The Precision Upon Which Civilizations Are Built

If you’re interested in making things (particularly metal things), you’re on a road that eventually leads to machine tools. Machine tools have a special place in history, because they are basically the difference between subsistence farming and modern civilization. A bold statement, I realize — but the ability to make very precise things is what gave us the industrial revolution, and everything that snowballed afterward. If you want to build a modern life filled with jet airplanes and inexpensive chocolate, start here.

Precision is more than just a desirable property. It’s a product. It has value, there is competition to create it, and our ability to create it as a species has improved over time. When your “precision product” is in the centimeter range, congratulations — you can make catapults and portcullises. Once you get into the millimeter range, guess what? You can make fine millwork in fancy houses, and indoor plumbing. Once you get sub-millimeter, now things get really interesting. It’s time for steam engines and automobiles. Once you get into the micrometer range, well, now we’re talking artificial heart valves and spaceships. Much like materials science, the ability to create precision is the unsung foundation and driving force of our standard of living.

Okay, so assuming I’ve sold you on the value of this product called “precision”, how do we make it? Machine tools are how humans currently get there, despite the dreams of the 3D printer crowd. Yes, drizzled plastic is great and the future is bright, but for right now, subtractive manufacturing is where it’s at when something has to be perfect.

How do You Make Precision Tools Without Using Precision Tools?

Okay, so let’s say it’s 1751 and you’ve got a pretty sweet French society under Louis XV, but you want to invent machine tools. There’s a chicken-and-egg problem here, because the secret sauce that allows machine tools to create precision is that they themselves are very precise. How do we make machine tools without machine tools?

The early history of the field is filled with incredibly clever people doing incredibly clever things to get around this problem. With careful designs, a machine tool can make a part that is more precise than the machine tool itself is. Yes, that sounds crazy, but we’re going to learn all the ways that this is possible. These techniques are how, starting in an age of wagon wheels and horsepower made from actual horses, Jacques de Vaucanson was able to invent the first metal cutting lathe that we would recognize as such. Or perhaps it was Henry Maudslay in Britain, or Andrey Nartov in Russia. The history books tend to reflect the citizenship of the people writing them, so it’s hard to know who was really first. Real history is messy, in any case, so maybe everyone was first.

Subtractive Manufacturing Crash-Course

Let’s put the history books down now and talk about the current state of subtractive manufacturing. Machine tools come in two basic forms- single-point, and multi-point. Single-point cutting machines include things like lathes, boring machines, and shapers. Multi-point tools include things like mills (horizontal and vertical), jig borers, and broachers.

Modern machine shops will be mostly CNC these days, of course. However manually-cranked machines are far from obsolete. They are used in many countries where labor is still less expensive than technology. Furthermore, we’re in the midst of a revolution of accessibility of manual machine tools for hobbyists. Thanks to Asian manufacturers, prices on new machines are orders of magnitude lower than ever before, and quality has reached a respectable level. Alternatively, extremely high quality used American machines from the early 20th century are plentiful and inexpensive (if you’re willing to do some tuning-up).

I hope that gives you a feel for where this class of tools we call “machine” originally came from. We’re at the start of a road that will be at times practical, at times theoretical, and at times heretical. Keep the history in mind, because we’re going to be talking about the machines and techniques used in modern practice, but they all have their roots way back in the 18th century. Much has changed, but I think you’ll find it surprising how much has stayed the same. This is the first in a series of articles aimed squarely at the reader who has never set foot in a machine shop, but is very interested to learn. Stay tuned!

97 thoughts on “The Precision Upon Which Civilizations Are Built

  1. ” Once you get sub-millimeter, now things get really interesting. It’s time for steam engines and automobiles. ”

    And if we were dropped into that environment, there would be the perception, “how primitive”.

        1. CMM is a coordinate measuring machine.
          It takes a machine to measure what a high tollerence machine can do. By the way the most accurate are EDMs. The numbers are at 5 digits under the decimal. .00002″

          1. Macona the most precise methods of machining we know is with particles. Electron beam machining EBM can be used for very accurate cutting or boring of a wide variety of metals. Surface finish is better and kerf width is narrower than those for other thermal cutting processes. Smaller than that only Laser . All that can be machined with the smallest particle, will be the most accurate machining process.

