PREMA EVROKODU 8 THE ANALYSIS OF APPLICATION OF SECONDARY SEISMIC ELEMENTS IN DESIGN ACCORDING TO EUROCODE 8

Projektovanje seizmički otpornih konstrukcija s ciljem zaštite ljudskih života, iako jeste najvažniji, nije i jedini cilj analize ponašanja i projektovanja objekata u seizmičkim područjima. Osim obezbeđivanja prostorne stabilnosti, predmet istraživanja su i performanse objekta tokom i nakon zemljotresa, naročito nekonstruktivnih delova fasade, pregrada, opreme, i uopšte povredljivost objekata [1]. Ipak, imajući u vidu potrebu za predstavljanjem zahteva i mogućnosti tehničkog propisa koji reguliše ovu oblast, Evrokoda 8 [2], u svetlu predstojećeg usvajanja ovog dokumenta kao nacionalnog standarda, fokus rada biće na rasvetljavanju jednog od aspekata primene Evrokoda 8 [2] u projektovanju objekata visokogradnje. Savremeni seizmički propisi, među kojima je i Evrokod 8 [2], nude mogućnost da se doprinos pojedinih konstruktivnih elemenata u obezbeđivanju prostorne stabilnosti objekta za dejstvo zemljotresa zanemari. Takvi delovi konstrukcije nazivaju se „sekundarnim” seizmičkim elementima [2] za koje nije neophodno ispuniti sve zahteve Evrokoda 8 [2], već je moguće primeniti samo odredbe Evrokoda 2 [3]. Nekoliko je razloga za uvođenje mogućnosti podele konstruktivnih elemenata na „primarne” i „sekundarne” u aseizmičkom projektovanju. Pre svega, na ovaj način proširene su mogućnosti utvrđivanja osnovnog nosećeg sistema konstrukcije, jasnom definicijom elemenata koji su ključni


INTRODUCTION
Design of structures for earthquake resistance with the purpose to protect human lives, although the most important, is not the only aim of behaviour analysis and design of structures in seismic regions.Apart from ensuring overall stability, the subjects of research are also building performances during and after earthquakes, particularly of non-structural elementsfacades, partition walls, mechanical and electrical equipment and resiliency in general [1].From the perspective of structural engineering society, there is a necessity to present requirements and possibilities of technical code that covers this field -Eurocode 8 [2], in light of the upcoming adoption of this document as a national standard.Therefore, the focus of this paper is on the presentation of one of the aspects of Eurocode 8 [2] implementation in seismic design of building structures.

