History of Geometric Methods

In Dimensioning and Tolerancing

image link-topic-sf0.jpg

1. Few Sources

Despite its importance, our knowledge of the origins of geometric methods in dimensioning and tolerancing is most unsatisfactory. It seems that every slide presentation on the subject has a quick section on history, but they all just quote each other - recirculating the same few factoids without any actual evidence.

Most of what we do know of the early history of geometric methods in dimensioning and tolerancing comes from the brief history of the field given by F. H. Rolt in his Foreword to Stanley Parker's Drawings and Dimensions {Parker 1956}. After that, it's a matter of connecting the dots of the surviving evidence.

Had I come to this field first, I would have thought that this was an unusual, even extreme, example of an industry which was devoted to information destroying the information about its own history with astonishing rapidity - the more so because this erasure of the past is making it harder to apply the concepts of this field in the present. But I started studying this history after having spent several years devoted to the study of the history printing and must conclude, sadly, that, no, it's typical.

2. F. H. Rolt

It is through F. H. Rolt that we know what little we do about the origins of geometric methods in dimensioning and tolerancing, so it seems appropriate to try to learn a little more about him. His direct contributions to geometric methods in dimensioning and tolerancing seem to have been administrative and managerial, but his strong technical background and the nature of his management confirm his reliability as a source for this early history.

A biographical sketch of Rolt was published in The Production Engineer: The Journal of the Institution of Production Engineers Vol. 31, No. 3 (1952): 130, { PE 1952} , upon the occasion of his giving a paper. The first page of this is online on the "ieeexplore.ieee.org" website; I have not yet had the time to track down the entire article (and don't feel like purchasing the digital version just now), so my knowledge of him through 1952 will be limited.

He was born 1888-11-01. { Nature 1946} From the WorldCat and Smithsonian Institution entries for his books we have his full name as Frederick Henry Rolt. He should not be conflated with L. T. C. Rolt, the noted author on the history of technology.

We know from { PE 1952} that he gained a Diploma from the City and Guilds Engineering College, South Kensington in 1908 and took a B.Sc. in 1913 (it doesn't specifically say where, but { Nature 1946} indicates that it was the University of London "as an external student."). He "received his practical training" in the shipping industry: the P. & O. Steam Navigation Co., Caird of Greenock, the Fairfield Shipbuilding & Engineering Co. (Glasgow). The PE sketch continues, noting that he "joined the scientific staff of the National Physical Laboratory in 1912, and with the exception of a period during the last war when he was seconded to the Ministry of Supply, he has been engaged in the Metrology Division since that date."

An entry in Nature confirms the note in the PE sketch which says that in 1946 he became the Superintendent of the Metrology Division of the National Physical Laboratory [UK]. It also indicates, from the form of reference to him in the title, that by 1946 he had been awarded the O.B.E. From the abstract:

[title:] Metrology at the National Physical Laboratory: Mr. F. H. Rolt, O.B.E

MR. F. H. ROLT has been appointed to succeed Mr. J. E. Sears. Mr. Rolt was born on November 1, 1888, and educated at Rutlish School, from which he obtained a Surrey County Council major scholarship to the City and Guilds Institute, London, where he obtained the diploma in civil and mechanical engineering, afterwards obtaining his degree in science at the University of London as an external student. { Nature 1946}

We know from his Foreword to Parker's Drawings and Dimensions {Parker 1956} that he (Rolt) was the chair of both the 1942 Admiralty Committee and the 1944 Inter-Services Committee which brought Parker's ideas on geometric methods in dimensioning and tolerancing to the point of application (more on this below).

He also published:

{Rolt 1928} The Accuracy of Commercial Screw Threads. [Series: GB Dept. of Scientific and Industrial Research, Engineering Research Board, Special Report No. 4.] London: H. M. Stationery Office, 1928.

{Rolt 1929} Rolt, F. H. Intro. by J. E. Sears. Ed. R. T. Glazebrook. Gauges and Fine Measurements. London: Macmillan, 1929

{Rolt 1952} Rolt, F. H. The Development of Engineering Metrology. London, Institution of Production Engineers, 1952. 56pp.

