Dynamic modeling of the performance of safety electrical circuits during a fire

1
Company reg. Name : WTW Power Solutions
Office Address : Atrium Centrum, al. Jana Pawła II 27, 00-867 Warsaw, Poland
Contact : E-mail: mail@wtwps.com, Internet: http://www.wtwps.com, Phone: +48 22 243 92 50; +48 600 190 084
Company reg. No : 142 36 16 30
Dynamic modeling of the performance of safety
electrical circuits during a fire -
study case based on the data of an actual
building.
Prepared for:
European Copper Institute and
Polish Copper Promotion Centre, members of the
Copper Alliance
Prepared by:
Dr. Jacek Wasilewski
Dr. Wojciech Wiechowski
WTW Power Solutions
Atrium Centrum
Al. Jana Pawła II 27
00-867 Warsaw, Poland
www.wtwps.com
Date / revision:
08 May 2012 / 4.0
Selected Object: Shopping Mall Galeria Tarnovia, ul. Krakowska 149, 33-100 Tarnów, Poland

2
TABLE OF CONTENTS
EX ECU TI V E SU MMA RY 3
STR ES ZC Z ENI E P O PO LSK U 8
ACKNO WL EDGEMENTS / PO DZ IĘK O WA NIA 14
1. Introduction 15
2. Selected object 16
2.1 Main building parameters 16
2.2 Brief description of the building in terms of fire safety issues 17
3. Electric power supply in the building 19
3.1 Fire safety switchboard RGPOŻ 21
3.2 Criteria for selection of fire safety circuits to be used in further investigations 24
3.3 Description of the selected circuits 24
4. Verification of cross section of cables based on IEC 60364 31
4.1 Criterion 1: Current rating & overload protection 31
4.2 Criterion 2: Permissible voltage drop 32
4.3 Criterion 3: Short-circuit strength 33
4.4 Criterion 4: Automatic disconnection of supply 33
4.5 Cable size verification for analyzed fire protection circuits 34
5. Dynamic simulations 40
5.1 Models of selected fire safety circuits 40
5.2 Analyzed scenarios 42
5.3 Simulations 43
5.3.1 Quasi-steady-state 43
5.3.2 Motor starting at no-fire conditions 45
5.3.3 Motor starting at 90 min. fire conditions 46
5.3.4 Single phase-to-ground faults 50
6. Summary and Conclusions 54
REF ER ENC ES 59
APP ENDI X 1 60
APP ENDI X 2 63
APP ENDI X 3 76
source: skyscrapercity.com

3
EXECUTIVE SUMMARY
This document describes the results of a project titled "Dynamic modeling of the performance of safety
electrical circuits during a fire - study case based on the data of an actual building - Galeria Tarnovia".
Overall aim of this study was to demonstrate and emphasize the importance of selection of proper cross-
section of conductors for fire safety circuits that ensure proper start-up and steady-state operation of fire
safety equipment when the electrical resistance of the wires becomes high due to a fire. It must be em-
phasized that the focus of work was to ilustrate system technical performance issues and not the perfor-
mance of particular components, for which a more detailed studies would be normally carried out by
equipemnt manufacturers during the design phase.
This study was based on the design data of an actual large building; a shopping Centre Galeria Tarnovia.
Detailed aims of the study were:
to verify whether or not the existing safety circuits in the building have been designed correctly
according to the existing regulations,
to verify whether or not the existing regulations ensure that the cross-section of fire safety cir-
cuits guarantee desired operation of electrical equipment and circuit protection during fire,
If not, to determine the cross-section of conductors that will allow a satisfactory performance of
the electrical and protection equipment during fire,
to perform the study for safety circuits for which methodology described in [4] for calculation of
circuit parameters has been applied and compare results,
to point out the most sensitive type of equipment to voltage drop.
At first, all of the fire safety circuits in the building are shortly described and a schematic diagram of the
entire fire safety switchboard (RGPOŻ) is shown. Number and types of loads, types of cables, their
lengths and cable routes with respect to fire safety zones are considered. As there are many of the fire
safety circuits supplied from fire safety switchboard RGPOŻ, it would be an extensive and pointless work
to model all of them. In order to select few most representative circuits for further investigations, criteria
for selection are defined and described. After application of these criteria three fire safety circuits are
selected for detailed simulations: circuit supplying sprinkler pumps smoke fans and compressors of hy-
drophore reservoir. Selected circuits are described in detail. The main parameters of the selected circuits
are shown in Table 1 below.
Table 1. Main parameters of investigated fire safety circuits.
Type of fire safety device Rated active
power
Supplying cable type Total cable
length
Water sprinkler pump 200 kW NHXH 4(3)x(2x1x120)/1x120 280 m
Smoke fan 45 kW NHXH 4(3)x(1x95)/1x50 250 m
Compressor of hydrophore re-
servoir
2,2 kW NHXH (5(4)x6) 90 m