          2. For smallest tolerance machines, think lithography of semiconductors. They are in the hard UV range nowadays, in the sub 10 nM range. In college we had electron beam litho, which promised to be even tighter, but never quite worked.

      1. Well, we only have very primitive “atom-range” manufacturing.But one could assume that if we could build any “atom-range” structure we can design, then you get things like robots that repair cells and DNA. As well as computers that are much more impressive than what we’ve got now.

        1. Once you get to the scale of DNA, it’s no longer possible to build a robot in the mechanical sense, because your building blocks become too large and their behaviour starts to become indeterministic. People think of nanomachines as if they’re miniaturized regular machines, but actually, they’re not machines but very carefully designed chemical reactions.

      2. High end semiconductor production arguably is atom range manufacturing. There’s various parts of the process where the placement of individual or small groups of atoms is vital to the success of the endeavour, like when doping or in certain transistor configurations. Functional parts can be seven atoms thick. Less or more atoms means a significant change.

        1. As the old joke goes: be suspicious of atoms, because they make up everything.
          Which is to say, it’s probably graphite. (Graphite is flat and easy to stick stuff onto.)

      3. The next step in modifying matter, because machining is only a modyfing matter mechanism. Will be “intelligent matter”, “programmable matter” , or “intelligent atoms”. A definition of atomic intelligence can be: The non-random behavior of particles. Human kind is starting to appreciate and studying the existence of intelligence at the fundamental level of atoms which can be termed as internal intelligence. A measure of this intelligence is its ability to form molecules and all matter is formed from molecules. Understanding the internal intelligence of atoms and matter that is responsible for the creation of complex matter , is the future of mankind.

    1. I got to the end of the post and thought I must have hit the page down button or something. Is this the beginning of a series or something? I really do want to know the answer to the questions you opened.

  2. When I was in Manchester, England last Spring, the Museum of Science and Industry claimed the Industrial Revolution started there. (I don’t really argue that claim) First, with the Bessemer (sp?) process of making steel, then, concurrently, steam engines, pumps, drive shafts, and machines tools (lathes, grinders, drills) and from there came the mechanized gins, thread spinners, and looms, for making cheap fabric, followed by sweat shops, and urban migration…

    1. I think you just touched on the other critical issue that happened within the evolution of precision – it’s not just the machines that are created but also the materials that they are made from (and can work on) – you need hard tools and rigid machines (as they rise, so can your precision and your ability to replicate them).

        1. Well, therein lies the subtle link: precision needs rigidity, rigidity usually commands heaviness, and heaviness requires excessive strength to move it and predictably stop it, in a reasonable time frame. But a fast correcting control loop might help shed some of inertia, and reduce the friction losses as well.

      1. If you cut something, eventually your cutter gets blunt. You need to replace it, or sharpen it.
        Either way, the position of the cutting edge needs to be re-calibrated.
        And let’s not forget accuracy…

        “An analogy for accuracy and precision is to imagine a basketball player shooting baskets. If the player has accuracy, his aim always takes the ball close to or into the basket. If the player has precision, his aim always takes the ball to the same place [which may or may not be close to the basket].”

        https://labwrite.ncsu.edu/Experimental%20Design/accuracyprecision.htm

    2. Freezing Londoners drove the industrial revolution. As the coal mines got deeper, water became to much for horses to handle. The first “modern” steam engines was made for pumping. “Coal: A Human History” by Barbara Freese is a recommended read, giving new perspectives of coals importance now and before.

    3. Whoa, Whoa, Whoa! Hold on a minute! Whilst Manchester, and Lancashire more generally, did certainly make numerous important contributions to the industrial revolution (Arkwright’s water frame, Hargreave’s spinning Jenny, Kay’s flying shuttle, Nasmyths works in Salford etc etc etc), I always thought that the Bessemer process was primarily developed in Sheffield, which is most certainly NOT in Lancashire. Its in Yorkshire (which is exactly the opposite of being in Lancashire).

      1. I have a hand saw made in Sheffield circa 1850’s, very good steel.
        I’m sure if I’d been in Sheffield and toured its museums, I would have received a different historical perspective.
        B^)
        As a Yank, I thought we fathered the Industrial Revolution!

