenths" - almost "half a thou".

        Many here can work to that order of magnitude of accuracy if needs be.

        I'd be a bit less unimpressed (perhaps) if there was another "0" ie 5.000034 (about 1/3 of a "tenth") or two ie 5.0000034 (ie 3.4 "millionths" of an inch)

        I'd like to see some one post a copy of a USA-made 5" sine bar inspection sheet/s for comparison.

        By the time any errors on other parts of the sine bar were accumulated (taking into account all the +'s and -/s) the end result may be significant.

        I'd be more impressed if the accuracy of the sine bar matched the accuracy of the "slip gauges" - Classes 0 (Laboratory), 1 (Inspection), or 2 ("Shop") - with which I'd expect that they'd be used.

        Sorry to be the party pooper.

        Not.

        I'd like to hope that one or more "0"s were left out by the OP ie as a "typo". _O"

        Comment


        • #5
          Originally posted by oldtiffie
          I don't think that 5.00034" for a nominal 5.0000" is any big deal and is an error of 0.00034" which is about 3 1/3 "tenths" - almost "half a thou".

          Many here can work to that order of magnitude of accuracy if needs be.

          I'd be a bit less unimpressed (perhaps) if there was another "0" ie 5.000034 (about 1/3 of a "tenth") or two ie 5.0000034 (ie 3.4 "millionths" of an inch)

          I'd like to see some one post a copy of a USA-made 5" sine bar inspection sheet/s for comparison.

          By the time any errors on other parts of the sine bar were accumulated (taking into account all the +'s and -/s) the end result may be significant.

          I'd be more impressed if the accuracy of the sine bar matched the accuracy of the "slip gauges" - Classes 0 (Laboratory), 1 (Inspection), or 2 ("Shop") - with which I'd expect that they'd be used.

          Sorry to be the party pooper.

          Not.

          I'd like to hope that one or more "0"s were left out by the OP ie as a "typo". _O"
          With a sine bar the actual distance doesn't matter, the point is that the inspector was (apparently) able to measure to 5 decimal points, which allows the thing to be set accurately.

          I have no idea what instrument or method was used in the measurement.

          Comment


          • #6
            That is a fine piece of work. I have several tool maker made pieces, a couple of sine bars, angle blocks and the like. I also feel like I am just a caretaker and hopefully I can find a suitable recipient when the time comes.
            James Kilroy

            Comment


            • #7
              Loply, jkilroy,
              Exactly correct. And I'm happy you can appreciate the talent shown by the builder.

              I didn't start this thread to "impress" or unimpress anyone. I'd simply like to know how it was possible 60 years ago to measure by proper inspection methods to that level of accuracy. And does metal change it's true dimensions over a period of time. I filled in what happened because I thought some might find it an interesting and odd occurence that just happens for no logical reason.

              But the man who built this does deserve to be remembered as a true craftsman.

              Pete
              Last edited by uncle pete; 03-21-2012, 08:08 PM.

              Comment


              • #8
                That is a nice piece of work and a special treat to get it from the original maker. I have quite a few pieces that were shop made, a few that I also got from the original maker. I enjoy using them and try to give tham a bit of extra respect.

                With that, fifty or sixty years ago is not the middle ages guys. A few here were starting on their apprenticeship then, and projects like this were often a part of that training. They were also made from cutoffs and drops while the machinist had time to spare while a long cut was being made.
                Jim H.

                Comment


                • #9
                  While I of course cannot say exactly what method was used to measure to within 5 decimal places one can do so with an inspection grade set of gage blocks, a .0001" indicator (splitting the graduations), an inspection grade granite plate and an environmentally controlled inspection room. These are items one would expect Los Alamos to have in its inventory.

                  And yes in metrology 50 to 60 years ago is ancient history when compared with the computer controlled measuring equipment available today. Just take a look at a 50 to 60 year old car and compare it with a 2012 model and you'll have a similar yard stick. I've helped make repair parts for machinery of that vintage and the tolerances were much looser than what we use today.

                  That said, it is an excellent example of fine craftsmanship and the current owner is blessed to have the sine bar.

                  As to metal changing size over time the answer is maybe. In the case of a sine bar it's only going to change due to the change in temperature and would be based on the coefficient of contraction & expansion for that material (assuming one did not try to use it on a 5 ton workpiece). Other items could/would have a different answer. If for example a piece of metal was hung vertically with a weight on the end, the forces due to tension would cause it to stretch over time. Another piece in compression would shrink in height over time. A beam supported in the middle and one end with a weight on the unsupported end will bend over time. Other examples would yield similar results.
                  Last edited by Dr Stan; 03-21-2012, 10:13 PM.

                  Comment


                  • #10
                    Nice sine bar (though it's true quality would require some slip gauges, surface table and comparator to check).

                    I made one fairly similar as a third year toolmaking apprentice, it was a standard project at our tech course and mostly rather simple. It was a good project because of the grinding skills required and particularly because it teaches about comparitive measurement, in this case comparing slip gauges to sine bar.

