The world of Pressure Equipment – The novelties of 2019

This year it was particularly rich in news for anyone working with European, American and also Italian standards: there is a lot of movement, we will try to explain why and analyze the consequences.

The harmonized Unfired Pressure Vessel standard EN 13445.

During 2018 several amendments of part 3 (Design and Calculation) were released, all of which will be integrated into the new 2019 edition of the standard. The most interesting is the one related to Clause 17, which deals with simplified fatigue analysis, that does not require a FEM analysis of the vessel, but merely calculates the stress variations on the basis of the normal formulae used for code calculations. Apart from the correction of the strong (and illogical) penalty provided in the previous version for the longitudinal weld of cylindrical shells, again there is the fact that the simplified analysis, previously possible only in the case of pressure fluctuations, is now also possible in the case of temperature fluctuations (or, better to say, in the event of fluctuation of the thermal stresses generated in the device by temperature differences between adjacent points). It should be noted, however, that the determination of the actual wall temperatures and their variation over time is not a particularly easy task: the risk is to be obliged to carry out a FEM analysis for the sole purpose of determining the temperature field and its evolution over time: we believe however, that this new standard can be applied in some typical cases (for example, the external jackets or coils used for cooling or heating tanks and reactors) in which the temperature field is, in any case, easily conceivable.

Another important change (long overdue!) Is the completion of the standard (contained in chapter 16) for the calculation of the stresses generated on shells and domed ends by the forces and moments that the pipes discharge onto the nozzles of the vessels: in the previous version this calculation was possible in the presence of only 3 load components (two bending moments and an axial load), while now also the case of two shear forces and a torsional moment has been foreseen. In this way it is possible to avoid the verification with an alternative standard (WRC or PD5500), a verification required previously in all cases where the piping analysis provided all the 6 load components (with endless discussions, often even with highly qualified companies), whose technicians were however ignorant – in the literal sense, that is, they simply ignored the limitations of the previous standard).

The 2019 version of EN 13445 (which will by now be dated 2020) is in preparation: the draft was subjected to a careful analysis by the HAS Consultant (the Commission consultant who verifies the compliance of the harmonized standards with the Essential Safety Requirements of the Directive). Unfortunately in this case the HAS Consultant, firmly intent on earning the loan that the Commission pays him, made a flood of comments, not only on the parts that the Amendments have modified, but also on everything else. It is the same type of analysis that is required by the CEN rules when a new standard is sent to all national standardization bodies for the Public Enquiry: the vote can be favorable, but each organization has the right to send its comments, to whom the TC is obliged to respond, either by accepting the comment and consequently modifying the draft standard, or by rejecting it; however in this case it must justify the reasons for the refusal.

It will then be the Formal Vote on the updated document to decide its approval as a European standard, binding for all national standardization bodies. In the case of the comments formulated by the HAS Consultant there is however a not negligible difference: the TC that has prepared the standard can accept or refuse a comment, but if a comment has been refused, the TC must convince the aforementioned gentleman to withdraw the comment; and must be particularly convincing, because if the HAS Consultant is not completely satisfied, the standard, even if approved in the formal vote, cannot be included in the list of harmonized standards for the reference directive, and therefore will not have the so-called “presumption of conformity” with the Directive. In the meeting of October 24th of our WG53 / CEN TC54, we held an online meeting with the HAS Consultant, at the end of which there was only one comment yet to be resolved, for which we asked for an opinion directly to the Commission: waiting for this opinion, I do not want to make advances. We hope well!

There are still a lot of amendments of EN 13445 which are now in preparation: even with the time allowed by the CEN bureaucracy, we will soon see the new Clause 18 of EN 13445.3 on detailed fatigue analysis (which required endless discussions  among the experts); always in part 3, the new Clause 9 on opening reinforcements (no substantial change, only the correction of some formal errors); more  important are the new part 11 on Titanium Alloys and the new part 12 on Copper Alloys. Hoping that the HAS Consultant will not create too many problems …


The American standard ASME Section VIII division 1

Although it is by far the most used calculation code in the world, it feels now the weight of the years and of its overly pragmatic approach (that is based on the respect for every minimum detail of the code rather than on a reasoned analysis by the designer), which makes it particularly suitable for less trained operators, who are not particularly encouraged to use their own brain, rather than rigidly applying the rules of the code. With this judgment I would not like to raise the anger of those Inspection bodies that are particularly active in Europe as ASME authorized agencies: but it is the same evolution of the code that justifies this judgment. The main features of the code are in fact the following:

  1. Low allowable stresses, heavily limited by the safety factor 3.5 on the tensile strength, with a consequent increase in the weight of the vessels.
  2. Low requirements on non-destructive tests: the use of the efficiency 0.7 (no volumetric control of the welds) is allowed also for large vessels.
  3. Rather superficial consideration of loads other than pressure (only a 20% increase in the allowable stresses for both the earthquake and the wind load is prescribed).
  4. Calculation for hydraulic test not required. The test pressure is imposed by the standard, but a corresponding calculation is not required, and in any case no allowable stresses are provided for this purpose.
  5. Particularly simple calculation rules, which still tend to favor obsolete configurations (reinforcement plates instead of self-reinforcing nozzles).
  6. Lack of rules for stress analysis and fatigue analysis.

