A Surface Finish Callout Carries a Date the Drawing Does Not Print

This page does not answer when Rt is used rather than Ra. The standard read here does not govern that choice, and we did not read the parameter standards that might. What it governs is how a requirement gets written down, and on that it has something the original question did not ask for and probably needs. The question came from a public engineering question site through its open interface, the browser tooling this site normally uses being unavailable again. We read the question and not its answers, and no username appears here.

The standard for writing it down, not for choosing it

ISO 1302:2002 is Geometrical Product Specifications (GPS), Indication of surface texture in technical product documentation. Fourth edition, 1 February 2002, prepared by ISO/TC 213, cancelling and replacing a third edition from 1992 that had been technically revised. Its scope is the rules for indicating surface texture in technical product documentation, given as drawings, specifications, contracts and reports, by means of graphical symbols and textual indications.

Notice what that scope does and does not cover. It is about the writing. Which parameter suits which surface is not in it. The nearest thing is Annex D, Minimum indications for unambiguous control of surface functions, and by the contents page that annex is informative, not normative.

By our count the standard has eleven numbered clauses, of which the first three are scope, normative references and definitions, leaving eight clauses of actual rules. It then carries ten annexes, A to J, exactly one of which is normative. Annex A covers the proportions and dimensions of the graphical symbols; everything from B to J is for information only. The clauses run to page 19 and the annexes run from page 20 to page 45, so the informative material is longer than the standard it accompanies.

One symbol, three families of parameter

The scope lists what the indication can carry. It is applicable to requirements expressed by means of:

• Profile parameters, according to ISO 4287, related to the R-profile for roughness, the W-profile for waviness and the P-profile for structural parameters
• Motif parameters, according to ISO 12085, related to the roughness motif and the waviness motif
• Parameters related to the material ratio curve, according to ISO 13565-2 and ISO 13565-3

Ra is one parameter, from the R-profile, from the first of those three families. The same graphical symbol on the same drawing can carry any of them. That is worth holding onto when someone says a drawing specifies surface finish: the symbol tells you a requirement exists, and the letters next to it tell you which of three quite different measurement traditions is being invoked.

A note in the scope adds a boundary that is easy to miss. Requirements for surface imperfections, given as pores and scratches, cannot be specified with surface texture parameters at all, and are referred to a different standard. So a roughness number is not a statement that the surface has no defects. That distinction has come up here before, in a standard that permits a particular kind of crack.

The part with no date on it

Now the introduction, which is where this standard says the thing worth carrying away. It explains that this edition was developed to go with the new surface texture standards issued in 1996 and 1997, and that those introduced “many radical changes compared with the content of the former surface texture standards issued in the 1980s.” Then the sentence quoted at the top: so radical that in some instances the drawing indications have a completely new interpretation.

The following paragraph draws the consequence. An indication applied under a former edition refers to the rules in force when it was issued, and can only be interpreted according to those standards. Annex I, titled Former practice, exists so that a reader can do that.

Read plainly, this means a surface texture callout is not self describing. It carries an implicit date, and the drawing does not usually print that date next to the symbol. Two drawings can bear the same mark and mean different things because of when they were drawn. The standard does not treat this as a curiosity. It devotes an annex to it and says the older indication can only be read the older way.

The introduction closes with a caution about its own contents: “Textual indications in this edition of ISO 1302 are under continuous development within ISO/TC 213 and a separate, detailed standard on this issue is under preparation. Consequently, the textual indications given may change in future editions of ISO 1302.” A standard published in 2002 telling you, in 2002, that part of what it prescribes is expected to move.

The definitions kept moving too

That expectation is easy to check on the neighbouring document, because a free preview of a draft amendment to the parameter standard is available. It is a draft, and we treat it here strictly as one, because its own cover page says in capitals that it is circulated for comment and approval, is subject to change, and may not be referred to as an International Standard until published as such.

The draft, put out for a five month parallel enquiry with voting from 11 April 2013 to 11 September 2013, is titled as an amendment adding two parameters. What it actually contains is a list of replacements: the definition of sampling length, the definition of height and spacing discrimination, which it splits into two, the definition of profile element, and the definitions of the mean height and mean width of the profile elements.

The interesting part is what the replacement notes contain. Both replaced parameters are given explicit thresholds: a minimum peak and valley height of ten percent of Rz, and a minimum spacing of one percent of the sampling length. Parameters that read like plain geometry turn out to depend on where you decide a peak stops counting as a peak, and that decision was being written into the text sixteen years after the 1997 edition.

We could not establish from free sources whether that amendment was published, or what the current status of the 1997 parameter standard is. The ISO catalogue page for it returns an error to a plain request. That is a gap in this page, stated rather than papered over.

Why this lands on a fastener supplier's desk

Surface texture reaches fastener work from several directions at once. It is a bearing surface condition under a head. It is a substrate condition before plating, where what the surface was blasted with changes what the coating does. It is a thread flank condition that sits underneath any friction figure. And it arrives as a symbol on a drawing that somebody else made, possibly a long time ago.

The practical shape of this page is therefore narrow and, we think, useful. If a drawing carries a surface texture indication and no edition reference, the standard itself says the indication has to be read under the rules of its own time, and provides an annex for doing that. Asking which edition is not pedantry invented by a supplier. It is the reading instruction printed in the standard.

The same shape appears elsewhere: a test can be run correctly and still not be the thing the buyer thought they had bought, as with a test the standard itself says is not an acceptance test. Here the document goes further and tells you the symbol can change meaning without changing appearance.