4
NHXH - halogen-free, flame-resistant cable. The cable bundle 4x(3)x(2x1x120) / 1x120 denotation
means:
4(3)x(…). – 3 phases and neutral conductors used in LV network (cable bundles connecting
RGPOŻ and local switchboards); phase 1, phase 2 and phase 3 (the neutral point is not used be-
tween the local switchboards and the motors driving the fire safety device)
(2x1x120) – two single-core cables (120 mm2
cross-section) per phase,
/1x120 – one single-core cable (120 mm2
cross-section) as a protective conductor.
Example drawing is shown below.
The cable bundle 4(3)x(1x95)/1x50 denotation means:
4(3)x(…). – phase 1, phase 2, phase 3 and neutral points used in LV network (cable bundles con-
necting RGPOŻ and local switchboards); phase 1, phase 2 and phase 3 (the neutral point is not
used between the local switchboards and the motors driving the fire safety device)
(1x95) – one single-core cable (95 mm2
cross-section) per phase and neutral point,
cross-section) as a protective conductor
The cable (5(4)x6) means the fifth-core cable of 6 mm2 cross-section including phase 1, phase 2, phase
3, neutral and protective point of LV network (cable bundles connecting RGPOŻ and local switchboards).
In case of cable connection between the local switchboard and the motor driving the fire safety device,
phase 1, phase 2, phase 3 and protective point are used (without neutral point).
Temperature change over time was modeled assuming the standard temperature-time curve, shown in
Figure 1.
Figure 1. Time – temperature for standard cellulose fires [4].
ProtectivePhase 3
Phase 2
Phase 1
Neutral

5
Initial verification of the sizing of conductors for selected fire safety circuits based on selection criteria
described in IEC 60364 was made.
The initial verification showed that under normal operating conditions (no fire) all the criteria are satisfied
and cable size selection is correctly made according to state-of-the-art methods and regulations applica-
ble today, for example the IEC standard 60364-4-41. These regulations are applicable to all of the build-
ings built today which are equipped with fire safety circuits and Galeria Tarnovia is an example of such
typical modern building.
The initial verification shows that for installations designed according to the existing regulations, for a
maximum value of temperature for the standard fire temperature–time curve, the maximum allowable
times of automatic disconnection of supply may be often exceeded, and the maximum levels of voltage
drop along circuits are often exceeded as well.
These results of simplified verification indicate that when the temperature of wires of the supplying cir-
cuits is high due to a fire, problems with operation of fire safety equipment as well as with fast discon-
nection of fire safety circuits by protective relays may occur. Therefore, in the next step the steady-state
and dynamic performance of selected fire safety equipment and of the protective relays have been veri-
fied by detailed dynamic simulations with the use of a professional simulation tool. For this study dynamic
models of the loads were prepared and the dependence of resistance of cables was automatically varied
according to the standard fire time-temperature curve.
Assumed study scenarios include both normal conditions (no fire) and fire conditions (90 min after fire
breaks out). The following simulation scenario types have been assumed:
during steady state operation of fire safety devices a fire breaks out and develops in the car park.
Performance of all devices is monitored.
equipment start-up during normal conditions and 90 min after a fire breaks out.
Phase-to-ground short-circuits at a switchboard or the terminals of the device both under no fire
conditions and 90 min after a fire breaks out.
For all the investigated fire safety circuits, RMS time domain simulations were performed and all the rele-
vant values like voltages currents, electric active and reactive powers as well as speed and mechanical
torque have been monitored.
Performed simulations confirm that:
Times of automatic disconnection of supply under fire conditions in a number of cases exceed al-
lowable limits,
Voltage drop under fire conditions is large, which results in large currents that cause induction
motor overload but not enough for the activation of thermal protection in the investigated scena-
rios,
Motor starting under fire conditions is often impossible as currents are large due to decrease of
supplying voltage as well as motor stall. The large current result in circuit tripping (despite that
soft-starters are used)
All the results are summarized in Table 2.