                    The "rollers' were made from one piece of bar, roughed out between centres with small diameter spigots each end and the middle parted but not right through, then hardened and ground as a single shaft between centres, final parting done on surface grinder with cut-off wheel. Because they are ground as one, diameter is not important, just that they are the same diameter, fairly easy to get on a cylindrical grinder.

                    Squaring up all sides of the sine bar on the grinder is a good skill, made easy if you have a square block (an essential tool to make and have if you are doing much grinding). Otherwise tedious.

                    Location for rollers is done using the relieved side of the wheel on surface grinder, so you learn about relieving and using a backstop.

                    The accuracy comes from checking it with a finger indicator on a surface table, sine bar vertical and comparing each roller location face to a stack of slip gauges. Comparitive checking like this is a basic, powerful tool, and would not have changed in 60 years. You can do it in your garage...

                    I still have mine wrapped up in oiled cloth...

                    BTW, you usually find tools like this are made by apprentices simply because you are not given time to fool about when you are a tradesman (Government work may be different!)
                    Last edited by Peter S; 03-21-2012, 10:49 PM.

                    Comment


                    • #11
                      I am certainly not at the skill or experience level of many on this board so my contribution here will be the love of "stuff" like this, "romanticized" perhaps but my mind just works imagining the whys, hows and wherefores, to me its a knowledge puzzle (see some of the comments in the EB thread as to the value of paper rather than electronic searches).

                      Anyway, with sine bars, which is the important part in terms of precision? Is it the flatness, the location of various holes or angles or is the the given length? Asking since, to me, the calculations involved change at great precision levels, so, could that account for the need to discern between 5" and 5.000034 (maybe a typo there, its mine) or 5.000035 say?

                      In terms of how did they measure this precise back then three thoughts:
                      (a) comparison to something more precise. With this you get into the discussion about how you get something as precise or more precise than what you have in your hand [I am thinking of the accuracy of index plates improving from a single hole...or so I have read] and so here it would be if it is this accurate, what did they compare it to? Or could there be something similar to the index plate thought...
                      (b) given the origin...when/how did they begin measuring a unit of time based on the decay of a given element? Would this have been something some of the same scientists were involved or would have known about? And if so, is there a way such knowledge could have been applied to measuring distance very, very accurately?
                      (c) regarding change in size/measured length, could the reason for such a precise measure be linked to a given temperature...I mean the most accurate gage blocks say, is there not a temperature as a "given" when those measurements are in fact true? So that only at that length would it be accurate...

                      Comment


                      • #12
                        Originally posted by JCHannum
                        That is a nice piece of work and a special treat to get it from the original maker. I have quite a few pieces that were shop made, a few that I also got from the original maker. I enjoy using them and try to give tham a bit of extra respect.

                        With that, fifty or sixty years ago is not the middle ages guys. A few here were starting on their apprenticeship then, and projects like this were often a part of that training. They were also made from cutoffs and drops while the machinist had time to spare while a long cut was being made.
                        First of all Pete congratulations for being lucky enough to be the 'keeper' of this fine piece of craftsmanship for as long as need be.
                        Of course you already know that now it will be up to you to select a worthy recipient when the time comes to pass it on.

                        I agree with Jim's comment about this date (1952) not being that long ago and also wanted to comment on the high degree of machining accuracy prevalent in so many industries even well before that time.

                        The date struck a chord with me as it was also the same year that the mighty B-52 heavy bomber first flew. Think of all the myriad systems in this one example alone and the degree of accuracy required in order to make it successful. So successful that a viable replacement has still not been found, 66 years after the design contract was first let out!

                        There are other examples I'm sure, but it does indeed amaze me of the high degree of precision that machinists of that era and before displayed almost daily.
                        Home, down in the valley behind the Red Angus
                        Bad Decisions Make Good Stories​

                        Location: British Columbia

                        Comment


                        • #13
                          The only real advantage CNC or CMM offer over methods used 50, 60 or 70 years ago is speed of measurement. The actual methodology is basically the same, comparing a part to a known standard.

                          Take a look around for some texts and catalogs from the late thirties and early forties to see how it was done. I have Pratt & Whitney and Van Keuren catalogs from this era and millionths were readily measured. I have a P&W toolmaker's flat that is lapped flat & parallel to 0.000010".

                          If you can, find Moore's Foundations of Mechanical Accuracy. It provides a great insight into just what mechanical accuracy is all about.
                          Jim H.

                          Comment


                          • #14
                            Great tool & the history with it, can't beat that! Thank for sharing this with us.

                            Comment


                            • #15
                              Well I've got a virgin set of NOS Mitutoyo grade 2 gage blocks I can use with this. But I personaly think it should be returned back to the U.S. when my times up. There might be a few specialised types of museums that might like it. I think the builder does deserve to be remembered. It's still tough to figure out to use it for what it was designed for or not. I'll have to at least once.

                              I also appreciate everyones posts about how it might have been measured. And as far as I know it's been properly stored on it's side since it was built.
                              No 60 years isn't that long ago, But it's a very long time ago compared to the standard high tech industrial measurement that's used today. JCHannum's point is of course correct, But our repeatable accuracy is far better today I think along with better abilitys to measure to much smaller limits.

                              Pete

                              Comment

                              Working...
                              X