KONCEPT PRIMARNIH I SEKUNDARNIH ELEMENATA
Osnovni koncept rada sa sekundarnim seizmičkim elementima zasniva se na zanemarenju krutosti sekundarnih elemenata pri analizi odgovora sistema u seizmičkoj proračunskoj situaciji.Da bi ovakav pristup mining the basic lateral-force-resisting system of the building, by clearly defining the elements which are essential for resisting seismic actionprimary seismic elements and those used only for supporting gravity loadssecondary seismic elements.Furthermore, there are some provisions of Eurocode 8 such as geometrical constrains, ductility requirements and detailing rules or capacity design conditions, which commonly cannot be satisfied as a result of architectural constrains regarding structural layout and dimensions of structural elements.If it is impossible to change the layout or at least crosssectional dimensions, designation of those elements as secondary can solve the problem, while ensuring good and clear structural concept.It is also an option to overcome the problem concerning the structural systems that are not covered by Eurocode 8 [2], such as prestressed concrete structures or systems of flat slab frames.The reason is that the existing experimental data and theoretical analyses are insufficient to explain their behaviour during earthquakes with adequate certainty and to establish reliable recommendations and requirements for design practice.Therefore, one option is to classify these systems as secondary seismic elements, guided by the principle -when the problem is impossible to solve in a satisfactory manner, maybe it can be eliminated [4].Finally, concrete building structures often consist of structural walls with only a few RC frames used for the purpose of bearing gravity loads (e.g. for supporting heavy facades).Strictly speaking, this structural system would be classified as a dual system according to current Serbian seismic design code [5] and a large portion of seismic load would be assigned to RC frames (at least 25%).As a result, column's and beam's dimensions are heavily increased which is contrary to the original designer's intentiononly structural walls resist seismic force and frames are used as gravity load-carrying elements.Furthermore, satisfying ductility demands would lead to an increase of required reinforcement area in those members.For this reason, classification of some elements as secondary seismic elements is certainly an appealing possibility in the framework of modern seismic design codes [2].Although this option seems to be the simplest solution, its application in structural design is unlikely trivial since a number of conditions and requirements should be met.
The aim of this paper is to describe the concept of secondary seismic elements considering EC8 [2] demands and requirements and to present the consequences of classification of some structural elements as secondary.In order to explain all steps in the seismic design of building structures with secondary seismic elements in detail, an appropriate application example of the reinforced concrete building is designed.Eurocode 8 [2] provisions are commented based on the analysis of design results which led to the important conclusions.
Sigmund i ost.
[6] pokazali su, primenom pushover analize na primeru kombinovanog sistema ramova i zidova (gde su ramovi klasifikovani kao sekundarni) da čak i pri zadovoljenju propisanih uslova, globalni stiffness of all secondary seismic members is unlikely to exceed 15%, which precludes the global response of the structure to change significantly.For the same reason, designation of some structural elements as secondary members is not allowed if it changes the classification of the structure from non-regular to regular [2].This provision serves as a preventive measure and it should suppress the possibility to conceal irregularities of building structures by designating them as secondary (e.g.structural walls that are continuous along the full height of the building except at the ground level).
On the basis of the fact that all structural elements should support and transfer gravity loads during earthquakes, the substantial difference in the design of primary and secondary seismic elements lies in the assumption of their behaviour when the structure is subjected to the same maximal displacements.It is well known that global ductility of a structural system depends on curvature ductility of its primary elements, which corresponds to the reduction factor of seismic action called the behavior factor q. Eurocode 8 [2] specifies the requirements for these elements in terms of design and detailing which refer to geometrical constrains, minimum and maximum values of longitudinal reinforcement ratios, as well as shear reinforcement ratio and confinement measures of boundary elements, in order to provide the sufficient curvature ductility.On the other hand, all elements classified as secondary members should withstand displacements governed by a primary system without clearly defined curvature capacity according to Eurocode 8 [2].The first option is to provide adequate design resistance of secondary elements corresponding to the assumption of their elastic behaviour during an earthquake, in order to prevent brittle failure modes when subjected to the expected displacements induces by the seismic action.As a result of applying these demands, internal forces in secondary elements are much higher than those obtained from the usual seismic designin which all elements are designated as primary, but they do not need to conform to the requirements of Eurocode 8 [2] specified for primary elements.Based on the fact that all structural elements with certain ductility, the other option is to design secondary elements for internal forces calculated from the analysis with adopted behaviour factor, which is lower than the one adopted for primary seismic elements.This would lead to lower internal forces in secondary elements than those obtained from the former option, i.e., based on the assumption of their elastic behaviour in the seismic design situation.However, the code [2] fails to provide the guidance for this type of analysis.Ultimately, the option for the analysis of secondary seismic elements is certainly some of the non-linear methods, e.g.pushover analysis, which takes into account ductility capacity of those members more realistically.Since the non-linear methods use predefined cross-sectional characteristics (in terms of longitudinal and transverse reinforcement), the analysis is iterative and the first option for the analysis of secondary elements can be used for the first iteration.Because of its simplicity and conservatism, this paper presents the application of the first method in the analysis of RC building structure considered in the numerical example.
There are only a few analyses of secondary seismic elements in relevant literature.Sigmund et al [6] conducted pushover analysis of the dual system of RC frames and shear walls, with frames taken as secondary elements.The results showed that global building response may significantly differ, depending on whether the frames are classified as primary or secondary even if the code requirements are met.Furthermore, plastic hinges development at ends of secondary columns was noticed.Since those elements are designed only in accordance with Eurocode 2 [3], it is crucial that the design of secondary elements is carried out by internal forces determined from maximal deformations of a primary seismic structure.Fardis [7] proposed a procedure for estimating these forces based on the ratio of inter-storey drifts: one in which the stiffness of secondary elements is not considered and another in which it is.