6
Table 2. Dynamic simulation results under normal and PH90 conditions.
Existing cable cross-sections
Circuit: Sprinkle pump Smoke fan Compressor
Existing cable cross-section 4(3)x(2x1x120)
/1x120
4(3)x(1x95)
/1x50
(5(4)x6)
Normal conditions
Cable resistance 0,0190 0,0236 0,2449
Steady state voltage drop 4,6% 2,4% 0,48%
Steady state current 1,0 p.u. 0,96 p.u. 0,87 p.u.
Steady state active power 1,0 p.u. 1,0 p.u. 1,0 p.u.
Time of supply disconection
(RGPOŻ/local switchb.)
0,070 s/0,047 s 0,007 s/0,008 s 0,003 s/0,003 s
Motor starting OK OK OK
PH90 fire conditions
Cable resistance 0,1013 0,1257 1,3042
Steady state voltage drop 14,9% 8,3% 1,8%
Steady state current 1,10 p.u. 0,95 p.u. 0,86 p.u.
Steady state active power 0,94 p.u. 0,94 p.u. 0,96 p.u.
32,707 s/12,588 s 3,988 s/0,38 s 6,246 s/2,980 s
Motor starting doesn’t start OK doesn’t start
Increased cable cross-sections
Circuit: Sprinkle pump Smoke fan Compressor
New cable cross-section 4(3)x(3x1x120)
/2x1x120
- (5(4)x16)
Normal conditions
Cable resistance 0,0013 - 0,2449
Steady state voltage drop 3,5% - 0,18%
Steady state current 0,99 p.u. - 0,86 p.u.
Steady state active power 1,0 p.u. - 1,0 p.u.
0,003 s/0,003 s 0,003 s/0,003 s
Motor starting OK - OK
PH90 fire conditions
Cable resistance 0,06751 - 0,4867
Steady state voltage drop 9,4% - 0,7%
Steady state current 1,02 p.u. - 0,85 p.u.
Steady state active power 0,98 p.u. - 0,97 p.u.
0,106 s/0,358 s - 1,531 s/0,047 s
Motor starting OK - OK
In case of other buildings, if all fire safety circuits are designed and built according to IEC 60364 stan-
dards, the fire safety equipment should operate properly in the steady-state. Thus, the fire safety devices
ensure its functions. However there are many factors affecting the motor circuit operation such a supply-
ing voltage in main switchboard, cable length and cross-section, motor apparent power, Speed-Torque

7
characteristic of a motor–driven device and protection device settings. Therefore each designed circuit
should be investigated individually.
The most sensitive type of fire safety equipment to voltage drop out of the three analyzed were the com-
pressors. Especially, during motor starting a risk of motor stall is observed. In case of large voltage drop
at the motor terminals, the initial mechanical torque is higher than the initial electric torque and the mo-
tor will not start.
Comparing the verification method described in [4] and the dynamic simulation method one can observe
that:
The results of voltage drop evaluation in the steady-state operation for both verification methods
are very close,
The simulation results of time of automatic disconnection of supply are consistent with the results
obtained from the method [4] only for two circuits, i.e. sprinkler pump and compressor. In case
of the smoke fan circuit, dynamic simulation results show that required time of disconnection is
fulfilled even under PH90 conditions. In contrast, the results obtained from method [4] show that
the condition of automatic disconnection of supply is not fulfilled under PH90 conditions when as-
suming that the tripping devices are electromagnetic releases or relays. If this aforementioned
assumption is rejected, all the results obtained from both methods of cable size verification (per-
formed dynamic simulation and the method described in [4]) are consistent each other.
The results of motor starting simulation as well as results of voltage drop evaluation based on
method [4] show that motors driving the sprinkler pump and compressor will not start under
PH90. A difference in results is observed for the smoke fan motor. Based on the method de-
scribed in [4], the obtained results show the voltage drop exceeds the permissible value of 10%
and the motor will not start. The dynamic simulation (including a softstart model) results show
that the motor will start properly under PH90 conditions.
Cable size verification method described in [4] may be used as a technical knowledge source for electrical
engineers designing electrical installations including fire safety circuits. Simulations with dynamic models
of electrical installation components may show more detailed results.
It is shown that in order to ensure the correct operation of fire safety equipment during a fire, it is essen-
tial to:
increase cross-section of conductors of the supplying cables - to decrease voltage drop
increase cable cross-section of the conductors of supplying cables and of the protective conductor
as well (or apply supplementary equipotential bonding) - to decrease phase-to-ground imped-
ance and ensure sufficiently fast automatic disconnection of supply
set thermal protection current settings not less than 1,1 times rated motor current in order to
minimize the likelihood of circuit disconnection in cases of slight motor overloading
The overall conclusion and recommendation of this study is that it is necessary to elaborate and evaluate
standards and recommendations for fire safety circuits that will include clear cable sizing criteria for the
expected fire conditions.