Since 2007, however, under the pressure of the European standard EN 13445, ASME has felt the need to modernize its pressure equipment standards. This was done starting with the complete remake of the ASME VIII division 2, which copied numerous points of the European standard, including:

  1. Allowable stresses (coefficient on the tensile strength reduced to 2.4).
  2. Test pressures (practically the same as for PED).
  3. Alignment to 13455.5 of the subdivision into groups with respect to materials and non-destructive controls (see table 7.1 Examination Groups for Pressure Vessels, substantially identical to table 6.6.1-1 of 13445.5, except for the lack of Testing Group 4).
  4. Possibility of using Stress Analysis, both elastic analysis and non-linear analysis.
  5. Fatigue analysis that is partially aligned with the European school (in addition to the classic method of Fatigue Strength Reduction Factors, a more modern method, based on structural stresses instead of total stresses including peaks, is now provided).

In the 2019 edition of ASME VIII a substantial approach of its two divisions (1 and 2) was made, which had already been in place for some components: in fact, an ASME Case already existed which allowed for the calculation of the opening reinforcements following the (more modern) rules of division 2. But in the 2019 edition there is even an Appendix (Mandatory) that allows you to calculate any component of a device having the U stamp (i.e. built in workshops that have the qualification only for division 1) with the rules of division 2, even using stress analysis and fatigue analysis (but always considering the allowable stresses of division 1): the fact of using division 2 for the calculation of opening reinforcements also allows the use of the various WRC Bulletins that deal with the stresses generated on the shells by the piping loads on the nozzles. It is also now possible to calculate the horizontal tanks on saddles according to the classic Zick method, present in division 2 but not in division 1.

It must be assumed that this trend, for the moment just outlined, must progressively continue in the future, with the gradual replacement of parts of the standard, or even with the integration of the two codes. A symptom of this intention could be the introduction in division 2, starting from the 2017 edition, of two different classes of equipment, class 1 with the lowest allowable stresses (coefficient 3 on the tensile strength), limited to temperatures outside the creep range, and class 2 with the highest stresses prescribed from 2007.


The Italian standards

After about a year of tug-of-war on the revision of the CTI Recommendations for the use of the old ISPESL Raccolte in the context of the PED, UNI, the Italian standard organization, after the work at its subsidiary CTI (Comitato Termotecnico Italiano, charged for all the standards concerning Pressure Equipment) had been practically completed, had vetoed the publication of the updated standard not only as “Recommendations”, but also as “TS – Technical specification” or “TR-Technical Report”. Thanks to a decisive stance taken by INAIL (the Italian Authority which took the competences of ISPESL), has finally accepted the publication of the ex-Recommendations as PdR (= Prassi di Riferimento = Reference practice): this year finally the publication of the new PdR55:2019 could take place, preceded by the making available (free) of the ISPESL Raccolte 1995 Edition, complete with errata, now directly downloadable from the INAIL website at the following link:

The last objection of UNI, which considered the publication of a standard making reference to other standards practically untraceable, was therefore superseded. In fact the old Raccolte, produced by the Italian Governmental Printing Office only in paper form (and only in Italian), were largely exhausted. I will not be here to list the advantages and disadvantages of a PdR with respect to another type of publication: I will only say that it will be valid for 5 years, after which it will have to be withdrawn or transformed into a real standard. For now, you just need to know that this is also downloadable for free from the UNI website:, like all the other documents published as UNI PdR.

But was this new document really needed in Italy? Maybe there are not enough pressure vessels on the world already? And isn’t it true that the PED Directive allows any standard? And in particular is the European standard EN 13445 not supposed to give presumption of conformity to the PED? All true, mind you: but it is also true that some norms that are most popular (ASME VIII division 1) are difficult to integrate with the PED, while the European standard EN 13445 is still incomplete as regards lamellar cast iron, copper alloys and alloys of titanium; and it is also true that for the requalification of existing vessels it is nevertheless an advantage to consider the standard with which they were originally designed. Moreover, if we look at our work in Sant’Ambrogio, the Raccolta VSR is still required by many of our customers, both in terms of software and design.

But what are the changes compared to the old Recommendations? The most important is the adjustment of the test pressure to that provided for by EN 13445.5: in other words the formula PH = 1.25 (fo / f) PS, where fo / f is the ratio of the allowable stresses of the material (ambient temperature with respect to design temperature) will remain the same, but in the old Recommendations it was required that the ratio of the allowable stresses, for a device made of different components, different materials and possibly also with different design temperatures, should be the highest among all possible stress ratios of the main components: if, however, this forced an over-dimensioning of the components with lower (fo / f) values, and if the apparatus had a minimum of radiographic examinations, it was allowed to decrease the test pressure up to the maximum value tolerable by these components: this was also the setting of the 13445.5 up to a few years ago, but then this setting had been modified: now the value of (fo / f) to be taken for the calculation of the test pressure, in the PdR as in 13445.5, is the lowest among all the possible stress ratios of the main components, without prejudice to the lower limit envisaged by the PED, still equal to 1.43 PS.

In the Raccolta VSR there is then the update of the calculation rules of opening reinforcements, which in the old Recommendations provided for a very low ratio between thickness of the nozzle and thickness of the shell, which severely limited the advantage of using self-reinforcing nozzles (a solution far preferable to reinforcing plates, cheaper as material cost, but much more expensive as labor costs); now this rule has been changed, at least in all cases where the component in question is not in a critical area from the point of view of fatigue. The method for calculating flat ends welded to cylindrical shells has also been changed, where the formula used to limit secondary stress on the shell due to the rotation of the end was strongly penalizing the end thickness: as in the case of openings, the old formula remains now valid only if a fatigue check is required. All this always with a view to aligning the VSR Collection with the harmonized standard 13445.3. Also interesting is the new formulation of the allowable stresses for vessels made of Titanium or Titanium alloys, based now on the creep characteristics of the material (where available), as foreseen in the draft standard EN 13445.11.

             Milano, December 5th 2019                                                                                       Ing. Fernando Lidonnici