This was the paper presented by Rolt as the first Sir Alfred Herbert Paper. It seems to have been published both in book form and in the Proceedings of the Institution of Production Engineers..

{Rolt 1961} Rolt, F. H. The Use of Light Waves for Controlling the Accuracy of Block Gauges. London: Hilger & Watts, 1961. 3rd ed. 1966.

3. Stanley Parker

We know very little about Stanley Parker, yet it would seem that we owe an entire field of modern industry to him - and, moreover, a field which stands behind almost every aspect of modern industry. This near-anonymity is to be regretted.

From what is known so far, he seems to have originated the field of geometrical methods in dimensioning and tolerancing with work beginning in 1938 and continuing through at least his book Drawings and Dimensions {Parker 1956}.

Modern sources say that he worked at the Royal Torpedo Factory in Scotland. While this may be true, I have not yet been able to find any actual evidence for it - or for the dates and capacity in which he was employed there. As he was involved with munitions manufacturing, contemporary sources are cagey about the details of that involvement. The earliest reference I have been able to find for this is a 1970 textbook which claimed that "Position tolerance theory originated with Stanley Parker at the Royal Torpedo Factory in Alexandria, Scotland" ( {Liggett 1970}, p. 4). For more on the Royal Torpedo Factory site (though little of its history as an institution), see the Royal Commission on the Ancient and Historical Monuments of Scotland's web page on "Alexandria, North Main Street, Argyll Motor Car Factory": http://canmore.rcahms.gov.uk/en/site/42488/details/alexandria+north+main+street+argyll+motor+car+factory/ The site began as the Argyll Motors Ltd. factory in 1906/1907 and is presently (2014) the Lomond Galleries (formerly Loch Lomond Factory Outlets shopping mall) and Loch Lomond Flea Market. Sic transit...

All that we do know about Parker with certainty comes (at present) from his own remarks and those of F. H. Rolt in Parker's Drawings and Dimensions {Parker 1956}. Rolt says that:

Parker says of himself:

"This book contains many criticisms of the faults of drawings of the past. They are based on the author's progressive experiences over many years as a craftsman, a designer of engineering products and their gauges, and as an inspector of various finished products. Two world wars have intervened during these experiences ..." (p. 8)

Parker's 1940 report, Notes on Design and Inspection of Mass Production Engineering Work. was published by the Gauge Design Drawing Office, Naval Ordnance Gauge Factory, Sheffield. {Parker 1940}.

Note: Google Books seems to think that he is Stanley Brampton Parker, author of The Vertical Vanishing Point in Linear Perspective (1947) and other works. But Stanley Brampton Parker (1881-1965) was an American. He was, for example listed in the 1919 Harvard Alumni Directory as of the Harvard class of 1903 and (in 1919) an architect in Boyleston, MA. It is unlikely that he is the Stanley Parker of the present discussion.

4. Early Work in England: 1938 to 1948

The early work on geometric methods in dimensioning and tolerancing is centered about Stanley Parker and various committees of the British Admiralty.

F. H. Rolt dates the origins of Parker's work to 1938, and its first publication to Notes on Design and Inspection of Mass Production Engineering Work (this was {Parker 1940}) See {Parker 1956}, p. v. The origins of this method thus spanned the outbreak of World War II in 1939.

Rolt continues:

"The value of this booklet [ {Parker 1940}] was appreciated by the Admiralty and led, in 1942, to the setting up of a Committee entrusted with the task of laying down a series of guiding rules for the re-dimensioning and tolerancing of the drawings of a particular naval weapon to facilitate its increased production by a number of new contractors. ..." ( {Parker 1956}, p. v)

Continuing further, Rolt says:

"... These rules [of the 1942 Admiralty Committee] were based up the general principles outlined in Mr. Parker's [1940] booklet, the subject-matter of which was subsequently extended by the Comittee in the form of two Admiralty manuals entitled Data Sheets for Designers and Draughtsmen, one dealing with components and the other with gauges." [See: {Admiralty 1944}] ( {Parker 1956}, p. v)

Rolt also notes that he (Rolt) was the chair of this Admiralty Committee.