8
STRESZCZENIE PO POLSKU
Niniejszy raport przedstawia wyniki badań prowadzonych w ramach projektu badawczego zatytułowanego
„Modelowanie dynamiczne pracy obwodów elektrycznych zasilających odbiorniki pożarowe w warunkach
pożaru – studium przypadku istniejącego budynku Galeria Tarnovia”.
Ogólnym celem przeprowadzonych badań było wykazanie i podkreślenie znaczenia prawidłowego doboru
przekroju kabli w obwodach pożarowych zapewniającego prawidłowy rozruch urządzeń pożarowych, ich
pracę w stanie ustalonym oraz skuteczności warunku samoczynnego wyłączenia zasilania jako środka
ochrony przeciwporażeniowej działającego przy uszkodzeniu izolacji. Proces doboru przekroju przewodów
rozważano w warunkach zwiększonej wartości rezystancji kabli zasilających ze względu na wzrost
temperatury otoczenia spowodowanej pożarem. Przeprowadzone badania oparte zostały na danych
pochodzących z dokumentacji powykonawczej istniejącego budynku handlowo-usługowego o dużej
kubaturze, tj. centrum handlowo-usługowe Galeria Tarnovia.
W ramach szczegółowych celów postawionych w niniejszych badaniach należało:
sprawdzić, czy wybrane istniejące obwody zasilające odbiorniki pożarowe zostały prawidłowo
zaprojektowane zgodnie z aktualnymi normami,
sprawdzić, czy aktualne normy zawierają odpowiednie wymagania odnośnie doboru przekroju
kabli pożarowych, zapewniające właściwą pracę urządzeń pożarowych w warunkach pożaru,
w przypadku stwierdzenia nieprawidłowości pracy urządzeń pożarowych, sprawdzić dla jakich
przekrojów kabli będą zapewnione właściwe warunki pracy odbiorników pożarowych,
przeprowadzić obliczenia dotyczące doboru przekroju kabli pożarowych uwzględniając metodykę
zaproponowaną w publikacji [4] i porównać uzyskane wyniki z wynikami przeprowadzonych
symulacji działania obwodów w warunkach pożaru,
wskazać na urządzenia ochrony pożarowej najbardziej wrażliwe na zmniejszoną wartość napięcia
zasilania ze względu na wzrost rezystancji kabli.
W pierwszej kolejności, opisano pokrótce wszystkie istniejące w rozważanym budynku obwody pożarowe i
przedstawiono w postaci schematu rozdzielnicy głównej pożarowej (RGPOŻ). Przeanalizowano zarówno
liczbę i rodzaj zasilanych urządzeń ochrony przeciwpożarowej jak również rodzaj, przekroje, długości oraz
sposób i trasę ułożenia kabli zasilających te urządzenia, w odniesieniu do stref pożarowych budynku. Ze
względu na duża liczbę istniejących obwodów pożarowych w rozpatrywanym budynku (zasilanych z
rozdzielnicy RGPOŻ), wybrano najbardziej reprezentatywne obwody, które posłużyły dalszym badaniom.
Kryteria wyboru obwodów do badań zostały ściśle zdefiniowane i szczegółowo opisane. Ostatecznie, do
badań wybrano trzy obwody, które zasilają pompę tryskaczową, wentylator oddymiający oraz kompresor
zbiornika hydroforu. Wybrane obwody szczegółowo opisano w niniejszym raporcie. Główne parametry
wybranych obwodów przedstawiono w Tabeli 1.1.
Tabela 1.1. Główne parametry rozważanych obwodów pożarowych.
Rodzaj urządzenia pożarowego Moc czynna
znamionowa
Kabel zasilający Całkowita
długość kabla
Pompa tryskaczowa 200 kW NHXH 4(3)x(2x1x120)/1x120 280 m
Wentylator oddymiający 45 kW NHXH 4(3)x(1x95)/1x50 250 m
Kompresor zbiornika hydroforu 2,2 kW NHXH (5(4)x6) 90 m