    4. One should not forget that early development of computers resulted in a major contribution to industrialization long before PCs existed. Charles Babbage, who built a mechanical computer in the dawn of the industrial revolution needed interchange-able screw threads. His contribution to standardization of screw threads is considered a major one and was part of the establishment of the Whitworth standard. One can read about his work in a book called “Faster than thought” wherein his mechanical computing “engine” is described. He was funded by Lady Lovelace who hoped that the computing engine would be able to predict the outcome of horse racing! And there were plans to build a computing engine that would have covered all of London existing at the time. Although his massive computing engine never materialized, it had all the functional attributes required by and found in a modern computer.

          1. Because they are the same thing in fact: matting grinded/scrapped surfaces tend to goes into a sphere (least energy point), so 3 surfaces tend to infinite radius sphere surface, a plane.

            You don’t need machinery to get high precision, a handfull lot of manual workers have done it through the history. But you need machine to have repeated precision manufacturing. Then you need accuracy to build what is known as interchangeable parts.

  3. This is a subject that always fascinates me. Whether it’s this specific topic of how precision tools came to be, or the equipment and stories of the great scientific experiments and discoveries of the 18th, 19th and early 20th centuries. Looking forwards to future articles about this on HaD.

    1. Root of precision machinery is in the clock making industry.
      A good compendium of all 16/17th technologie is “Encyclopédie ou Dictionnaire raisonné des sciences, des arts et des métiers” from Diderot and D’Alembert (you can read it online, french only).

      1. Have you seen a watchmaker’s lathe, it looks just like a woodworker’s lathe but much smaller.
        The key to precision machine tools is the invention of the way to make a precision lead screw – and that, to my mind, was pure genius.

  4. Nice intro. Make this a regular column? I’ll help…

    Anyway, one little point I push when I do this kind of intro, modulo all the talk of heavy metals and solid fastenings, is that 3d printed (plastic) objects can be treated much the same was as 3d cast (aluminum) objects — say your motorcycle engine block. One can rough out a casting and machine finish it to some more accurate alignment, then insert the appropriate bearing materials… Industrial revolution techniques are still quite valid.

    1. Using 3d printed parts as a casting to be machined is a topic I am very interested in. Maybe people tried it with little success due to technical reasons (like the type of plastics used do not lend themselves well to machining) or maybe not enough people tried it but I didn’t see a lot of content out there. I think it can be a very useful tool for rapid prototyping where you need precision only at some features of the part, like a precision bore through a coupler, etc.

      I tried turning a PLA printed part a while back but could not get a good surface finish. Maybe I did not have the right feeds and speeds or maybe I needed a much sharper HSS bit instead of a carbide insert, etc. Please share what you know!

      1. I’m not sure if I’ve tried PLA, but ABS machines ok (although there are some stringy bits). I was milling it, but if you’re turning you might want to look into polished or honed inserts like they sell for aluminium.

        1. Thanks. I’ll give ’em a try but I find that insert tooling has a lot of upfront investment. More often than not, you need to buy a box (10 of them generally) for at least 6-7 buck per insert, Not a machine shop so don’t have any tool representatives I know that could give me some samples.

      2. Grinding versus machining of 3D printed plastic parts could produce a better finish.
        Freezing the part to make it more brittle may also help. Rubber components can for instance be made brittle, thereby allowing them to be machined and ground.

      3. It happens daily in Dental Laboratories. Frames for partials are designed in cad, printed, then invested and cast using the ‘lost wax’ technique, and finished by hand before teeth and pink acylic are added.

        There are plenty of labs still waxing these by hand, but that is changing as the speed and knowledge of cad/cam increases.

        Mamy are also cad designing then milling wax or printing patterns to be used in lost wax methods again, but for pressing glass (Lithium Disilicate) into the form as an alternative to direct milling of Zirconia (yytrium stabilized, CZ, and other flavors)

        If you want more information, head over to dentallabnetwork.com and read all you like.

        1. My dentist is directly CNCin’g crowns now. They 3d scan the old tooth in the mouth, grind it down to a post that is then scanned as well. The two scans are put together, edited for clarity/errors/fill in missing bits (I had a chunk missing from biting down on something hard) and fed to a CNC holding an uncured ceramic blank. Once the CNC is done, the blank is cut from the remaining sprue with a sanding disk, touched up, and test fit. Once it fits, it goes into a kiln to be fired/cured. 20 minutes later, it’s glued to the post and you’re on your way home $1,000 US poorer.