Rolt continues:

"These manuals [ Data Sheets for Designers and Draughtsmen] were later used as a foundation for the work of an Inter-Services Committee [The Inter-Services Committee for Dimensioning and Tolerancing of Drawings], set up by the Ministry of Production late in 1944 for establishing a basic system for dimensioning and tolerancing the drawings of interchangeable engineering products of various degrees of precision. The Committee's work culminated in the publication by H. M. Stationery OFfice in 1948 of another manual under the title of Dimensional Analysis of Engineering Designs." [See: {Inter-Services 1948}] ( {Parker 1956}, pp. v-vi)

Rolt also notes that he (Rolt) was the chair of this Inter-Services Committee.

It is interesting to note that while the 1943 revision of British Standard 308, Drawing Office Practices, says almost nothing about tolerancing (it merely states in one paragraph that when necessary maximum and minimum limit dimensions can be given together), it provides this explanation in its foreword:

"The difficult subjects of tolerancing, the application of limits and fits, and grades of surface finish have merited the attention of special Technical Committees and separate publications will be issued in due course relating to these matters." ( {BS 308 1943}, p. 5)

This comment about tolerancing and limits is obviously a veiled reference to the 1942 Admiralty Committee. (It's only obvious to us, today. At the height of the Second World War, when these new methods were being used to increase naval ordnance production, this would have been classified information.) To the best of my knowledge this is the earliest reference to geometrical methods in dimensioning and tolerancing to appear in any national standards document.

The earliest more direct reference (that I have discovered so far) in the British civilian literature during this period is probably Gladman's article "Drawing Office Practice in Relation to Interchangeable Components" {Gladman 1945} (but I haven't read it yet).

It is important for the 21st century reader to realize that this early work by Parker and the various Committees, through at least Parker's 1956 Drawings and Dimensions, did not use symbols. Parker and others at the time expressed it using words and then-conventional drawing notations. 21st century texts often claim that GD&T is a language for technical communications - sometimes comparing it to natural languages such as French or Japanese (e.g., {Neumann 2009}, p. 1.2). Manifestly this is not so, as it was used without this symbolic overlay for over a decade, and in this form its originator wrote what is probably the finest book ever on the subject. The geometric methods originated by Parker are not a language but rather a system of ideas which remain valid independently of the language with which they are expressed.

5. Early Adoption in the US: 1945 to 1949

Gladman's article "Drawing Office Practice in Relation to Interchangeable Components," which had appeared originally in the English Proceedings of the Institution of Mechanical Engineers in 1945 was reprinted in the same year in the US by the Society of Automotive Engineers as SAE Technical Paper 450238. {Gladman 1945}.

The US Army Ordnance Corps. Standard for Dimensioning and Tolerancing, ORD 30-1-7, appeared in 1946 {ORD 30-1-7 1946}. Honsinger says that "this standard is truly unsophisticated" ( {Honsinger 1994}, p. 3 and several times thereafter). But Krulikowski credits it as having "introduced the use of symbols (rather than notes) for specifying form and positioning tolerances." ( {Krulikowski 1991}, p. 3) I haven't actually found a copy of it to read.

The Society of Automotive Engineers also introduced greater coverage of dimensioning - curiously, in a standard for aeronautical drafting {SAE Aero 1946} (see {Krulikowski 1991}, p. 4). I have not yet seen this.

US MIL-STD-8 seems to have made its first appearance in 1949 {MIL-STD-8 1949}. I haven't yet seen it, and do not know the degree to which it incorporated geometric methods. It was revised by {MIL-STD-8A 1953}. The handouts for a presentation by Applied Geometrics, Inc. for Carl Zeiss claims that MIL-STD-8 (1949) introduced the concept of GD&T "Rule 1" {AGI 2003}, foil 7.

Giesecke, 3rd ed. (1949): One way to trace the adoption of geometric methods in dimensioning and tolerancing is to consider how they were incorporated into any of the several long-running series of drafting textbooks by the major publishers. For example, editions of Giesecke's Technical Drawing have been published by various publishers (primarily Macmillan) from 1933 to the present. (See {Giesecke}, which redirects to the more detailed bibliographic information on Giesecke in the Drafting and Drawing Bibliography ). The 1949 Third Edition of Giesecke makes explicit reference, in its chapter on "Dimensioning" to ASA Z14.1-1946, Drawing and Drafting Room Practice (and in its Appendix 2 appears to reproduce much (all?) of the material from this standard). Yet neither it nor the material reproduced from Z14.1-1946 make any reference to geometric methods. (Interestingly, this edition of Giesecke also refers explicitly to {ORD 30-1-7 1946} (in a footnote on p. 424), but only for an unusual method of unilateral tolerances.)

