New working method for the drafting, approval and publication of the amendments to EN 13445 (unfired pressure vessels) and EN 13480 (piping)

It is now more than 30 years that I have the honour of being the president (sorry! I meant the coordinator) of Working Group 53 (Design and Calculation) of CEN Technical Committee 54, which is in charge of the most important of all the harmonized standards of the PED Directive: EN 13445 (Unfired Pressure Vessels).

When, back in 1990 (the PED directive was still in preparation), the Commission gave to CEN the task to draw up the harmonized standards of the new directive, we had a choice that was certainly not easy: we could slavishly copy one of the many European codes that deal with this subject (the British BS 5500, the French CODAP, the Italian VSR, the German standards AD-Merkblätter, the Dutch Stoomwezen, if we want to stay in Europe; without forgetting the American ASME standards, certainly the oldest in the world); or we could create something completely new, trusting in the fact that our group included the best brains of the small (but not too much small) world of European pressure vessels.

The decision to follow this second and certainly more difficult path was not taken lightly: what played a decisive role was the existence, in our field, of different schools of thought: the most important of which was largely linked to the American standards, which in the past had influenced the French CODAP, the Italian VSR, and, to a lesser extent, the British BS 5500: these standards tended to favour the use of high thicknesses; but there was also a German school, much more modern than the American one: this school was certainly winning in terms of costs, since, allowing higher allowable stresses, based the safety of the vessels on the greater use of non-destructive tests and more accurate qualification procedures both for welding and for all other operations (forming, heat treatment, etc.).

The synthesis of these different schools of thought was not easy at all: to this we must add the fact that the vastness of the topic made it practically impossible to entrust all the work to a single group of experts: therefore the various clauses and annexes of part 3 were entrusted to individual experts of different nationalities, certainly very well versed in their field, but not always equally versed in the use of the English language (which was initially chosen as the only working language), and not even in Bill Gates’ software. The result was (and the latest version 2021 of part 3 of the standard, “Design and Calculation” is still there to prove it) that there was no common imprint, both in the style of the text and in the presentation of the arguments, of the figures, equations and graphs; moreover, at the beginning we were forced to limit the work to only some materials (steel and nodular cast iron) and to skip important topics, such as design in the field of high temperatures. Moreover, precisely because of the difference in the schools of thought in which the various experts had trained, we often found ourselves embroiled in interminable discussions (sometimes more philosophical than technical): discussions that, however, if on the one hand made us waste months of time, on the other hand they forced each of us to deepen all the topics covered: so I can honestly say, without fear of denial, that I have learned more in these last thirty years of professional life than I had ever learned in the previous years.

Therefore, the first edition of part 3 of the standard (2002) was largely incomplete: this caused, from the beginning, the need for continuous fine-tuning, which was made through the issue, in a continuous stream, of subsequent modifications (precisely the amendments). Among other things, the above defects were the first (but not the only) cause of the difficulty of this standard to establish itself, given that there were still other standards available on the market that could guarantee (in principle) compliance with the essential safety requirements of the Pressure Equipment Directive: in fact, it should be remembered, the PED never obliges the use of a detailed standard, but merely requires that the standard used by the manufacturer conforms to the aforementioned requirements, guaranteeing however, at least in theory, the so-called “presumption of conformity” with the Directive only to the harmonized EN standard. But there was another factor that made the affirmation of EN 13445 difficult: that is the error committed by the politicians of the Commission, who counted on the disappearance of the old national rules only for the fact that, according to the constitution of CEN (European Committee for Standardization), the national standardization bodies are obliged to withdraw the corresponding national standard every time a new EN is approved: it is a pity that, in almost all countries, the standards on pressure vessels were not drafted by these bodies, but by other entities, certainly not at all obliged by the CEN statute: the association of manufacturers (France), the national surveillance body (Italy), the consortium of inspection bodies (Germany). Even in those countries (United Kingdom) where it was precisely the standardization body (BSI) that drafted the British code, the lack of completeness of the first edition of the harmonized standard allowed BSI to obtain a derogation, with the simple condition of changing the name of its standard: no longer BS 5500 (British Standard), but PD 5500 (Published Document). With the result that practically all of the European calculation codes have continued to change and evolve, perhaps taking as a model the new ideas that were the basis of EN 13445, and often thanks to the work of the same experts who, at TC54, worked on EN 13445; so that in the end, in every single country, the old national standard was in fact able to guarantee the same presumption of conformity to the Directive as the harmonized standard. Added to this is the fact that the Commission’s funds were soon cut (around 2005), which transformed the aforementioned experts into volunteers who barely manage to bring not just a fair salary, but in many cases not even travel and subsistence expenses for attending meetings.