Geometry and design parameters
The classification procedure as well as the analysis, and design of primary and secondary seismic elements are described for symmetric reinforced concrete building, presented in Figure 1.The building has eight storeys and the story height of h s = 3,5 m.
The design ground acceleration of a g = 0,2g is adopted as a design parameter.The Type 1 design spectrum applied for Ground type B is used, according to EN 1998-1 [2].A preliminary static analysis is conducted in order to determine the fraction of seismic base shear taken by the walls.It was concluded that vertical structural walls resistance exceeds 65% of the total shear resistance of the whole structural system in both directions (approximately 92%).Therefore, the system is classified as a "wall system" [2].The structure is regular both in plan and in elevation, which enables Lateral force method of analysis.The behaviour factor is adopted as q = 3.0 for ductility class DCM [2].
The columns B1 and B2, perimeter beams at an intersection of axis 1 and axes B and C, as well as their connections, are designed by linear-elastic analysis in accordance with EN 1992-1-1 [3] and EN 1998-1 [2].

The analysis of secondary seismic elements
Secondary seismic elements and their connections should be designed and detailed for internal forces which occur at the maximum displacements during earthquakes, in order to have sufficient bearing capacity to support and transfer gravity loads included in seismic design condition [2].Maximum deformations should be calculated in the analysis which neglects the contribution of secondary elements to the lateral stiffness of the structure while primary elements are modelled with their cracked flexural and shear stiffness.The analysis should also include P-Δ effects.
The Eurocode 8 [2] requirements mentioned above imply that it is necessary to conduct two separate analyses of the building structure, for each of two horizontal directions: one, in which the stiffness of all structural elements is considered and, another in which the lateral stiffness of all secondary elements is neglected.For this reason, it is necessary to build two separate numerical models of structure [7]:  a model which includes the stiffness of primary and secondary elements -SP model, and  a model which includes only the stiffness of primary elements -P model.
The P model is built based on neglecting lateral stiffness of those structural elements which are intended to be classified as secondary by the designer.This could be accomplished by modelling secondary elements without flexural stiffness (by reducing the moment of inertia or modulus of elasticity) or by modelling them with moment releases on their ends.Maximum displacements calculated from the P model are used for estimation of internal forces in the secondary elements in the SP model, by the procedure described in 3.2.2.
Besides being used for determination of maximum displacements of the structure, P model is also used for the purpose of classification of secondary elements as well as for the design of primary elements in seismic design situation (Figure 3a).On the other hand, SP model is used for the design of secondary elements in the seismic design situation, and for analysis and design of whole structure in all other design situations.