What this settles and what it does not

• The standard covers how surface texture is written down, in drawings, specifications, contracts and reports, by graphical symbols and textual indications
• The same symbol can carry three different families of parameter: profile parameters, motif parameters, and parameters from the material ratio curve, each from a different standard
• Indications made under former editions can only be interpreted under the standards in force when they were issued, and an informative annex on former practice exists for that purpose
• The 1996 and 1997 revisions were described by the standard as so radical that some indications acquired a completely new interpretation
• Ten annexes, one of them normative, and by our count the annexes occupy more pages than the clauses
• A 2013 draft amendment to the parameter standard rewrote five definitions and wrote explicit thresholds into two of them, ten percent of Rz and one percent of the sampling length. It is a draft and says it may not be cited as a standard
• We could not confirm the current status of the parameter standard from free sources, and we did not read it, nor the parameter standards for motifs or the material ratio curve
• Nothing here says when to specify Rt rather than Ra. The standard read for this page does not address that choice

The question started at which parameter. The document underneath it answers a different question first: whichever parameter you write, the reader has to know which edition you meant, and the symbol will not tell them.

This page covers step 6, the documentation. The whole order is substrate, thread, head, drive, finish, documentation, and why doing it out of order is rework rather than a tweak is in specifying a screw.

Common questions

Does a surface finish symbol mean the same thing on every drawing?

Not necessarily. The introduction to the indication standard says that indications applied under former editions refer to the rules given in the surface texture standards issued at the time, and can only be interpreted according to those standards. It also states that the 1996 and 1997 revisions were so radical that in some instances the drawing indications have a completely new interpretation. An informative annex titled Former practice is provided for reading older indications.

What is ISO 1302?

Geometrical Product Specifications, Indication of surface texture in technical product documentation. Fourth edition dated 1 February 2002, prepared by ISO/TC 213, replacing a third edition of 1992. It specifies rules for indicating surface texture in drawings, specifications, contracts and reports by means of graphical symbols and textual indications.

Does it tell me which roughness parameter to use?

No. It governs how a requirement is written, not which requirement to choose. The closest thing in it is an annex on minimum indications for unambiguous control of surface functions, and by the contents page that annex is informative rather than normative.

Can the same symbol carry more than one kind of parameter?

Yes. Its scope covers profile parameters according to one standard, covering roughness, waviness and structural profiles; motif parameters according to another; and parameters related to the material ratio curve according to two more. Ra is one parameter from the first family.

Does a roughness value tell me the surface has no defects?

No, and the scope says so. A note states that requirements for surface imperfections such as pores and scratches cannot be specified using surface texture parameters, and refers them to a separate standard.

How much of the standard is mandatory?

By our count it has eleven numbered clauses and ten annexes, A to J, of which only Annex A is normative. Annex A covers the proportions and dimensions of the graphical symbols. The clauses run to page 19 and the annexes from page 20 to page 45, so the informative material occupies more pages than the clauses.

Have the parameter definitions themselves changed?

A draft amendment to the 1997 parameter standard, circulated for a five month enquiry with voting from April to September 2013, replaced the definitions of sampling length, height and spacing discrimination, profile element, and the mean height and mean width of the profile elements. It is a draft and states on its cover that it may not be referred to as an International Standard until published as such.

What did that draft change about the thresholds?

It wrote them into the text. The replacement notes for the mean height and mean width of the profile elements give a minimum peak and valley height of ten percent of Rz and a minimum spacing of one percent of the sampling length. Those parameters therefore depend on a discrimination setting, not on geometry alone. This is reported from a draft, not from a published requirement.

Is the 1997 parameter standard still current?

We could not establish that from free sources. The ISO catalogue page returns an error to a plain request, and we did not read the parameter standard itself or any later replacement. This page makes no claim about its status.

So when should Rt be specified instead of Ra?

This page does not answer that. The standard read here does not govern the choice of parameter, and the standards that discuss parameter selection were not read.

References

• ISO 1302:2002, Geometrical Product Specifications (GPS), Indication of surface texture in technical product documentation, fourth edition, 1 February 2002, read from the freely previewed national adoption of the European version. Fifteen pages covering the contents, foreword, introduction and clause 1
• ISO 4287:1997/DAmd 2, draft amendment, voting 11 April to 11 September 2013. Free preview, nine pages, including the replacement clauses
• The question, on a public engineering question site. Four votes, one answer

Both previews were downloaded complete, and the contents page, the introduction and clause 1 were each checked against rendered page images. This is the first time this site has used either document, although Ra appears on twenty four of our existing pages and roughness on fifteen, which is part of why this one was written. The draft amendment is treated throughout as a draft, because its cover states that it is circulated for comment and approval, is subject to change, and may not be referred to as an International Standard until published as such. Nothing from it is presented as a current requirement. We could not establish the current status of the 1997 parameter standard or whether that amendment was published, because the ISO catalogue page returns an error to a plain request; that gap is stated in the body rather than filled by inference. The following are our own counts, not statements by the standard: that there are eleven numbered clauses of which eight carry rules; that there are ten annexes of which one is normative; and that the annexes occupy pages 20 to 45 against pages 1 to 19 for the clauses. Clauses 4 to 11 and the text of every annex were not read, only their titles and page numbers from the contents page. The parameter standards themselves, including those for motif parameters and the material ratio curve and any later replacement, were not read and nothing is summarised from them. No roughness value, parameter recommendation or measuring instrument brand appears here, and this page does not say when Rt should be specified. The question came from a public engineering question site rather than the forum tooling this site normally uses, which remains unavailable. We read the question and not its answers, and no username appears here.