The trouble with Roundness

One of the geometrical tolerance characteristics is 'Roundness' (or 'circularity' if you are working to ASME Y14.5).

 

The way in which Roundness has always been defined is that it applies to the line element created in any plane perpendicular to the axis of the feature.

 

Thus roundness is measured perpendicular to the axis of the feature, and not perpendicular to the surface of the feature.  This is inconsistent with other geometrical tolerances (e.g. run-out, profile, etc) which are all measured perpendicular to the surface.

While this works out just fine for cylinders (where perpendicular to the axis is pretty much the same as perpendicular to the surface), it does not always work out well for cones.  The greater the angle of the cone, the greater the cosine error in any measurement.

 

For these reasons, a proposal was put forward in the ISO technical committee TC213, several years ago, to change the definition of roundness, so that it would be defined as a tolerance which applied perpendicular to the surface.

 

ADVANTAGES:-

• it makes more sense from a design point of view, when defining cones, as it is more likely to correspond to the type of variation which affects the functionality of the part.
• it makes more sense from a measurement point of view, as it will reduce measurement uncertainty.
• it is more consistent with the definitions of other geometrical tolerances.

 

DISADVANTAGES:-

• it is a change
• it introduces another difference between ISO and ASME
• the transition from the old definition to the new would require managing in industry (changes to working practice, changes to measurement procedures, changes to software, etc).

ISO have flip-flopped over this for a while.  At first the change was going to be introduced, then it wasn't.  Every time it was discussed, it caused a great deal of debate.  Eventually, a new definition for roundness was to be introduced in the 2012 revision of ISO 1101, but at the last minute there was another change of heart.

 

At the time of writing, the plan is to leave the definition of Roundness unchanged.  As a way around this, an alternative proposal has been put forward, which is actually a rather elegant way of tackling the issue.

 

 

The idea is that if you need to control the 'roundness' in a direction normal to the surface, you can specify this by using a Circular Run-out tolerance, with the feature used as its own datum.  Although not an 'illegal' drawing indication under current rules, most experts in this subject would normally discourage this kind of practice, as it creates a 'circular reference'.  With just about any other tolerance characteristic (position, coaxiality, parallelism, etc), this would create a meaningless and uncheckable requirement.

 

Why is it different with run-out?  Run-out tolerances have always covered a combination of geometrical characteristics.  Applying Circular Run-out to a cylinder controls the combined effect of any roundness errors, and any coaxiality errors.  By using the cylindrical feature as its own datum, the coaxiality factors are eliminated, leaving you with a roundness tolerance.  Apply the same tolerance arrangement to a cone, and you have a kind of roundness, but applied normal to the surface of the cone.

 

Some kind of illustration covering this is likely to find its way into a future edition of ISO 1101 (unless people change their minds again).



BS EN ISO 1101:2013 now published.

The latest version of ISO 1101 is now available, having completed its approval as an EN standard, and is now released as a British Standard.

BS EN ISO 1101:2012 delayed

ISO 1101:2012 was published by ISO in April 2012.  You can buy a copy today by ordering it directly from ISO.

For member states of the European Union, the next stage for this standard is for it to be adopted as an EU standard by the European Committee for Standardization (CEN), when it will become EN ISO 1101:2012.

I would have expected this to be a simple 'rubber stamping' exercise, but I have recently been told that this process started in June, and will take at least 5 months. I have no idea why it would take this long - it seems excessive even in terms of the glacial speed at which most standards organisations operate.

Members of the EU, such as the UK, will not adopt this standard until it has been approved as an EN standard. When it is an EN standard, BSI will then adopt it as a British standard with the title BS EN ISO 1101 (just as Germany will adopt it as a German standard called DIN EN ISO 1101, and so on). This process should be fairly rapid.

Due to the very slow movement at CEN, the earliest that the new BS EN ISO 1101 could appear would be very late in 2012, and it probably won't appear until 2013 (in which case it may well be titled BE EN ISO 1101:2013).

Most unfortuately, this means that the CD version of BS 8888:2011, which includes all the ISO standards that it refers to, and which should now be available from BSI, will include the previous version of ISO 1101 (BS EN ISO 1101:2005).

Constituency boundary changes

I don't normally blog about politics, but this radical approach to boundary reform looks very interesting ... New boundary changes

A bit of history - the first atomic clock

The National Physical Laboratory (NPL) recently turned up this old piece of film from the 1950s.

NPL developed the first atomic clock, and this short film explains how it worked.

click here and the film should appear in a new window

NPL is well worth a visit. They hold open days every so often, and have a wealth of really useful information freely available on their web site. They also offer excellent training courses (including our Geometrical Tolerancing course).