The classification of secondary seismic elements
The total contribution to the lateral stiffness of all secondary seismic elements should not exceed 15% of that of all primary elements, according to the requirement 4.2.2 (4) of Eurocode 8 [2].However, the procedure for estimating the contribution of the stiffness of secondary elements is not defined, which allows two approaches to be used.The first method, and a simpler one, is based on the estimation of the fraction of base shear taken by secondary elements [7].The second method uses inter-storey drift ratios of building structure δ r,P /δ r,SP obtained from the analysis of P model and SP model at each building level.Interstorey drifts are calculated in accordance with EN 1998-1: 4.3.4[2], for the same system of horizontal forces in each of the two horizontal directions [7], where: δ r,P is the design inter-storey drift obtained from P model, and δ r,SP is the design inter-storey drift obtained from SP model.
This method analyzes the contribution to the lateral stiffness through flexibility, which is a simpler approach in practical application when using software for structural analysis.The method is based on the analysis of two systems with a single degree of freedom (SDOF), corresponding to P model and SP model defined in Section 3.2.The requirement that limits the contribution of the stiffness of secondary elements, according to Eurocode 8 [2], can be presented by the Expression (1): gde su: K S krutost sekundarnih seizmičkih elemenata, K P krutost primarnih seizmičkih elemenata koja odgovara P modelu.Kako se klasifikacija sekundarnih seizmičkih elemenata sprovodi na osnovu pomeranja P i SP modela, uslov za klasifikaciju se može prikazati preko odgovarajućih fleksibilnosti: where: K S is the lateral stiffness of all secondary seismic elements, K P is the lateral stiffness of all primary seismic elements which corresponds to P model.Since the classification of secondary seismic elements is conducted for the displacements of P and SP models, the condition for the classification can be presented by using the corresponding flexibility: gde je: K SP ukupna krutost sistema, koja obuhvata krutost primarnih i sekundarnih seizmičkih elemenata, K SP = K P + K S .Konačno, doprinos krutosti sekundarnih elemenata u ukupnoj krutosti sistema, izražen preko odnosa fleksibilnosti δ P /δ SP , glasi: where: K SP is the total lateral stiffness of the system, which comprises the stiffness of both primary and secondary seismic elements, K SP = K P + K S .Finally, the contribution to the global lateral stiffness of secondary elements can be expressed in terms of the ratio of corresponding flexibilities δ P /δ SP : Imajući u vidu definiciju krutosti konstrukcije, akcenat je na istom sistemu horizontalnih sila -iste raspodele po visini, ali i istog intenziteta.Prema preporuci nekih autora, raspodela opterećenja po visini treba da odgovara seizmičkom opterećenju [7].Međutim, vrlo često se pri aproksimaciji krutosti sistema koristi i jednako raspodeljeno opterećenje po visini, što može biti jednostavnije za unos u proračunski model.Ghali i Gayed [8] pokazali su, na primeru konstrukcije od 12, 25 i 50 spratova, da je uticaj primene ove dve raspodele na odnos međuspratnih pomeranja manji od 1,0%.U ovom numeričkom primeru, razlike su manje od 1,7%, pri čemu raspodela koja odgovara seizmičkom opterećenju daje konzervativnije rezutate.
In order to compare the lateral stiffness of two structures (P and SP models), the same system of horizontal forceswith the same distribution along the height, but of the same intensity also should be applied.The distribution of horizontal forces should correspond to the seismic load i.e. to the height-wise linear one [7].However, it is common practice to use uniformly distributed load along the height to estimate lateral stiffness of the system, which arises from its simple application in the numerical analysis.Ghali and Gayed [8] showed that the influence of application of these two load distributions on the inter-storey drifts is less than 1.0%, based on the analysis of building structures with 12, 25 and 50 storeys.In the current numerical analysis, the differences are less than 1.7%, and the load distribution which corresponds to seismic load gives slightly conservative results.
The discrepancies of the analysis results arising from the application of these two methods can be significant.They are the result of the different deformed shape of the certain structural elements that are a part of a lateralforce-resisting system, which the other method takes into account.The difference between the deformed shapes of the elements is especially noticeable (at higher levels), considering the selection of the elements which are analyzed as secondary (perimeter frames and interior columns).The analysis of inter-storey drifts of both models (curves S1 in Figure 2), calculated for the same system of seismic load, have shown that the stiffness contribution of all frames (perimeter frames and flat slab frames) fail to fulfill the code requirement in both horizontal directions -δ r,P /δ r,SP > 1,15.For the purpose of comparison, these elements resist only 8.9% of total seismic base shear in X direction and 8.6% in Y direction, which would satisfy the code requirements.There are few possible solutions that satisfy code requirements in terms of the ratio of inter-storey drifts: (1) to increase the contribution to the lateral stiffness of primary elements, (2) to decrease the contribution of secondary elements, or (3) to classify only one system of frames as secondary (system of perimeter frames or system of flat slab frames).In this particular case, the contribution to the lateral stiffness of secondary members is decreased, by decreasing the cross-sectional dimensions of perimeter column to b s /h s = 25/40 cm, Slika 2. Doprinos krutosti sekundarnih seizmičkih elemenata Figure 2. Contribution of secondary seismic elements

Internal forces in secondary seismic elements
The Eurocode 8 requirement for the design of secondary elements is determined on the basis of "equal displacement" rule which considers different (reduced) ductility capacity of secondary elements (and their connections) in comparison with ductility capacity of primary elements.If the ductility capacity of all secondary members is not determined precisely, it is crucial to provide adequate resistance corresponding to the assumption of their elastic behaviour during the earthquake action.Moreover, the internal forces in these elements are determined from seismic displacements of the system which is more flexible (P model), in order to take into account the most unfavourable design condition (Figure 3.a).In other words, the internal forces in secondary elements are higher than those obtained from the analysis of the whole structure under seismic actions with the assumed elastic behaviour, proportionally to the inter-storey drift ratio of P and SP models.Milev [9] presented similar approach for calculation of internal forces in secondary members.A relatively accurate estimation of internal forces in secondary elements throughout the structure can be obtained by using the inter-storey drift ratios d r,P,m /d r,SP,m (Figure 3.b).Unlike the case of the interstorey drift analysis for the classification of secondary elements, the interstorey drifts d r,P,m and d r,SP,m are calculated under the actual design seismic load acting on corresponding structural model [7].Finally, the internal forces in secondary members at the floor level m, are computed in the SP model for the seismic combination [10], with introducing the coefficient α as a multiplier of the seismic action:
The procedure described above is inadequate for the analysis of rigid, short-period structures (with fundamental period of vibration smaller than T c [2]), and instead of the "equal displacement" rule, the so-called "equal energy approximation" is used.In this case, the calculation of interstorey drifts on the basis of the displacement ductility factor μ δ (given in 5.2.3.4 (3) [2]), for the purpose of determining the coefficient α, is considered as a reasonably accurate.
In the current numerical analysis, the fundamental periods of SP model and P model are approximately equal to 0.80 s and 0.85 s, respectively, which are larger than T c = 0.5 s (Ground type B).The corresponding seismic forces, determined by Lateral force method of analysis, are about 4860 kN and 4575 kN.Therefore, the internal forces in secondary members are computed by using "equal displacement" rule.
the seismic action effects from 3.0 to 3.3 times in relation to the effects obtained for the primary elements, as per Equation (5).The following sections are focused on the differences between the analysis results of certain structural elements when considered as (1) primary and (2) secondary seismic elements.

