Talk:Engineering tolerance/Archive 1

Application to Wikipedia[edit]

It seems to me that tolerances can also be applied to quality control in article editing. For example in stating a name of an individual in an article, instead of writing John, the editor may want to research to find that the individual not only had a middle name never used in public, but abbreviated his first name from Johnathan.--Mrg3105 (talk) 11:16, 20 December 2007 (UTC)[reply]

And in engineering and safety, to the physical distance or space (tolerance), as in a truck (lorry), train or boat under a bridge as well as a train in a tunnel (see structure gauge and loading gauge). Peter Horn 04:12, 17 February 2008 (UTC)[reply]

Margin of error[edit]

It was suggested that "engineering tolerance" be merged with "margin of error". These two are only losely related. Margin of error deals mostly with the variation of the measurement process and statistical data presentation. Engineering tolerance is the amount of variation the end-use will tolerate. These are different. Rlsheehan (talk) 19:58, 9 May 2008 (UTC)[reply]

Tag now removed. --Salix alba (talk) 17:04, 10 May 2008 (UTC)[reply]

Examples and History[edit]

Are there any good sources on orders of magnitude in tolerances for various processing techniques over manufacturing history? —Preceding unsigned comment added by 164.67.237.253 (talk) 22:50, 2 September 2008 (UTC)[reply]

Dubious[edit]

I'd be hard pressed to find any source that most tolerances are applied to shafts and holes. I apply tolerance to all sorts of non-shaft and -hole features everyday. As such, I feel that part of the sentence should be removed. Wizard191 (talk) 12:25, 27 August 2010 (UTC)[reply]

I'm not sure which specific sentence you refer to. Certainly the shaft and hole is only an example and not a limiting statement. Feel free to clarify this. Rlsheehan (talk) 13:48, 27 August 2010 (UTC)[reply]

It would be the sentence with the dubious tag on it. Wizard191 (talk) 15:13, 27 August 2010 (UTC)[reply]

Yes but it seems to idicate now that tolerances are used for many other types of items. Again, feel free to clarify the statement. Rlsheehan (talk) 16:21, 27 August 2010 (UTC)[reply]

Done. Wizard191 (talk) 16:28, 27 August 2010 (UTC)[reply]

10 years later..this article still has too much emphasis on shaft/hole fits which is really distracting from the basic concept. Gjxj (talk) 23:23, 12 May 2020 (UTC)[reply]

Evolution of precision?[edit]

There was a first part accurate within a micron, and since it was the FIRST such part it was made by a machine NONE of whose parts were that precise (in the common meaning of the word "precise"). Obviously this (a machine making parts that somehow, paradoxically, have less inaccuracy than every part in the machine making them) has occurred. But I'm bewildered as to HOW it occurs. Could someone write this article or if the info is alrady there point me to it?69.86.130.90 (talk) 19:23, 21 February 2011 (UTC)Christopher L. Simpson[reply]

If I understand correctly, you are asking how humans have managed to make increasingly precise machines. As you point out, usually machines make things less precise than themselves, which suggests a contradiction in how high-precision machines could come to exist in the first place. It is a very good question, probably better suited to precision engineering, but I'll take a stab here.

Overall, there are many clever ways to increase qualitative precision, and then there are clever ways to shrink measurement tools.

For qualitative precision, grinding two flatish surfaces against one another in random directions will tend to produce surfaces more planar than they started by way of abrading asperities. Similarly a spherical surface occurs naturally when you rub a convex surface against a concave surface, which is the traditional way of manufacturing lenses. Also, I think inertia will help you make increasingly precisely-round parts on a lathe (i.e., I think you could turn a bearing race more round than the races on the lathe you are using). That's at least a start on how you'd get qualitatively increasing precision.

For quantitative precision, two approaches come to mind: mechanical advantage and calibration. Mechanical advantage means that I can use a lever or a screw or the like to use one measuring device to make much smaller measurements. One simple device is a vernier scale. Similarly, even with a simple lens you could do photolithography to effectively manufacture very-small features. By calibration, I mean that if you have a device that is precise but not accurate, if you can measure its error, you can calibrate that error out. So once you have a way to make a very flat gage block that just isn't the right thickness, then you can think about adjusting the thickness.

I can't think of any examples of increases in precision that are due to engineering cleverness like the above. Unfortunately, history of measurement and precision engineering are a bit thin on this stuff. —Ben FrantzDale (talk) 01:10, 22 February 2011 (UTC)[reply]

I was too long in coming back to get this answer. I would like to express my thanks for these pointers.69.86.131.64 (talk) 11:20, 31 March 2011 (UTC)Christopher L. Simpson[reply]