New ISO 1101:2012

ISO have just published the new version of ISO 1101, the geometrical tolerancing standard.  The BSI implementation, which will be called BS EN ISO 1101, will appear shortly.

The new standard is mainly concerned with ensuring that all geometrical tolerancing requirements can be applied to 3D CAD models as well as to 2D engineering drawings. Although most of the 2D annotation can already be applied directly to a 3D model without any change of notation or meaning, this is not true of all of it. The standard now includes some additional 3D annotation elements, such as annotation planes and orientation planes, to ensure that all requirements can be fully implemented in a 3D environment.

The new revision also includes some minor changes and clarifications to the wording of the standard, as well as some new functionality.  The use of projected tolerance zones is now incorporated into ISO 1101, instead of a separate standard (ISO 10578), and there is now provision for unilateral and unequal bi-lateral tolerance zones with profile tolerances, using a new UZ modifier.

Explaining Londoners

Picked this up from a tweet from Evan Davis.  A New York Times article explaining Londoners, but much of it applies to Brits anywhere:
http://www.nytimes.com/2012/03/04/magazine/explaining-londoners.html?_r=3&src=tp

A sense of proportion

If you have not seen this before, it is really worth checking out:

http://www.freewebarcade5.net/media/the-scale-of-the-universe-2.swf

You may have to click to skip an ad, but then you just start the app and move the slider - awesome!

Beijing

Trying to catch up with some blogging for the first time in ages.

I had my first trip to China in September last year, with the ISO TC213 meeting series being held in Beijing.  The city is a fascinating mix of the traditional, the modern, and in the case of the traffic, the downright terrifying.  Parts of the city look as if they have remained unchanged for hundreds of years, whereas the efficient underground railway is full of young Chinese glued to smartphones and media players.

We were treated to the standard tourist trip to the Forbidden City, which is as vast as it is impressive, although many of the interiors are off-limits, which is a shame.

Food was interesting.  In the hotels we were paying western prices for food that was often very poor, whereas a few hundred yards away were local restaurants offering fantastic food for a fraction of the price.

The highlight of the trip was a banquet at the Baijia Dayuan ('Garden of the King') restaurant.  This is set in a beautiful courtyard, reputedly several hundred years old, full of pools and flower beds, and illuminated after dusk with a wonderful array of lanterns.  The staff are all in fabulous traditional dress, and put on a variety show of music and dance to accompany the food.  Highly recommended for anyone who gets the chance to go there.

BS 8888: 2011

We (that is committee TDW/4/8 at BSI) have recently finished putting the final touches to the next revision of BS 8888. With the new revision, why have tried to do something significantly different with the standard, which we hope will make it much more useful for industry.

BS 8888 first appeared in the year 2000, when BSI withdrew BS 308 and adopted the ISO system of standards for technical specification.

The ISO system for technical specification is documented in a large number of ISO standards (probably around 250 in total).  At present this is like a large, disorganised technical library, consisting of numerous documents from different authors with very different styles of writing.  Some documents provide a broad overview of a large area of technical specification, while others go into minute detail in a very specialised topic.  Some documents deal with specific manufacturing processes, such as casting and forging, while others deal with general topics such as dimensioning or drawing views.  Some documents have overlapping content, and in many cases, you would have to use two or three documents together to properly understand a particular topic.

ISO are making some efforts to bring structure and organisation to this collection of standards, but this is a slow process, and will take many years.

BS 8888 was introduced to help engineers work with the ISO system, and initially did so by providing an index, along with some additional commentary to help explain some topics.

With the new revision of BS 8888 we are taking this a step further.  There were two fundamental reasons behind our approach to the new revision:

• Industry has in many cases found BS 8888 unsatisfactory to work with, because they expected BS 8888 to be a replacement for BS 308, and to contain the same level of information.  Some poor marketing from BSI early on encouraged this misperception, and the subsequent disappointment felt by many who purchased the standard.
• The ISO system is difficult to work with because of the way in which it is spread over a wide range of different standards, which are interrelated, but often in a fairly haphazard manner.

Our aim within the technical committee has been to use BS 8888 as a vehicle to address both of these problems.

In order to make the ISO system more accessible, we have taken the core content of a number of key ISO standards, such as ISO 1101 and ISO 5459, and incorporated it directly into BS 8888.  In this way, we intend to make the essential elements of the ISO system available together within a single document.

