Use of advanced calculation methods to complement methods of effective cross-section in natural fires

Les je naraven gradbeni material, ki ima relativno na svojo težo dobre mehanske lastnosti. Poleg tega ima tudi dobro požarno odpornost, saj zogleneli zunanji del lesa ščiti notranji del pred hitrim naraščanjem temperature. V praksi se za račun požarne odpornosti lesenih konstrukcij pogosto uporabljajo poenostavljene računske metode, ki jih podaja standard SIST EN 1995-1-2 [9]. Kot kažejo zadnje raziskave, pa so te metode lahko tudi nekonservativne oziroma z uporabo teh metod ne zagotovimo ustrezne požarne varnosti lesenih konstrukcij v primeru naravnega požara, kakšrnega predstavlja parametrična požarna krivulja SIST EN 1991-1-2 [8]. Razlog za to leži v debelini nenosilnega sloja, ki je skladno s SIST EN 1995-1-2:2004 [9] podana in umerjena na standardno požarno izpostavljenost in znaša 7 mm. Skladno s poenostavljeno metodo [9] se enako debelino nenosilnega sloja upošteva tudi v primeru parametrične požarne krivulje. Na podlagi napredne toplotne in mehanske analize za račun požarne odpornosti konstrukcij, pa je bilo v tej magistrski nalogi ugotovljeno, da so vrednosti nenosilnega sloja običajno večje od 7 mm, v primeru ko obravamo parametrično požarno krivuljo. To pa pomeni, da je skladno z poenostavljeno metodo podano v [9], požarna odpornost za parametrično požarno krivuljo precenjena, posledično pa je metoda nekonservativna. V magistrski nalogi smo obravnavali lesen nosilec izpostavljen različnim parametričnim požarom. Temperaturo po prerezu nosilca smo določili z napredno toplotno-vlažnostno analizo. Požarno odpornost lesenega nosilca pa smo določili po dveh metodah, in sicer po poenostavljeni in napredni metodi. Pri poenostavljeni metodi smo nosilnost izračunali na podlagi efektivnega prečnega prereza. Debelina zoglenele plasti smo določili glede na izotermo 300 oC, ki je bila določena na podlagi temperaturnega polja izračunanega s toplotno-vlažnostno analizo. Glavni kriterij za določite debeline nenosilne plasti je bil, da je upogibna nosilnost po poenostavljeni metodi ob koncu oglenenja manjša ali enaka upogibni nosilnosti po napredi metodi ob koncu oglenenja. S tem smo zagotovili, da da smo bili na varni strani glede na napredno mehansko analizo. V nalogi smo podali nove potrebne debeline nenosilne plasti, pri tem pa tudi ugotovili, da se debelina nenosilne plasti skozi celoten razvoj požara spreminja, in ni konstantna, kot je to v primeru standardne požarne izpostavljenosti. Poleg tega smo odkrili še ostale pomanjkljivosti metode zmanjšanega prečnega prereza. Prvi problem je v določitvi končnega časa oglenenja, drugi pa s tem povezan končni čas računa. Ta se zaključi, ko prerez neha ogleneti in s tem ne izgublja več nosilnosti, vendar se v realnosti temperatura v notranjosti prereza še nekaj časa povečuje in kot je bilo prikazano z napredno metodo, nosilec še vedno izgublja nosilnost tudi po tem, ko ne ogleni več.Timber is a natural structural material, which has good mechanical properties compared to its weight. Additionally, it has good characteristics in a case of fire exposure since the charring layer of the cross-section protects the inner timber from sharp temperature increase. In design practice, simple methods for fire resistance of timber structures are commonly used according to standard SIST EN 1995-1-2:2004 [9]. However, the latest research show, that this methods can be non-conservative and using them do not guarantee the adequate fire safety of timber structures in the case of natural fire such is a parametric fire curve given by SIST EN 1991-1-2:2004 [8]. The reason is the thickness of zero strength layer, which is calibrated to standard fire exposure, and is, according to standard SIST EN 1995-1-2:2004 [9], 7 mm. When using simple design method [9], the same thickness of zero strength layer is also taken in the case of a parametric fire curve. However, based on the advanced thermal and mechanical analysis, it was discovered in this master thesis, that the values of zero strenght layer are usually greater than 7 mm in the case when the parametric fire curve is considered. This means that according to simple method [9], the fire resistance for the parametric fire curve is overestimated and consequently the method is non-conservative. In this master\u27s thesis a timber beam exposed to various parametric fires is analysed. The temperature distribution across the cross-section was determined by advanced heat-moisture analysis. Further on, the load-bearing capacity of timber beam was determined based on two methods – simple and advanced method. According to the simple method, the load-bearing capacity was determined based on the effective cross-section, where isotherm 300 °C is implemented to estimate the char layer thickness,. The main criterion for determining thickness of the zero strength layer d0 was that the bending capacity determined by the simple design method at the end of charring is less or equal than the bending capacity according to advanced method. In the thesis new values for the thickness of the zero strength layer were given. It was also discovered,that the thickness of the zero strength layer is not constant during the fire exposure, as it is the case for standard fire exposure. Furthermore, some disadvatanges of the reduced cross section method were disclosed. One of them is time when charring and calculation end. According to the simple design method, the end of calculation is associated with the end of charring, since afterwards the load-bearing capacity does not decreases anymore. In reality the temperature inside the cross-section still rises also after charring end, and consequently the load-bearing capacity of the timber beam decreases