The analysis of the design results
As pointed out before, the foundation for the design of primary elements is their capability of developing ductile behaviour which enables them to sustain large deformations in inelastic range under the seismic action.This is achieved by proper detailing of those members, especially in certain ("dissipative") zones i.e. "critical regions" [2].Unlike, the secondary members rely upon their strength, instead of ductility, to support gravity loads when subjected to the same displacements as primary members.
For the purpose of comparative study, the perimeter column B1 and beam BC-1, interior column B2 and its connection to flat slab are analyzed and discussed in cases of their classification as part of (1) primary and (2) secondary system for resisting seismic action.

The design results of column B2 and corresponding slab-to-column connection
The design results for column B2 under the extreme design combination of actions at each floor level are presented in Figure 5.It clearly shows the influence of the classification of column as primary (PSE) and secondary (SSE) element, in terms of required longitudinal reinforcement ratio.In case of its designation as primary element, the required reinforcement ratio is clearly lower than a minimum value required for Ductility class DCM (Figure 5.a), which arises from its small contribution to the global lateral stiffness.The same (minimum) amount of reinforcement is sufficient to ensure the required resistance of the column when it is designated as secondary, except at the top storey which is critical for section design.Furthermore, it is possible to decrease its cross-sectional dimensions, considering the fact that they are not governed by ductility requirements in terms of the maximum value of normalized axial force.Instead, the cross-sectional dimensions can be determined by limiting the compressive stresses in serviceability limit states [3].In this case, the crosssectional side length is decreased from 40 cm to 35 cm.For comparison, the cross-sectional area of the same column of 11-storey building with the same structural layout can be decreased by 45 %.Razlike u potrebnoj količini uzengija prikazane su na slikama 5b i 5c, kao i u tabeli 1 pomoću mehaničkog zapreminskog procenta armiranja ω wd,m određenog za ceo m-ti sprat.Na osnovu prikazanih rezultata može se da zahtevi za armiranje uzengijama primarnog seizmičkog stuba u kritičnim oblastima imaju rezultat u značajnom povećanju količine uzengija, i do 75% u nivou osnove gde je potrebno obezbediti adekvatno utezanje preseka.Jasno je, takođe, da klasifikacija u sekundarne rezultuje većom podužnom armaturom, ali kada se povede bitka za svaki kvadratni centimetar (skupog) slobodnog prostora, verovatno će opcija jače armiranih stubova manjih dimenzija dobiti prednost nad stubovima većih dimenzija s minimalnom armaturom.
The differences in required shear reinforcement area are shown in Figure 5b and 5c as well as in Table 1, in terms of mechanical volumetric ratio ω wd,m calculated for the whole storey m.Based on presented results, it can be concluded that the detailing requirements of primary seismic column in critical areas give an increase of the shear reinforcement, up to 75 % at the column base where adequate degree of the confinement is needed.It is also clear that an increase of longitudinal reinforcement is a consequence of classification of the column as secondary.However, in the discussion for each square centimetre of (expensive) available space, the option of more heavily reinforced columns with smaller cross-sectional dimensions will probably prevail over the option of column with larger cross-sectional dimensions and minimum reinforcement ratios.
Apart from the column, the slab-to-column connections should also be designed and detailed when subjected to the maximum displacements due to earthquakes, as mentioned in Section 3.2.This implies, above all, that punching shear stresses need to be checked.It is well known that these stresses are a function of the intensity of gravity loads, but they also increase during earthquakes, as a consequence of an increase of the load eccentricity presented with the coefficient β [3]. Figure 6 presents the values of coefficient β as well as the shear stresses at the basic control perimeter along the height of the building as a function of different classification of column B2, for persistent and seismic design situation.It can be concluded that the slab has sufficient punching shear resistance (v c,Rd = 0,094 kN/cm 2 ) under gravity loads in the persistent design situation.The increase of bending moments in secondary columns leads to increase of coefficient β (almost two times than minimum prescribed value of 1.15 [3]) and punching shear stresses, which are higher than corresponding values calculated in persistent design situation.Moreover, the punching shear resistance is exceeded and, therefore, punching shear reinforcement is required.Slika 6. Vrednosti koeficijenta β i napona smicanja od probijanja u funkciji klasifikacije stuba B2 Figure 6