          1. Oh absolutely. Just was limiting myself to the printing of parts as the post above was mentioning. The most common in office systems are just now starting to get close to the level of the things we have in dental labs with scanners claiming resolutions and repeat-ability in the single digit micron range as opposed to the scan wands 50+ micron ranges for the good ones.

            When the in-office systems first started with the intra-oral scanning, if you printed a model from the scan and made a crown from the scan and not the model, the crown wouldn’t fit the model.

            Things are getting much better. And far more fascinating. We are less limited now than ever in what is possible, and making things from Peek, Pekkton, fibre reinforced composites, lithium disilicate (what yours likely was), yttrium stabilized zirconia, cubic zirconia, plus all the metals we used 20 years ago. On top of that (sorry, bad pun) we can layer and stain these things with ever larger variants of porcelains, colorants, and other fun things before or after putting them in 1700C furnaces sitting on the desk near 5 axis mills.

            Not to mention all the gaps being bridged(another pun, sorry)(not sorry) between fixed crowns and removeables like dentures that can be installed so that they are functionally indistinguishable from natural teeth and will not come loose like granpa’s did. All with increased accuracy and new nanocomposite materials that ease the metamerism challenges and employ greater lifelike properties of fluorescence, chroma variations, translucency, and flexural strength.

            It’s a great and fascinating time to be a dental technician.

            If this gets your geekybits excited, head over to dentallabnetwork.com and read some.

  5. I own copies of both “Machine Tool Reconditioning” and “Foundations of Mechanical Accuracy” which together outline the process of going from rocks and dirt to machine tools with sub-micron accuracy.

    They are interesting reads, and have surprisingly broad scope since an understanding of alignments helps even when assembling something as simple as a 3d printer.

    I believe both books can be read for free online.

    1. Owning both of those you listed (and many more) they’re great at building up first principles towards achieving accuracy. https://www.amazon.com/Precision-Engineering-Chris-Evans/dp/1871315018/ Precision Engineering: An Evolutionary View is unfortunately out of print and so copies are expensive. It is by far the best book for a historical perspective as it works its way through history talking about the work done to develop lead screws for astronomical observation and horology and onto the development of the first machine tools and continues on up to relatively modern times such as the large optical diamiond turning lathe (LODT.) (The book is from the early 90s.) It also heavily cites sources after each chapter for further deep dives into primary sources.

  6. Any discussion of precision machine tools should include the Holtzapffel lathe. Manufactured from 1794 to 1928. Starting in the late 1980’s, some replicas have been made, mostly adhering to the mechanical dimensions of the originals so the old Holtzapffel manuals can be used.

      1. No its not, A hammer can be used to duplicate itself.
        The history of the blacksmith is an interesting one with early blacksmiths using rocks as anvils and forming tools to create their primary tools that are then used in their craft.

        Precision was also well known in the days of early forge work with blacksmiths hand forging friction fit tenons. Then came the scraping methods that were used to create highly accurate slide ways and guides, all hand done and these processes were used to build the early day machine tools.

        Progress gave birth to the mechanical sharper’s.

        1. Scraping is still used today. Some machine tools are still hand finished by scraping and a lot of machine reconditioning is done by scraping. No matter how big a machine gets, a patient and skilled guy with a scraper blade can still finish the slideways to their final precision.

          1. This is so true. And I know :/ had to do this on my CNC machine. a really slow process but well worth it. Scraping was also used in knife and sword making although the scrapers were a little more aggressive. and IIRC some of the early gear cutters were essentially a mechanical scraper.

  7. Oooh I’ve been waiting ages for somebody to tackle this subject! The area of metrology I find most fascinating is interferometry because it allows for extreme levels of precision with remarkably simple equipment, once you have worked out how to make decent optical components.

    1. I completely agree. I was hoping it wasn’t just a teaser though. QD – a really well written article – It’s apparent that every word was weighed and measured for effect. Please share your sources for data too, I’m sure there are some fascinating reads there too.

  8. Please make the next few articles a little less poetic. Not every dimension on an aerospace part is +- a few microns. And just because the process is capable of high precision doesn’t mean that you should tolerance it that tightly. There are real costs that are associated with increased precision (e.g. manufacturing, inspection, and/or process validation to name a few). Increased precision drives costs. However, it does not always drive performance. A 2 meter walking stick toleranced to +-1 cm probably performs equally to one that is toleranced to +-.01 mm. However, the later will be more costly.