In order to make the ISO system more useable, we plan to use BS 8888 as a filter.  We will include the main elements of the ISO system, which are relevant to most users, but screen out some of the more specialised aspects of the ISO system, which would have very little relevance to most engineers (and would only tend to confuse matters).  BS 8888 will continue to index and reference all these ISO standards, but the aim is to maintain a central body of content which provides access to the fundamental elements of the system.

The 2011 revision of BS 8888 is currently scheduled for publication in December.  It is not perfect, but we hope that does mark a significant step forward in terms of the usability of both BS 8888 itself and the ISO system.

I will lay out some of our plans for the further development of this standard in a future post.

ISO 8015 & ISO 14405

First the bad news.  With a couple of minor tweaks, none of which addressed the fundamental objections put forward by the UK and USA, ISO 8015 has now been approved.  The UK, USA and Canada all opposed this, but it was passed with a comfortable majority of other nations voting in favour.  The invocation principle, which requires all mechanical engineering specifications to be interpreted according to ISO GPS rules unless an alternative standard is specified, is now enforced.  The voluntary nature of standardisation, which has always applied in the past, has in this case been removed.  You no longer have to opt in to the use of this standard, you now have to opt out, which places a responsibility on anyone involved with mechanical engineering specifications to be aware of the standard and its rules.

With the new version of ISO 8015 now published, much of the content of the previous ISO 8015, dealing with size and the envelope requirement, has been moved to another standard, ISO 14405-1.

ISO 14405-1 is the first standard to really address the whole issue of size properly, and gives a thorough, and long overdue, breakdown of the different ways in which size can be defined.  It also provides tools which enable size to be defined in different ways on a specification when necessary.  In the absence of any other indication, the previous definitions of size still apply by default.

This is a good standard, which brings clarity to an area of specification which is often poorly understood.

There is also in the pipe-line an ISO 14405-2 standard.  This is really concerned with explaining how a specification will inevitably be ambiguous if datums and geometrical tolerances are not used.  The UK felt that this content should be produced as a technical report rather than a standard, as it is really just an informative document, but again were over-ruled.  However, the content is perfectly valid, and may be of interest to anyone who is not convinced of the necessity of using geometrical tolerancing to specify products.

Multitasking

An interesting article here, suggesting that while we are all under constant pressure to multitask more and more, it may not be a good idea.

http://www.pbs.org/wgbh/pages/frontline/digitalnation/interviews/nass.html

see what you think ...

Symbols

Mas'ood asked about some of the symbols I mentioned in the last post.  I'm working on a document to publish here which will show where all the different symbols are to be found in the ISO GPS system.  Real soon now!

Paris and Helsinki in February

February was a busy month for travel.  The ISO TC 213 conference was held in Paris, so I was there for five days.  At one meeting, with a fairly tight schedule, it was tentatively suggested that lunch could be restricted to 50 minutes, but the French delegation vetoed this idea immediately.  They take their lunchtimes seriously over there.

I also had to deliver a geometrical tolerancing course in Espoo, just outside Hesinki in Finland.  Helsinki in February rarely gets warmer than -20C, so that was an interested experience.  Many of the cars have cracked windscreens over there, which I assumed was due to the extreme temperatures.  However, locals told me that it is nothing to do with the temperature, but due to spikes from snow tyres being flung up from the road.  They went on to point out that you never see motorcyclists in Finland in the winter!

The picture on the left shows the view from the back of the training venue.  This is on the coast, and between the camera and the trees in the distance is an inlet from the sea, but completely frozen at this time of year.  Just behind the 'stop' sign on the left of the picture, you may just be able to see a boat frozen in.

On one evening we went to the Zetor restaurant in Helsinki, which has an unusual kind of farming theme.  Some of the tables are built onto old tractors.  Farming implements cover the walls.  Old buckets are used for lampshades.  The food was very good.

Common Zone modifier

First post since November! It has been a busy few months, but I must try to do better than that.  I'll put up some more posts about what has been going on shortly.

Mas'ood occasionally sends me interesting (or awkward, depending on your point of view) questions about geometrical tolerancing.

I am going to put some of his questions here - feel free to comment if you have a view.

First question is about the use of the 'common zone' modifier, CZ.

This is placed in a geometrical tolerance frame to indicate that two or more features have to satisfy the requirements of a single tolerance zone.

CZ used with a flatness tolerance

Mas'ood's question was whether the CZ modifier can be used with the envelope requirement.

Personally, I don't see why not.