The design results of perimeter frame in axis
1-column B1 and beam BC-1 The influence of the perimeter frame stiffness on the magnitude of deformations and the shape of deformed structure is explained in Section 3.2.1.The same conclusion can be drawn from the analysis of design results of perimeter frame in axis 1, presented in Figure 7 and Table 2.As a result of the frame action and high axial forces in perimeter columns under lateral loads, the implementation of Equation (1) in design of perimeter columns, classified as secondary, leads to reduction of axial forces followed by an increase of bending moments.The design for such internal forces gives a large amount of reinforcement area, especially in lower storeys (Figure 7a).On the other hand, the normalized axial forces in primary columns are high (ν d,max = 0,53) due to narrowed cross-sectional width (determined from ductility demands) which subjects them to strict rules for detailing and confinement of the concrete core.The values of mechanical volumetric ratio ω wd,m are from 33% to 93% higher than those determined for secondary perimeter columns.However, this is insufficient reason for their classification as secondary, mainly for economic reasons, because of a large amount of longitudinal reinforcement.In this case, the reduction of crosssectional dimensions is unlikely an option since it would further increase the reinforcement ratio, above the maximum value of 4% [3].It is evident that the desired moguće postići željene rezultate i da ih je najbolje razmatrati kao deo primarnog sistema.results could not be achieved by classification of the column B1 as secondary, and that it is reasonable to treat them as a part of primary system for supporting seismic loads.Slični zaključci mogu se primeniti i na grede koje su deo fasadnih ramova.Rezultati proračuna grede BC-1 pokazuju očigledan uticaj povećanja momenata savijanja u sekundarnim seizmičkim gredama, dobijenih primenom izraza (1), koji rezultuje povećanjem armature i do tri puta.U ovom primeru, uzengije u primarnim gredama određene iz uslova kapaciteta nosivosti praktično su iste kao uzengije sekundarnih greda određene iz elastičnih uticaja (tabela 2).Similar conclusions apply for the perimeter beams.The design results of the beam BC-1 indicate that the bending moments are significantly increased by application of Equation (1) for secondary seismic beams, which increases the required reinforcement area up to three times.Table 2 shows that, in this case, the amount of transverse reinforcement is similar to the different classification of the beam i.e. the capacity design shear forces in primary beams are almost equal to the elastic shear forces in secondary ones, calculated by an implementation of Equation (1).

CONCLUSIONS
The analysis of reinforced concrete structure with secondary seismic elements presented in this paper highlighted the advantages and disadvantages of application of this interesting concept in aseismic design of building structures.Although this concept may seem appealing to the designer because secondary members can be designed and detailed according to Eurocode 2 [3] "only", its application in design practice is rather demanding with uncertain outcome.The expected benefit in terms of the simple design procedure is compromised by classification procedure and methods for calculation of internal forces which are based on comparative analysis of two numerical models of the same structure.The results of the analysis showed that, in some cases, design requirements for secondary elements are almost the same as Eurocode 8 requirements [2], which apply for primary (ductile) elements.Further, it is shown that the choice of this classification has significant influence on the behaviour of slab-to-column connection and on the values of punching shear stresses.The decrease in crosssectional dimensions of secondary members can be expected if their contribution to the lateral stiffness is low.However, this will increase the amount of longitudinal reinforcement.The implementation of this concept in Eurocode 8 [2] provides the possibility for complex analysis of certain structural elements, and the interpretation of the code rules is certainly a challenge for both designers and researchers.

Figure 5 .
Figure 5. Design results of column B2: a) interaction diagram, b) cross section of the column B2 (PSE), c) cross section of column B2 (SSE)

Figure 7 . 1 Figure 8 .
Figure 7. Design results of the column B1: a) interaction diagram, b) cross-section of the column B1 (PSE), c) cross section of the column B1 ( SSE)