  9. Great read so far…..funny for me I’ve never been on hackaday before today.i see the word precision and I’m in…Machine tools. …I interviewed 2 month ago @Simmons Machine Tool Manufacturing in Albany N. Y…….I start on Monday 01/29/18….I read Fabio one of the contributors apparently has been taking things apart since he could hold a screwdriver. ….my cover letter = I’ve been building thing’s since I was 5year’s old and disassembled my father’s lawnmower on my mothers sheets. No lie….can’t wait to get in that facility. It’s unbelievable I feel like the stars are starting to line up in my lifes ongoing pursuit of precision skills. .by the way I’m also average accomplished Billiards player

  10. How precision screws were made ages ago. A craftsman would hand cut two screws as accurately as he possibly could. Those two would be mounted side by side and geared or chained together, used to drive a cutter along a rod, also connected to the two hand cut screws to turn at the same speed.

    Errors in the dual screws would average out. The next step would be replacing the hand cut screws with the first two machine cut ones. Iterate the process two or three times and the setup should be producing lengths of threaded rod with a decent level of pitch accuracy.

    1. I cannot view the video at work but i briefly jumped through it and it shows CE Johansson gauge blocks.

      I have read that he converted his wives sewing machine into a grinder and made the first gauge blocks on it, at home. For me that would be one of the greatest hacks of all time but I have not found any image or description on how it exactly worked or was modified.

      Does anyone know? Would be a great HAD-article on the same topic.

  11. I might be the only one, but I got stuck on this line:
    “Machine tools come in two basic forms- single-point, and multi-point. Single-point cutting machines include things like lathes, boring machines, and shapers. Multi-point tools include things like mills (horizontal and vertical), jig borers, and broachers.”

    The two families of tools have always been “move the part” vs “move the cutter”. The lathe moves the part against a cutter, the mill and most other machines move the cutter against a stationary part. That line has gotten blurrier with time – but the point remains.

      1. Manual vs CNC really doesn’t come down to labour cost. It’s more to do with programming time vs time to run the job. Many simple operations will take longer to do CAD and CAM than to just put the part in a manual machine and do it. So for one off, manual machines often win. Once you go into production, CNC wins hands down. For many jobs the time to set up the work in the machine is far longer than the actual machining time.

        Often these days the difference in cost of good quality manual machines and a CNC machine is insignificant. So you may as well go CNC. That said, good quality manual machines last forever. I have a 75 year old lathe in my shop and a 50 year old mill. These days your hard pressed to find new machines built to the standards of that era.

        The hybrid solution is conversational CNC machines. Where the operator programs short operations as he goes. It skips the whole CAD and CAM process. The operator just selects from a list of macro’s and enters in the parameters. Stuff drill a round hole pattern centered on current position to z depth, 9 holes at N radius.

        Machining has been a big rabbit hole for the past 5 years or so. I’ve always had an interest and thanks to guys like Keith Fenner on youtube I’ve been able to explore it. It all started with a 3d printer kit at the local makerspace, then I got mini mill and mini lathe. Now I’ve got the big machines. I also now run a CNC router at work. By background is electronics, AV and computer networking.

      2. That was pretty much my exact point.
        “Stationary part” tools (mill, drill press, etc) have certain behaviors regardless of the number of cutting points on the bit – single point (boring bar, fly cutter, etc) vs multi point (drill bit, end mill, insert cutter, etc) doesn’t change those behaviors.
        “Moving part” tools (mostly lathe, with a few variations) behave differently, and again the number of cutting edges on the tool is not the relevant detail here.

    1. Oops, hit ‘report comment’ rather than ‘reply’ yet again. While I agree with you that rotate-the-part vs rotate-the-tool is how I think about mills vs. lathes, I’m not sure that’s a fundamentally useful way of dividing up processes: it’s just traditional. Sure, I do plenty of single-point operations (boring head) on my mill, but I’ve stuck a boring bar in the mill vise to bore a concentric toolholder attached to the spindle, and I do a lot of line boring on the lathe, where the workpiece is on the cross-slide, and modern vertical and horizontal machining centers are like two lathes and half a dozen mills in one work envelope. When I think about this, it seems more challenging/rarer to do multiple-point operations on a lathe than to invert the rotate-the-part concept.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.