ISO devised the CZ modifier with a view to it being used in the tolerance frame only, but as long as the meaning is clear, I see no problem with applying it to the envelope requirement as well.  It would be a way of indicating that two features-of-size, of the same size and in alignment with each other, should be treated as a single feature.

CZ applied with the Envelope Requirement to two features

I would feel less comfortable with this approach if it was being applied to two features which were produced with separate machining or manufacturing operations, but we are well into the realm of 'designer's descretion', as the standards provide no guidance in this area.



CZ applied with Envelope Requirement to two features produced with separate operations



ISO have developed several different symbols in recent years, such as Common Zone (CZ), Separate Zone (SZ), United Feature (UF), Contacting Feature (CF), etc, and are currently reviewing these to see if they can minimise this 'symbol inflation'.  The 'UF' symbol was to be used to indicate that several features were to be treated as if they were a single feature, but ISO TC213 will be reviewing whether the CZ symbol could be used as an alternative, or even remove the requirement for the UF symbol altogether.

FDIS 8015 gets negative vote

The Invocation Principle - see previous post - was part of the latest revision of ISO 8015.  The FDIS (Final Draft International Standard)  of 8015 has just recently been out for vote, and the results are back.

In order to be approved, two thirds of member organisations must vote in favour of the FDIS (66.66%), and no more than 25% of members must vote against.  The actual results were 65% in favour, and 32% opposed.  The FDIS has thus received a negative vote, and has been 'disapproved'.

Phew!  But far too close for comfort!

The Invocation Principle

How do you know which standards govern the interpretation of a specification, such as an engineering drawing?  That is the problem that this principle is intended to address.

When we had our old British BS308 standard, it was easy. You were asked to put a note on the drawing along the lines of ‘DRAWN TO BS 308’. With the new British Standard, BS 8888, we suggest a similar note, such as ‘CONFORMS TO BS 8888’, and the Americans require that drawings produced to the American standard show a reference to that standard, ‘ASME Y14.5-2009’.

ISO have never had a formal method of indicating that drawings or specifications conform to ISO standards. One of the key ISO standards is ISO 8015, which defines the Envelope Requirement and how size tolerances are interpreted. ISO 8015 itself contains the requirement to mark specifications ‘TOLERANCING ISO 8015’ if they are to be interpreted according to its rules. As a result, this marking has become the way in which many organisations indicate that they are working to the ISO system, and is even now a requirement of BS 8888.

Arguably, specifications which are to be interpreted according to the ISO system should list all the other applicable ISO standards on the specification as well, or at least list them in a document referenced by the specification. For most organisations, this would be a lengthy list, including references to standards such as ISO 1101, ISO 2768, ISO 5458, and many others. Unsurprisingly, many organisations don’t bother.

The invocation principle was intended to avoid this difficulty, by introducing a rule which stated that if you are using one part of the ISO system, you are using all of it. This seems sensible to me. This approach does not require you use bits of the ISO system that you were not using previously, it just means that if you make use of any particular part of it, such as surface texture specifications, then all the relevant ISO standards apply. You don’t have to list them all, you just have to have some means of showing that one part of the ISO system applies, such as the ‘TOLERANCING ISO 8015’ indication.

The contentious stuff starts with a second part to this rule, which states that use of any ISO GPS symbol is sufficient to indicate that you are working to the whole ISO GPS system.  In other words, you don't need to state ‘TOLERANCING ISO 8015’ or anything like that, you only need to use a geometrical tolerance, or a boxed dimension, or some other GPS symbol, and that makes it an ISO GPS document unless otherwise stated.

This is a REALLY BAD IDEA.

Reasons why:

1. In the UK, standards have a voluntary status. Their use cannot be imposed, and they work on an 'opt in' basis. This requirement would mean that many ISO standards would be imposed, quite possibly without the document owner being aware of this, unless they specifically 'opted out'. A requirement which is enforced by the absence of any alternative marking is liable to lead to confusion, misunderstanding, and resentment (and bring the system into disrepute).

2. ISO do not have ownership of all the symbols which are used in ISO GPS standards. Many symbols are shared with ASME, and many national standards organisations have used them in national standards which are still current.  Some symbols are even used differently in other ISO standards

This requirement would mean that if someone forgot to put ‘ASME Y14.5’ on a drawing, it would then be regarded as an ISO GPS drawing, irrespective of the intention. It means that a sketch on the back of an envelope, which happened to include a dimension with a box around it, would suddenly be subject to the rules and requirements of an ISO GPS standard. It means that civil engineering or architectural drawings with tolerances might become subject to ISO GPS rules and requirements, without the relevant professions even being consulted.