If nothing else, this situation has improved with the spread of the current Covid 19 pandemic, which has forced us to replace face-to-face meetings with online meetings: these meetings obviously do not involve travel and accommodation expenses, and therefore allow participation of a greater number of experts (well, at least some advantages the coronavirus has brought us).

However, something is changing: under the pressure of large users and engineering companies, thanks to this continuous refinement of the standard that has made it more and more complete, the use of EN 13445 is increasingly generalizing in Europe: and this has therefore happened thanks to the continuous issue of changes and updates, precisely the so-called “amendments“. Surely part 3 of the standard was the one that generated the greatest number of amendments: which obviously caused confusion. In practice, up to the last 2021 edition, the system worked like this: each new amendment, once approved, was published independently of the existing text, so it was immediately applicable, even in case that it could have been partially or entirely in contradiction with it (of course you had to know this, and not everyone knew it). It was then necessary to wait for the next issue of the standard, which every year consolidated in the text all the amendments issued after the publication of the previous issue. A completely new edition of the standard was then prepared every five years. Since each of the standardization bodies used to indicate the new issues with the year in which they were published in their own country (a year in many cases subsequent to that of publication by CEN), this constituted a further reason for confusion. Added to this is the significant complication of the procedure, valid both for the issue of a new standard and for that of an amendment:

  • The TC decides to open a Work Item (by voting within the TC).
  • The WI can be «preliminary» (3 years to make it formal).
  • Within 8 months the draft of a formal Work Item is drawn up in English in the TC or in its groups (WGs: it is possible to request an extension of another 9 months).
  • The draft approved by TC / WG is translated into the other two official CEN languages ​​(French and German).
  • After the preparation of the translation, the Public Enquiry is launched, in which the standardization bodies members of CEN accept or reject the standard with a weighted vote (weight proportional to the number of inhabitants of the country) and any comments are recorded.
  • There are 4 months to process comments (non-accepted comments require justification).
  • The Formal Vote is launched on the modified document updated with the accepted comments (also weighted). It can be avoided if no comment has been accepted and the outcome of the Public Enquiry was positive.
  • During the processing of the standard, the “Standstill Agreement” is triggered (Suspension agreement = no work must be carried out at national level on the subject covered by the European standard in preparation).
  • Once the EN standard approved in the formal vote has been published, the CEN member bodies undertake to withdraw any corresponding national standards that may conflict with it.

Lately, the Commission, in order to ensure the compliance of each harmonized standard with the essential safety principles of the Directive, has demanded the intervention of a “HAS Consultant“, who is also entitled to comment on the drafts of the new standard, or of the new amendment: which obviously caused further delays.

In an attempt to simplify and speed up the procedure, starting from the 2021 edition of the standard, CEN has inaugurated a new working method. As I will try to demonstrate below, however, there are considerable doubts that the new working method envisaged by CEN can bring advantages compared to the current situation, at least as regards part 3 (Design and Calculation) of EN 13445.