6. B.S. 308: 1950 - 1953

It would seem that work to this point (England, before 1950) was driven by wartime needs. Rolt says of the transition to peacetime needs:

By this time [1948] Service [that is, Admiralty, etc.] drawings embodying the new system became fairly widespread in industry, and a Committee was set up by the British Standards Institution in 1950 to consider its introduction into British Standard 308 on Engineering Drawing Practice, which was at that time due for a general revision. Representatives of Industry and of the three Services sat on this Sub-Committee, Mr. Parker representing the Admiralty, and its persevering effots were eventually rewarded by the publication, in 1953, of this revised British Standard. It is significant htat more than half of this Standard is taken up by recommendations relating to dimensioning and tolerancing ... These recommendations made in this British Standard stemmed mainly from those set out in Mr. Parker's original booklet, and must be regarded as a complete justification and result of his pioneer efforts, started some fifteen years earlier." ( {Parker 1956}, p. vi)

Finally, Rolt goes on to note that Parker "played a full part" in the introduction of these ideas into American, Canadian, and European standards.

By way of background, British Standard 308 was first published in 1927 (as Engineering Drawing Office Practice). It was revised in 1943 under the same title {BS 308 1943}, without any incorporation of geometric methods. It was revised again in 1953 (as described here) to include geometric methods, with further revisions in 1964, 1972, and 1985. In the 1972 revision it was divided into three parts (General Principles, Dimensioning and Tolerancing of Size, Geometric Tolerancing). The 1985 revision was metric. It was withdrawn in 2000 in favor of BS 8888, Technical Product and Document Specification, which is essentially an interface into various ISO standards ( {Simmons 2004}, p. vi)

7. Parker's "Drawings and Dimensions" (1956)

Stanley Parker's work culminated in two documents: the 1953 revision of B.S. 308 and his 1956 book Drawings and Dimensions {Parker 1956}. He intended this book to provide "explanations of the principles and geometry upon which the dimensioning and tolerancing methods in B.S. 308 are based" (p. ix), and it does. This book remains, over half a century later, the finest book ever written on geometric dimensioning and tolerancing. Indeed, sad to say, it is probably the only really good book on the subject ever written.

This is all the more remarkable, to the modern eye, because he employs no special symbols at all. The complex symbolic lexicon of the later standards documents wasn't a part of his system at all. Yet all of the elements of geometric methods of dimensioning and tolerancing are there, derived directly from the inspection and gauging situations which inspired and required them.

It is also a gracefully written book (no small task given its thoroughness) and it clearly reflects a lifetime of work by someone dedicated to getting it right, but reasonably.

8. Other Anglophone References in the 1950s

Beyond B.S. 308: 1953 ( see above) and Parker's 1956 book, the 1950s saw the first widespread diffusion of Parker's geometric methods into industry.

It isn't yet clear to me (as I don't have it) if Gilson's 1951 A New Approach to Engineering Tolerances {Gilson 1951} is relevant.

The Society of Automotive Engineers published {SAE Auto 1952}, an automotive version of {SAE Aero 1946} (see {Krulikowski 1991}, p. 4). I have not yet seen either of these.

US MIL-STD-8 of 1949 was updated early in the decade to MIL-STD-8A 1953. I don't actually have this yet, so I can't say for certain that it involved geometric methods. Its fourth-generation successor, MIL-STD-8C 1963 certainly does.

It is interesting to note that Healy and Rau's Simplified Drafting Practice of 1953 says nothing at all of geometric methods in tolerancing, even though it devotes a chapter to tolerancing. {Healy & Rau 1953},

Buckingham's 1954 Dimensions and Tolerances for Mass Production, {Buckingham 1954}, probably is relevant (see the 1956 The Production Engineer review, below). It would be interesting to compare this to his Principles of Interchangeable Manufacturing from three decades earlier {Buckingham 1921}.