9
NHXH – nie zawiera halogenu, ognioodporny
Oznaczenie wiązki kabli 4x(3)x(2x1x120)/1x120 należy rozumieć następująco:
4(3)x(…). – faza 1, faza 2, faza 3 oraz punkt neutralny sieci nn (kable łączące rozdzielnicę RGPOŻ
oraz rozdzielnice lokalne pracują w układzie trzech faz i punktu neutralnego); faza 1, faza 2 oraz
faza 3 (nie stosuje się punktu neutralnego między rozdzielnicami lokalnymi a silnikiem
napędzającym urządzenie pożarowe)
(2x1x120) – dwa jednożyłowe kable (o przekroju 120 mm2
) na fazę i punkt neutralny,
/1x120 – kabel jednożyłowy (o przekroju 120 mm2
) użyty jako przewód ochronny.
Pokazane na rysunku ponizej.
The cable bundle 4(3)x(1x95)/1x50 denotation means:
4(3)x(…). – faza 1, faza 2, faza 3 oraz punkt neutralny sieci nn (kable łączące rozdzielnicę RGPOŻ
oraz rozdzielnice lokalne pracują w układzie trzech faz i punktu neutralnego); faza 1, faza 2 oraz
faza 3 (nie stosuje się punktu neutralnego między rozdzielnicami lokalnymi a silnikiem
napędzającym urządzenie pożarowe)
(1x95) – jednożyłowy kabel (o przekroju 90 mm2
) na fazę i punkt neutralny
/1x50 – kabel jednożyłowy (o przekroju 50 mm2) użyty jako przewód ochronny.
Oznaczenie kabla (5(4)x6) należy rozumieć jako kabel pięciożyłowy zawierający fazę 1, fazę 2, fazę 3,
punkt neutralny sieci nn oraz punkt ochronny (kabel łączący RGPOŻ i rozdzielnicę lokalną). W przypadku
połączenia między rozdzielnicą lokalną a silnikiem napędzającym urządzenie pożarowe, uwzględnia się
jedynie trzy fazy i przewód ochronny (nie używa się punktu neutralnego).
Zmianę temperatury otoczenia w czasie rozwoju pożaru zamodelowano w oparciu o standardową krzywą
tempetarura-czas, którą przedstawiono na Rys. 1.1.
ProtectivePhase 3
Phase 2
Phase 1
Neutral

10
Rys. 1.1. Standardowa charakterystyka temperaturowo-czasowa pożaru.
Dokonano wstępnej weryfikacji przekrojów kabli zasilających wybrane odbiorniki pożarowe w oparciu o
kryteria z odpowiednich arkuszy normy IEC 60364.
Wyniki wstępnej weryfikacji przekrojów kabli pożarowych wskazują, że dla stanu normalnego (brak
pożaru w budynku) wszystkie kryteria doboru przekrojów kabli są spełnione w odniesieniu do wymagań
aktualnych wytycznych, np. wieloarkuszowej normy IEC 60364. Zatem dla takich warunków, wytyczne te
są adekwatne dla projektowania instalacji elektrycznych w nowych budynkach wyposażonych w
urządzenia ochrony przeciwpożarowej, takich jak nowoczesne centrum handlowo-usługowe Galeria
Tarnovia.
Rezultaty wstępnej weryfikacji przekrojów kabli zasilających odbiorniki pożarowe pokazują, że przy
uwzględnieniu aktualnych wytycznych projektowych, dla maksymalnej temperatury otoczenia po czasie
90 min. wg standardowej krzywej pożarowej (Rys. 1.1), mogą być przekroczone zarówno czasy
samoczynnego wyłączenia zasilania, jak również wartości spadku napięcia wzdłuż kabli.
Wyniki przeprowadzonej weryfikacji przekrojów kabli wskazują, że w przypadku wysokiej temperatury żył
kabli spowodowanej pożarem, mogą wystąpić problemy związane z zapewnieniem poprawnej pracy
urządzeń pożarowych oraz niedotrzymaniem w tych obwodach wymaganego czasu samoczynnego
wyłączenia zasilania. Z tego względu, w następnym kroku badań, przeprowadzono szczegółową symulację
pracy rozpatrywanych obwodów zarówno w stanie ustalonym, jak i przejściowym uwzględniając nie tylko
modele odbiorników, ale także zabezpieczeń nadprądowych. Stosowne symulacje przeprowadzono przy
użyciu profesjonalnego narzędzia. W tym celu zbudowano szczegółowe modele elementów obwodów
uwzględniając zależność rezystancji kabli zasilających odbiorniki pożarowe od temperatury otoczenia
zmieniającej się wg standardowej krzywej pożarowej.
Dla celów badań założono odpowiednie scenariusze zdarzeń uwzględniające warunki normalne pracy
obwodów pożarowych (brak pożaru) oraz warunki pożaru (90 min. od momentu powstania pożaru).
Rozpatrzono następujące klasy scenariuszy symulacji:

11
podczas pracy ustalonej urządzeń pożarowych, powstaje pożar w garażu (na trasie kabli
zasilających urządzenia), który trwa 90 min. Analizuje się pracę wszystkich rozpatrywanych od-
biorników pożarowych,
następuje rozruch silników napędzających urządzenia pożarowe, zarówno w warunkach
normalnych oraz po 90 min. od powstania pożaru w garażu,
następuje zwarcie doziemne (L-PE) na końcu kabla zasilającego lokalną rozdzielnicę oraz na
końcu kabla zasilającego urządzenie pożarowe, zarówno w warunkach normalnych oraz po 90
min. od powstania pożaru w garażu.
Dla wszystkich rozpatrywanych obwodów, przeprowadzono symulacje w czasie przebiegów wartości
skutecznych (RMS) różnych wielkości fizycznych, takich jak napięcie, prąd, moc czynna i bierna oraz
moment i prędkość silników napędzających urządzenia pożarowe.
Wyniki przeprowadzonych symulacji potwierdziły, że:
W większości przypadków, w warunkach pożaru na trasie kabli, czasy samoczynnego wyłączenia
zasilania przekraczają dopuszczalne wartości,
Spadki napięcia w warunkach pożaru osiągają wysokie wartości. Mogą występować nieznaczne
przeciążenia silników, niepowodujące przy tym zadziałania zabezpieczeń przeciążeniowych,
Rozruch silników podczas pożaru na trasie kabli zasilających te silniki jest często niemożliwy.
Występują zjawiska utknięcia silnika oraz pracy silnika w niestabilnym punkcie charakterystyki
prędkość-moment ze względu na niską wartość napięcia zasilającego, co skutkuje prądami
przeciążeniowymi powodującymi aktywację zabezpieczeń nadprądowych i wyłączenie całego
obwodu (nawet pomimo stosowania układów łagodnego rozruchu).
Wyniki symulacji zostały podsumowane w Tabeli 2.1

12
Tabela 2.1. Wyniki symulacji dynamicznych w warunkach normalnych oraz PH90.
Istniejące przekroje kabli
Obwód: Pompa tryskaczowa Wentylator
oddymiający
Kompresor
Istniejący przekrój kabla 4(3)x(2x1x120)
/1x120
4(3)x(1x95)
/1x50
(5(4)x6)
Warunki normalne
Rezystancja kabla 0,0190 0,0236 0,2449
Spadek napięcia w stanie
ustalonym
4,6% 2,4% 0,48%
Prąd w stanie ustalonym 1,0 p.u. 0,96 p.u. 0,87 p.u.
Moc czynna w stanie ustalonym 1,0 p.u. 1,0 p.u. 1,0 p.u.
Czas samoczynnego wyłączenia
zasilania (RGPOŻ/rozdzielnica
lokalna
0,070 s/0,047 s 0,007 s/0,008 s 0,003 s/0,003 s
Rozruch silnika OK OK OK
Warunki PH90
Rezystancja kabla 0,1013 0,1257 1,3042
ustalonym
14,9% 8,3% 1,8%
Prąd w stanie ustalonym 1,10 p.u. 0,95 p.u. 0,86 p.u.
Moc czynna w stanie ustalonym 0,94 p.u. 0,94 p.u. 0,96 p.u.
lokalna
32,707 s/12,588 s 3,988 s/0,38 s 6,246 s/2,980 s
Rozruch silnika nie startuje OK nie startuje
Zwiększone przekroje kabli
Obwód: Pompa tryskaczowa Wentylator
oddymiający
Kompresor
Istniejący przekrój kabla 4(3)x(3x1x120)
/2x1x120
- (5(4)x16)
Warunki normalne
Rezystancja kabla 0,0013 - 0,2449
ustalonym
3,5% - 0,18%
Prąd w stanie ustalonym 0,99 p.u. - 0,86 p.u.
Moc czynna w stanie ustalonym 1,0 p.u. - 1,0 p.u.
lokalna
0,003 s/0,003 s 0,003 s/0,003 s
Rozruch silnika OK - OK
Warunki PH90
Rezystancja kabla 0,06751 - 0,4867
ustalonym
9,4% - 0,7%
Prąd w stanie ustalonym 1,02 p.u. - 0,85 p.u.
Moc czynna w stanie ustalonym 0,98 p.u. - 0,97 p.u.
lokalna
0,106 s/0,358 s - 1,531 s/0,047 s
Rozruch silnika OK - OK