3. As many of the symbols which ISO regards as ‘GPS symbols’ are either in the public domain, or widely used in different ways, in different contexts, an attempt to impose a requirement like this without consulting as widely as possible (other ISO technical committees, professional bodies, industry organisations, etc, etc,) is simply irresponsible.

4. A further consequence is that we would then need to introduce an exemption symbol or marking, such as ‘NOT AN ISO GPS SPECIFICATION’ for cases where it didn’t apply, at which point it is starting to get a bit silly.

5.  What happens in the case of a nation which is not a member of ISO?  What happens if ASME introduce a rule saying 'in the absence of any other marking, a technical specification is to be regarded as an ASME Y14.5 specification'?

The imposition of such a requirement would benefit no one, and simply lead to confusion, and quite possibly to litigation.  ANSI have made their objections well known within ISO, but a number of people from other national bodies seem determined to force this through.  I'm not clear on their motivation, but at least in part it seems to stem from a fear of 'Americanisation', and a desire to give ISO a higher status as the 'ultimate default'.  These motivations may be sincere, but they are misguided.  Industry will not benefit from this, and it will actually lead to the diminishment of the reputation of ISO standards if it goes through.

Prague in September

Painfully overdue for a post here, so here is a quick up-date, to be followed by something more detailed in due course.

The latest TC213 conference met in Prague, capital of the Czech Republic, early in September.

Prague is a rather beautiful city.  The old town has many famous landmarks, and it is clearly popular with tourists.  The place was absolutely heaving, with large gaggles of people following tour guides wherever you went.  Most of these groups were equipped with ear pieces, so they could hear the tour guide above the general din.  Most of them also seemed to be following the lead of the guide in singing 'Good King Wenceslas', with varying degrees of tunefulness, as they marched around.

Tributes to Franz Kafka are also to be found throughout the city, such as the statue of The Man Who Wasn't There (see left).

I like the name of this bar as well (see below).

This was a very brief visit on my part.  I flew in on Wednesday evening, chaired a meeting on Thursday, and flew out again on Friday.

The hot topic of debate was the 'invocation principle' which has been proposed for a revised draft of ISO 8015.  Quite what this is, I will explain in the next blog post.  The American delegation is outraged by this, and circulated a letter listing their objections prior to this series of meetings.  Some other nations are strongly in favour, and one German delegate in particular has been mounting a charm offensive to cultivate support.  The UK delegation is strongly opposed to this new principle, supporting both the American objections, and a number of our own.  Many national representatives are remarkably ambivalent over the whole issue, considering its potentially far reaching effects.  As I say, more on this to follow ...

New Courses

It has been a busy summer so far.

We have developed and piloted a new Stress Analysis course, introducing the basic principles of stress analysis for calculations with calculators and spreadsheets.  The idea has been to give engineers and designers the tools to carry out basic calculations for stress, strain and deflection in fairly straight-forward structures.  This is really covering a lot of the same ground that engineering students cover at degree level, but compressed into three days, with a strong emphasis on practical application.

We had hoped to compress this into a two-day course, but that proved overambitious.  We will be trialing the latest version of the course with the client who requested it within the next few weeks.   Once all the creases have been ironed out, we will be publicising this with the other courses on our main web site.

When running the geometrical tolerancing course for one of our main aerospace clients, we frequently get questions and feedback comments about datums and tolerances applied to assemblies.  There is not time to cover this adequately in the standard course, so we have proposed working with the client in questions to develop an advanced, follow-on geometrical tolerancing course.  This new, advanced course would look in particular at assembly datum structures and assembly-level tolerances.  If any other organisations out there (particularly aircraft manufacturers) would be interested in collaborating with us on the development of such a course, please get in touch.

First NPL GD&T course

The first NPL GD&T course was held in the splendid setting of Bushy House in Teddington, where the National Physical Laboratory began over a hundred years ago.  This is not far from Heathrow Airport, but you would never know it.  Surrounded by the extensive grounds of Bushy Park, including cricket pitches, orchards and deer parks, we felt a bit like extras on a Jane Austin movie set.

This was one of the views from the museum room where the course was held.  There are a herd of deer out there somewhere, but you can't see them in this picture.

The museum room itself was full of old bits of measuring equipment mapping some of the history of NPL since 1900, although some of the equipment looked older.

We had a great mix of people on the course, including three from Scotland, three from Ireland, and one from the Isle of Wight.  They represented an equally diverse range of industries, including aerospace, boat building and medical equipment.

Feedback from the course has been highly positive, and we hope to run the next one in August or September.