Three days in San Antonio

The latest ISO TC213 conference was held in San Antonio, Texas. Other commitments meant that I was only able to fly in for three days of the conference, which lasted about 10 days altogether.

I flew from Manchester via Atlanta, a most efficient airport, where it took no more than about 30 minutes to pass through immigration, collect luggage, redeposit luggage, pass through security and make my way to the correct terminal and gate for the next leg.

I arrived in San Antonio, which had been enduring weeks of heat wave recently, to find it raining. Their first proper rain for 18 months, timed to greet a visitor from Manchester! It rained on each of the following three days as well.

Most of the trip was spent inside the Holiday Inn, where all the delegates were staying, and all the meetings were being held. I was in meetings from 9:00am to 8:00pm on the first day, and 9:00am to 6:00pm the second day, so I did not see a lot besides the inside of the hotel until day 3.

San Antonio itself is smaller than I expected, as American cities often seem to be. I expect them all to be huge, and most of them aren't. According to Wikipedia, its population is around 2 million (smaller than Manchester), and it is still the seventh largest city in the USA. Rain showers each morning cut the humidity to comfortable levels, and left it pleasantly warm and even fresh while I was there.

We saw the Alamo, of course, which is close to the centre of the city.

Also in the heart of the city is Riverwalk: one storey below street level, there are walkways along either side of the San Antonio river, where there are clustered a vibrant collection of bars, shops and restaurants. We strolled along here each evening to find somewhere to eat, and always ate well.

My flying visit was quickly over, and then it was back to Manchester (which had enjoyed some of its sunniest weather in my absence).

The next TC213 conference will be in Vienna in 2010, which I will look forward to. I just hope that the ISO habit of scheduling meetings to run all evening as well as all day will still leave a little time to see the city.

Next BS 8888 now scheduled for 2011

Been very remiss with postings recently. Somewhat shocked to see the last one was on 1st July! My only excuse is that I was away for half of July, and spent the other half doing some major updates to training course material (lots and lots of new diagrams, which always take ages to get right).

Anyway, we had a meeting last week of the BSI committee looking after BS 8888 with a view to getting preparation of the next revision (then scheduled for October 2010) underway.

We had a useful review of the current edition, and are making plans to restructure the content (to make the information more accessible), add more diagrams and illustrations, and to cull the annexes, which now make up almost 75% of the document. These annexes have been added over a period of time, with intention of providing additional explanation and useful background information, but they are getting a bit out of hand. Some will stay, and may even be extended, but some will be incorporated into the main body of the standard, and some are no longer necessary and can be removed altogether.

We then turned our attention to a number of forthcoming ISO standards and amendments, which I have already mentioned on previous posts. These are bringing through some significant changes and additions to the system of technical specification. The justification for the next revision of BS 8888 lies in providing coverage of the changes and developments that these new standards and amendments will bring.

- ISO 1101 amendment 1 is bringing in the additional annotation elements required for fully annotating 3D CAD models.

- ISO 1101 amendment 2 is introducing additional symbols, bringing in some new, and rather technical functionality to geometrical tolerancing.

- ISO 5459 is a complete overhaul of the definitions and rules for datums.

- ISO 14405 is a new standard on linear sizes, which will replace ISO 8015.

In each case, we reviewed the timetable for the new document - how many comment, discussion, amendment and voting stages each has still to go through, and the likely timescale for that to take place. Our conclusion was that there is a faint chance of ISO 1101 amendment 1 being ready for inclusion in a 2010 revision of BS 8888, but that none of the other will see the light of day before 2011 at the earliest.

While we could bring out a 2010 revision of BS 8888 with tweaks, adjustments and improvements, there is no technical justification for a revision of the standard until these new technical changes and developments can be included. I am pleased to say that we have decided to delay a revision of BS 8888 provisionally to 2011. Pleased, because I have wanted all along to make BS 8888 a more stable standard that changes less frequently.

Our decision has also been influenced by the fact that some of these future developments are still fairly contentious, and have a rocky road ahead of them before they are finalised. In several cases, the new standards are introducing new possibilities and symbology to address very specialized and rarefied applications, and they run the constant risk of making the whole system horribly complicated in order to address issues which are of largely academic interest.

The UK representatives on the ISO committees have always sought to avoid unnecessary complexity, and to take a pragmatic, working engineer's approach to these new developments, but we don't always manage to get our own way. Amid the complexities, however, there is also much genuinely useful work taking place. Some of these new standards are now clearly defining rules and principles that have previously often been implied rather than stated, and rarely properly understood. Our challenge is to introduce these in a way that is practical and useable for industry at large.

In just over 4 weeks I shall be going to Texas for the next ISO TC213 conference (having finally managed to find some affordable flights), and will report back on how things develop there.

Catching up

I was away training last week. The course went very well indeed, but I'm always a bit drained afterwards. Friday night was then spent at a charity ball/auction to raise funds for my childrens' school - a terrific night out at The Lowry in Manchester, but a very late finish.

Saturday was when Friday night caught up with me. Sunday was a fund-raising sponsored walk (the childrens' school again - who schedules this stuff?) with me trying to catch up with the kids. And this week has been mainly catching up with admin so far.

I have a meeting coming up next week at BSI for TDW/4/6, the committee dealing with education. We have been slowly working towards a fully accredited training scheme in Geometrical Product Specification for some time now. The IED are going to be the accreditation body, but we still have a long way to go in sorting out the details of how the process will work, how it will be funded etc. I need to catch up with some outstanding actions on structuring the curriculum.

