It's a common misconception that tolerances amount to "how badly the thing can be made." Tolerances and allowances are needed to ensure fit and function of independently manufactured parts.
I retired as Chief Technologist in one of NASA's largest machine shops, after a career as machinist, metrologist, tool and instrument designer. The largest part of my job was to advise on the manufacturability of parts and assemblies. Any time I saw a design with 10 micron tolerances, I required a detailed tolerance analysis to demonstrate the need.
Reliance on tight tolerances to achieve fit and function is a hallmark of bad design.
I have installed an inexpensive pair of DRO's on my mill and found them entirely satisfactory. Details here:
http://groups.yahoo.com/group/GrizHFMinimill/files/D.C.%20Clark%20stuff/
David Clark in Southern Maryland, USA
--- In GrizHFMinimill@yahoogroups.com, "\"hanermo\" - CNC 6-axis Designs" <gcode.fi@...> wrote:
>
>
> > You are correct, the iGaging DROs are only accurate to .001" but for
> > 95% of machine work, they work fine.
> Please dont say that to a machinist (I am not, truly. Maybe a journeyman).
> And note that the real guaranteed results are +/- 0.002" worst case, or
> +/- 0.04 mm.
>
> Making any kind of "mechanical" part to a print, with a 0.04 mm error,
> is a reject for more than 90% of work.
> Mechanical assy just dont work at that, even screw holes dont align, and
> parts dont fit within each other, etc.
>
> For a bracket, box, mount of course even a 1 mm error is usually
> immaterial, but the whole idea about machine tools, for me, is trying to
> make good stuff well.
> If it works, looks good (shiny, stainless) and is better than store
> made, then I am happy.
>
> > If you are trying to make accurate cuts to press fit, you better be
> > measuring with mics anyway.
>
> Well- for me maybe, for most here possibly, for machinists, Absolutely not.
>
> The reason for DROs is so that you can make accurate stuff *without*
> measuring with mics, and so that bend, springback, slop, mounts, screws
> and bearing errors are eliminated.
> DROs are really good and useful, precisely because they allow one to
> actually get to 0.01 mm accuracies, easily, fast, and without measuring
> with mics.
>
> The chinese scales, igaging stuff, and digital calipers are not proper DROs.
> They just look like them.
>
> I dont have DROs, yet.
> When I get the money (february probably), I will install real glass scales.
> I will get the more accurate ones, at 5 microns, as they are almost the
> same price.
>
> PS.
> I make almost everything in stainless 303, 304 or 316L.
> Given that it takes longer, costs more, and I need to spend the $ to
> make the widget (as compared to store bought chicom junk), I want to
> make it good, strong and shiny as well.
>
> My inspirations are things like lee valley hand tools,
> Hitachi Koke industrial power tools,
> Haas indexers and mills,
> and so on.
> I have no interest in making stuff "almost as good" as a cheap chinese
> cast iron piece, weak, light, less-than-very-good, rusting, from soft
> cheese steel.
> If I need a basic junky utility tool I just buy it from the nearest
> chicom retailer, quite happily.
>
> Example:
> I will shortly be making wrenches.
> Short, long, thin, and normal thickness.
> My inspiration will be stahlwille.
> My wrenches will be much stronger (by 50%), lighter (20%) and better
> looking.
>
> A full set (metric, imperial), in both slim and normal, long and normal
> length will be about 12 pcs each, x 4, or 48 pieces.
> Material will be billet titanium, and material costs will be about 1000$.
> Note that equivalent wrenches woudl cost over 10.000 $, and the
> Stahwille ones would cost over 4000$.
>
> The wrenches will hang on my office wall, and will be used as a demo of
> what I can make.
> Accuracy will be to +0.01 mm - 0.00 mm,
> And I will only measure the first few with mics (I will be using mills I
> can trust for accuracy).
>
> If the wrenches were to be off by 0.03 mm or an igaging/chicom scale
> permitted error, they will round nuts, and will not work well.
>
> I fully understand and support saving and very well understand the
> reason people install igaging stuff.
> Also, it makes it intuitively good and nice to see the "almost good"
> results.
> The problem is, that you then expect to get "real" accuracy (nothing is
> accurate, truly) and expect to have DROs, when these are really just scales.
>
> The most important point, imho, is that people will buy these without
> actually knowing that they are NOT acccurate and you cannot use them to
> make stuff to *size*.
> So for those of us (minority, yes, but still) the money is totally
> wasted, and the disappointment factor high, when yuo cannot really use
> it for what you wanted it for.
>
> I ahve a large 600 mm (20") chinese digital caliper.
> It can be (by spec sheets and in reality) upto 0.08 mm off.
> Totally useless for measuring size, in mechnical fab, with machine tools
> I wasted the approx. 150$ it cost.
>
> It IS however good for measuring relative size, in large pieces, to
> about +/- 0.02 mm.
> Thigs like measuring two like pieces for relative errors.
>
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/GrizHFMinimill/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/GrizHFMinimill/join
(Yahoo! ID required)
<*> To change settings via email:
GrizHFMinimill-digest@yahoogroups.com
GrizHFMinimill-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
GrizHFMinimill-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/