Abbott's 1953 The Dimensioning of Engineering Drawings, {Abbott 1953}, seems to have been popular (to judge from the number of secondhand copies still available). It was referenced in the 1956 The Production Engineer review of {Buckingham 1954} (see below) as a "more polished" successor to {Inter-Services 1948}.

A book review in The Production Engineer (UK) in 1956 discussed Buckingham's Dimensions and Tolerances for Mass Production {Buckingham 1954} The reviewer complained that Buckingham covered much more than his title suggested, including also details on gauging. For a more directed study of just dimensioning and tolerancing, the reviewer suggested the "early" {Inter-Services 1948} "or its more polished successor," {Abbott 1953} { PE 1956}

The American standards made the transition from the older Z14 series (which did not incorporate geometric methods in any of their revisions, so far as I can tell) to the Y14 series (with Y14.5 devoted to "Dimensioning and Notes" in its first, 1957, edition). I haven't yet had the opportunity to read it, but the evidence from Giesecke (see below) indicates that it did include geometric methods. Y14.5-1957

Giesecke 4th ed. (1958): The Fourth Edition of Giesecke's Technical Drawing {Giesecke}, notes in its Preface that:

The chapter on Dimensioning has been completely rewritten and expanded into two chapters, "Dimensioning" and "Tolerancing." Both are in accord with the new ASA Y14.5-1957 Dimensioning and Notes . Particulary important is the addition of complete information on geometric and positional tolerances.

But while its acceptance here indicates that by 1958 geometric methods had gone mainstream, the treatment in Giesecke is limited. In particular, the use of datums is rudimentary. He uses notes (as Parker did) rather than symbols. I haven't yet checked this against Y14.5-1957 to see if Giesecke isn't just reflecting the standard as it then was. I suspect, though, that this was the case; even the later MIL-STD-8C of 1963 (which I have checked) does not treat datums systematically as the basis for dimensioning.

The appearance of {Bibeau & Sweet 1959} (which I have not yet seen) would seem to indicate that the name "true position dimensioning" for this field was still current in the US throughout the 1950s.

At the end of the decade, MIL-STD-8 was updated: {MIL-STD-8B 1959}

9. Anglophone References in the 1960s

I haven't really researched this period yet. Sources I've found but not yet examined include:

The change in title from {Foster 1963} to {Foster 1966} (the omission of the "and Positional" from the latter title) is perhaps an indication of the consolidation of the overall name for this field to "geometric dimensioning and tolerancing."

I do have {MIL-STD-8C 1963}, and can affirm that it does indeed cover geometric methods.

10. Historical Studies

There aren't many.

David H. Honsinger. Geometric Dimensioning and Tolerancing - 1946 to 1982: The Differences in the Standards (at the old Watervliet arsenal, I believe; officially it is from the US Army Research, Development and Engineering Center, Close Combat Armaments Center, Benét Laboratories). It is available freely online, and I have reprinted a copy in the bibliography; see {Honsinger 1994}. As noted in the bibliography, it is basically a close technical report on the different interpretations of individual types of features. It is useful also because it associates clearly the contemporary (1940s, 1950s) use of the terms "datum dimensioning," "true position dimensioning," and "geometric dimensioning and tolerancing" with each other. It would have been more useful still if it had contained any bibliographic information at all on the various standards it referenced.

11. Notes on Modern Texts

Foster was a relatively early US entrant into this field who continued to publish commercially for some time. The "Acknowledgments" sections of his works contain a number of interesting references to the literature of the 1950s and 1960s, not all of which have been tracked down here. See (at least) the relatively available {Foster 1970} and {Foster 1986}.

Krulikowski's Fundamentals of Geometric Dimensioning and Tolerancing is one of the few texts of the last quarter century to devote more than an obligatory bullet point to the history of this field. {Krulikowski 1991}

12. Standards Revisions

(Some of this information duplicates that given elsewhere in this Notebook. I find it handy to have a reference all in one place, though.)

B.S. 308

SAE:

ORD 30-1-7 1946

MIL-STD-8

Z14.1 / Y14.5


Select Resolution: 0 [other resolutions temporarily disabled due to lack of disk space]