In TDW/4/6 we have learnt a great deal from the National Physics Laboratory (NPL) who have had a system of accredited training for metrologists in place for sometime now. When I catch up with everything else, I'm planning to become a deliverer of the NPL training. It is a near-perfect fit with the geometrical tolerancing training I'm providing now.

Something more technical next time ...

ISO 1101 amendment 1

ISO 1101 is the principle ISO standard for geometrical tolerancing. It was last revised in 2004, just five years ago, but there are already a number of amendments queuing up to be incorporated into it.

The first of these, Amendment 1, is concerned with 3D annotation, and is going through the final stages of approval prior to publication. It is likely to come into effect some time in the next 12 months.

What is '3D annotation'?

It is established practice in most industries to use a 3D CAD model as an essential part of a component specification. The 3D model represents the 'nominal' geometry of the part, and as many manufacturing operations, and some inspection operations, can be driven directly from this 3D model, there is no longer a necessity to duplicate this data with dimensions on a 2D engineering drawing.

This has paved the way for the 'minimum content drawing' approach that is now widespread. The 2D drawing is used to convey information about datums, tolerances, surface finish and sometimes inspection requirements, but no longer needs to define the fundamental geometry of the part.

The next stage in the evolution of this process will be to apply the information about datums, tolerances etc directly to the 3D model, eliminating the requirement for a 2D drawing at all. This is what is meant by '3D annotation'.

Eliminating 2D drawings does not really become a practical proposition until everyone involved in the specification, manufacture and inspection of the component can work fully in 3D. For most of industry, that prospect is a distant one, so 2D drawings are going to be around for many year yet.

However, CAD software is increasingly providing the facilities to apply all specification requirements directly to the 3D model, and some manufacturing organisations are starting to work in this way.

The CAD software is currently a little ahead of the standards in this respect. While most 3D annotation requirements can be met quite happily by applying the 2D annotation elements in a 3D environment, there are some situations where this is not sufficient. Some specification elements depend on the positioning of a geometrical tolerance within a drawing view to convey part of the requirement. Working in 3D, we no longer have drawing views to position or orientate the geometrical tolerance, so other means have to be found to convey that information.

Amendment 1 provides some illustrations to indicate how a geomtrical tolerance requirement can be indicated in a 3D specification as well as in a 2D specification. More importantly, the amendment introduces some new annotation elements to enable all geometrical tolerance requirements to be fully specified in 3D. Symbols have been introduces to define 'intersection planes' and 'orientation planes', which can be used to define the location or orientation of a tolerance requirement relative to the 3D model.

The amendment also ties up one or two other loose ends. It will indicate how the 'all around' symbol can be used in 3D (although in my view there is no need to use it in 3D at all). There is also a new symbol to help make it clear when a derived feature, rather than a surface, is being used as a datum feature.

Finally, there will now be the facility to define offset tolerance zones. This last was something of a disappointment for those of us who argued in favour of adopting the American (Y14.5) symbology in the interests of simplicity, clarity and universal harmony. The ISO approach will be different from the Y14.5 approach, but at least we managed to avoid using the same symbol as the Americans', but with a different meaning (a very real possibility at one stage). Sometimes, these tiny, common-sense achievements rank as great victories when developing international standards.

BS 8888:2010

We are now starting to plan the next revision of BS 8888, which is due out in October 2010.

BS 8888 is currently up-dated every two years. I think this is far too frequent, and in principle, I would like the standard to be much more stable, and only up-dated every four or five years. This has been one of my objectives since I took over the chair of TDW/4/8, which is responsible for BS 8888.

However, BS 8888 acts as an interface to the ISO system of standards, and has to be kept up-to-date with changes in the ISO system, and this is what fundamentally drives the revision cycle.

In recent years, there have been a lot of developments within the ISO standards, and BS 8888 has had to change frequently to keep abreast of these changes. Right now, there are some more major changes in the pipeline, and a further revision of BS 8888 for 2010 is unavoidable.

So what changes are on the way? Four major ones at least:-

1. Amendment 1 to ISO 1101 (the main ISO standard for geometrical tolerancing).

This mainly deals with 3D annotation.

2. Amendment 2 to ISO 1101.

This introduces some new symbols, and also extends the system in several ways. For instance, some of the filtration options which can be used with surface texture specifications will be available for form tolerances.

3. ISO 5459

This is a major up-date of the datum standard, which is long over-due.

4. ISO 14405.

This is a new standard for linear sizes. It will replace ISO 8015, and provide a more comprehensive range of options for how we define sizes of features.


I'll give further details about some of these developments in future posts.

Geometrical Tolerancing, standards and related stuff

Does the world really need another blog?

I'm about to find out. This is my first experiment with blogging, and it really is an experiment. If it generates some interest and dialogue, then I will keep it going. If it doesn't, then I'll quietly close it down at some point.

I run my own business. I provide training and consultancy in geometrical tolerancing, engineering drawing, and related issues. I am also involved in the development of the standards which govern this field, through technical committees at BSI and ISO.

The people I work for are mainly in the aerospace and defence sectors, companies like Airbus, BAE Systems, and MBDA.

I also work for organisation like Toyota, Siemens, Thales, several in the Oil and Gas sector, and numerous SMEs across a wide range of industries. My clients include some of the Formula 1 teams, and scientific establishments like the Rutherford Appleton Laboratory - all of them fascinating in different ways.

I will be recording here, on an occasional basis, my thoughts and experiences on running a business, on working with tolerances and specifications, and on developing the British and ISO standards.

This is not going to be a free help-line on tolerancing issues, although there may well be discussion about the standards like BS 8888, ISO 1101, ISO 5459, etc. In part, I hope that this blog will throw some light on how these standards are developed.