In practice, according to the new method, TC54 is required to limit the number of amendments to no more than one per year for each part of EN 13445, grouping in that single amendment all the changes of all the clauses of each part: the experience of past editions has however shown that this is not possible in a working group like WG53 which manages such a vast topic: WG53 is in fact divided into several subgroups: we have a “Design Criteria” Subgroup (General criteria of design), a “Non-Pressure Loads” Subgroup (loads other than pressure) and a “Tubesheets” Subgroup (dealing only with heat exchanger tubesheets): furthermore, as already mentioned above, each subgroup includes experts in charge of handling only one clause, or even to edit single parts of the same clause. These groups (and in particular the SG “Design Criteria”), having to deal with different topics, are therefore forced to work independently: therefore, collecting all the changes involved in the various subgroups once a year in a single amendment would practically cause the stoppage of the works, or at least of some of them; to this we must add the fact that clause 19 of part 3 (Creep Design – design in the field of creep) is under the responsibility of another working group, WG59, which deals only with this topic.

Furthermore, if we consider that, in the past, many changes to the standard have arisen from users’ comments, I do not see how it is possible, with the new procedure, to immediately proceed to correct those errors that are not simply printing errors, and therefore need to urgently carry out an unscheduled technical update: corrections of this type, with the new system, risk being postponed for years. Please note that this idea is by no means shared by the CEN standardization bodies, which, on the contrary, really appreciate this new procedure: obviously for a standardization body it is much better to issue fewer amendments, moreover, scheduled not so frequently. I simply wish to note, in this regard, that the complication of the work on Part 3 experienced in the last two years is mainly due to the need to deal with the comments expressed by the HAS Consultant on the various amendments under discussion: this has de facto delayed the approval of many of these. Just to give an example, think of the deplorable story of the alleged non-compliance with the Directive of the use of higher nominal design stresses for austenitic stainless steels having a rupture elongation ≥ 35%, where the same directive stated > 35% (greater then, and not greater than or equal): this (apparently) negligible detail caused more than a year of delay for the new edition of the standard, scheduled for 2019, and published only in mid-2021. This delay was precisely due to the need to resolve this single comment: a battle that the entire WG53 had to fight to save the competitiveness of the standard, which, at least in the case of stainless steels, would have been irremediably compromised, if we consider that 35% is exactly the rupture elongation provided for in the specifications of many austenitic steels commonly used by industry: acceptance of the comment would therefore have caused an increase in thickness (and therefore in weight) of all austenitic stainless steel vessels, with the additional consequence of making it impossible to modify the vessels designed with the previous edition of the standard, whose thicknesses, if recalculated with the new edition, would now have been insufficient. It should be noted that the battle was not against the consultant of the Commission, who in the end had only done his duty, but against the management of CEN: in fact, in order to stop discussions, CEN had decided to vote on the approval of the comment and the consequent modification of the standard, making it known that failure to accept it would have resulted in the loss of the presumption of conformity with the Directive of the entire standard. The vote, under the shadow of such a threat, was obviously resolved in favour of the HAS Consultant. It was only thanks to the intervention of the Commission’s Desk Officer, who in the end accepted the presentation of a document demonstrating the substantial equivalence, from the point of view of safety, of the formulation provided for in the standard with that provided for in the Directive, that we laboriously managed to solve the problem.

Considering this unfortunate precedent, one can now easily imagine what would happen to our single annual amendment if a similar stalemate were created only for one of the various topics concentrated from now on in the same amendment, to be launched all together in one single public enquiry: in the presence of the failure to resolve a single comment, the entire amendment would be stopped, thus also delaying the correction of the other arguments, possibly containing technical errors already recognized or necessary clarifications and improvements.

It goes without saying that this is just my personal opinion: it will only be the experience of the first years of work that will tell us whether my concerns are well founded or not. For the moment, the practical result of this approach was that in the last session of TC54 it was decided to group in a single document all the amendments of part 3, relating to the previous edition 2014, which had already passed the public enquiry and the revision of the comments, and for which only the formal vote was missing: the fact is that this made it impossible to include the same amendments in the first Issue of the 2021 edition. Therefore, if the formal vote is organized shortly, and if obviously the same ends with a positive outcome, these amendments will see the light this year, with the second Issue of the 2021 edition (a good record, considering that some of these were under discussion since the end of the now distant 2017).

January 24th, 2022

Fernando  Lidonnici

                                                                                                                 (Convenor of WG53 / CEN TC 54)