Testing steel at high temperatures

10th July 2013

Ratio Rp0,2/Rm at elevated temperature for different steel grades
Ratio Rp0,2/Rm at elevated temperature for different steel grades
Relationship between design-values and mechanical properties (A387-22-2)
Steel grade A387-22-2, Hot Tensile Test Values versus Calculation Limits of ASME II-D, Tables U and Y-1
Steel grade A387-22-2, Tensile strength values versus Temperature, including Calculation Limits of ASME II-D, Table U

Jörg Maffert and Dr Ingo Detemple look at the criteria for tensile tests at elevated temperatures on CrMo(V) steel plates.

Technical requirements for petrochemical reactor steels have proliferated in the past decade. The need to increase economic benefits together with higher operating temperatures and pressures are leading to the construction of higher capacity reactors with thicker walls.

Also, more severe requirements in these steel specifications leave the steelmaker little scope to find a good balance between all the influencing parameters, in order to achieve the best possible steel design.

The design of pressure vessel for petrochemical industry is based on mechanical properties and the design method, which are given by the construction code. The design values such as ASME II-D, Tables U and Y-1 [1] are to be used for design purpose only. Specification writers whereas often specify acceptance criteria for tensile tests at elevated temperatures based on these tabulated values. We have seen that the level can differ between 80 and 90 per cent of the values of table U.

This paper targets the sensibility of users of codes to take into account the real potential of high sophisticated steels, such as CrMo-steels, in accordance with ASME SA387 and SA542. The method to verify the suitability of the given values for testing purpose will be described in this article. The modern technology of steel production as well as an appropriate method have been taken into account.

Hot tensile test

Recommendations concerning elevated temperature tensile tests be­come sometime mandatory, the acceptance criteria are somehow aligned to the code values, which are given for calculating purpose only. The tested values shall not be compared to the design values for rejection purpose. When the requirement is on a 90 per cent level of the table U, the manufacturer would try to improve the hot tensile properties. Such a product would have quite instable tensile properties especially after long time post weld heat treatment (PWHT), which is not acceptable for an equipment running a high temperature and high pressure. 

As an example we refer to APIRP 934 A [2] recommendation: “Elevated temperature tensile tests, when required by the purchaser, should be performed at the equipment design temperature. Test specimens should be in the maximum PWHT condition. Acceptance values should be as specified by the owner/user…”

The most efficient way to verify the suitability of the acceptance criteria is the ratio Rp0,2/Rm (please note: Rp0,2 and Rm are determined together within the same tensile test). Subsequently, we determine typical ratios for different steel grades at ambient temperature (Fig. 1) and elevated temperatures (Fig. 2). In Fig. 1 the requirements according ASME are included as solid lines. In Fig. 2 the design values according ASME II-D, Table U and Table Y-1 are shown as a function of the testing temperature.

It comes clear that the ratio Rp0,2/Rm is characteristic for a given steel, by way of example, for steel grade A387-22-2 this typically is between 0.65 and 0.85.

Fig. 3 shows for the SA 387-22-2 the relation of Rp0,2/Rm, calculated both with the design values and the recently measured values of our production. The large difference between the data population and the curve calculated with the design values is obvious. In terms of the related Rp02, the design values according Table Y-1 are much lower than the measured yield-strength data points, whereas for the tensile strength the design values according Table U is much too high, compared to the tested values. We plotted a 85% (dotted) and a 90 per cent (solid) line to better visualise different acceptance criteria levels. This indicates a certain misfit according the tabulated design values and the mechanical properties of the examined steel grades. Furthermore, the ratio Rp0,2/Rm calculated with the design values of Table U and Y-1 leads to extremely low value of about 0.45, which is more typical for an A516-60 than an A387-22-2.

This all may raise reasonable doubts, that the tabulated design values should be used as acceptance criteria.

The graph in Fig. 4 highlight for SA387-22-2 the measured hot tensile test values versus calculation limits of ASME II-D, Tables U and Y-1 in a temperature range up to 550°C. Having in mind that the tabulated tensile values are too high (compared with the yield strength values), they are the most challenging to achieve for testing. The 85 per cent and 90 per cent reference lines are plotted in the graphs.   

EN 10314

As the above ratios of Rp0,2/Rm, calculated with the tabulated values don´t fit with the long term experience of one the most experienced plate mill, ways have to be found how to specify acceptance criteria in order to guaranty a safe product with excellent properties.  

Some of the specification writers understand the above described issue, specifying values, which fit more with the modern steels.

An appropriate statistical method for calculation of the minimum hot tensile or yield strength is given by EN 10314 [3]. Using this method will leads to curves, which are less or equal 85 per cent of the values listed in table U of ASME II-D.

The same method is used to determine the derivation of minimum values of proof strength included in the harmonised European standard for flat products made of steels for pressure purposes (EN10028-3)

In the following graphs we plot the curve, resulting from EN 10314, into the measured data, in order to enable the reader to compare them with the calculation limits of ASME II-D, Table U. We can see that the graph fits well with the 85 per cent level.


The values of ASME II-D, tables U and Y-1 are to be used for design purpose only. In chapter 2 we have seen that there is a misfit between the tabulated values and the measured mechanical properties of the examined steel grades. Rp0,2 values of table Y1 are quite below the mechanical properties of the evaluated steels, and vice versa for Rm.

It is the task of design people to define the test method as well as the acceptance criteria. In the case of pressure vessels running at high temperature, appropriate values have to be specified from existing standards and specifications. We motivate the user to verify the aptitude of the chosen method. The above described philosophy of EN 10134 can help and simplify the evaluation.
The one who is trying to fulfil a too high acceptance criteria, such as 90 per cent of ASME II-D, Table U values, can have a negative impact because it can impair the well balanced profile of mechanical properties.

A calculation of the minimum hot tensile strength according to EN 10314 leads to curves less or equal to 85 per cent of the values listed in table U. As described in the article this level fits quite well with the evaluation based experience of one of the leading steel makers.

Enter X at

Jörg Maffert, Technical Marketing, and Dr Ingo Detemple, Metallurgical Department, are with Dillinger Hütte GTS, Dillingen, Saar, Germany.

References: 1. Boiler and Pressure Vessel ASME Code, Section II (Materials), Part D (Properties) [last edition]: tables U, Y1; 2. API Recommended Practice 934-A [last edition]: Materials and Fabrication of 21/4Cr-1Mo, 21/4Cr-1Mo-1/4V, 3Cr-1Mo, and 3Cr-1Mo-1/4V Steel Heavy Wall Pressure Vessels for High-temperature, High-pressure Hydrogen Service; 3. EN 10314 (last edition): method for the derivation of minimum values of proof strength of steel at elevated temperatures

Abbreviations: Rm - Tensile strength; Rp0,2 - Yield strength; T -  Testing Temperature (°C)


Previous Issues















twitter facebook linkedin © Setform Limited