13
W przypadku innych budynków, zakładając, że wszystkie obwody pożarowe zostały zaprojektowane
zgodnie z wytycznymi wieloarkuszowej normy IEC 60364, urządzenia ochrony przeciwpożarowej powinny
pracować poprawnie w stanie pracy ustalonej. Tym samym, urządzenia te będą pełniły swoją funkcję
podczas zakładanych 90 min. pożaru na trasie kabli zasilających te urządzenia. Z drugiej strony, należy
mieć świadomość, że istnieje wiele czynników mających wpływ na poprawność pracy tego typu urządzeń
silnikowych, takich jak napięcie na szynach rozdzielnicy głównej, długość i przekrój żył kabla, moc
znamionowa silnika, charakterystyka prędkość-moment silnika oraz ustawienia zabezpieczeń w obwodzie.
Zatem, każdy obwód powinien być rozpatrywany indywidualnie.
Stwierdzono, że najbardziej wrażliwymi urządzeniami ochrony pożarowej na zmniejszoną wartość
napięcia zasilania z trzech rozpatrywanych typów urządzeń są kompresory, zwłaszcza podczas rozruch
silnika, gdzie zaobserwowano jego utknięcie. Zbyt mała wartość napięcia zasilającego silnik spowodowała,
że początkowy (przy postoju silnika) moment elektryczny był mniejszy od początkowego momentu
mechanicznego i tym samym silnik nie wystartował.
Porównując wyniki weryfikacji, której metodyka została przedstawiona w publikacji [4], z wynikami
przeprowadzonych symulacji pracy urządzeń pożarowych zaobserwowano, że:
Wyniki szacowania spadku napięcia w stanie ustalonym pracy odbiorników pożarowych dla obu
metod weryfikacji są bardzo bliskie,
Wyniki otrzymane podczas symulacji czasów samoczynnego wyłączenia zasilania są zgodne z
wynikami metody opisanej w [4] dla dwóch z trzech rozważanych obwodów pożarowych, tj.
zasilających pompę tryskaczową i kompresor. W przypadku obwodu wentylatora oddymiającego,
wyniki symulacji stanów dynamicznych pokazują, że czas samoczynnego wyłączenia zasilania w
tym obwodzie mieści się w granicach wymaganych wartości nawet w warunkach PH90. Z drugiej
strony, otrzymane wyniki weryfikacji kryterium samoczynnego wyłączenia zasilania, przy
zastosowaniu metodyki [4], wskazują na niespełnienie tego warunku podczas 90 min. od czasu
powstania pożaru, przy czym przyjęto założenie, że urządzeniami powodującymi samoczynne
wyłączenie zasilania są jedynie wyzwalacze i przekaźniki elektromagnetyczne (zabezpieczenia
zwarciowe). W przypadku odrzucenia tego założenia, wyniki obu metod weryfikacji są ze sobą
zgodne.
Wyniki symulacji dynamicznej rozruchu silników napędzających urządzenia ochrony
przeciwpożarowej oraz szacowania spadku napięcia wg metody opisanej w [4] pokazują, że silnik
kompresora nie uruchomi się w warunkach PH90. Obserwuje się różnice w wynikach obu metod
jedynie dla obwodu wentylatora oddymiającego. Na podstawie metody opartej na publikacji [4],
wyniki otrzymane wskazują, że spadek napięcia podczas warunków PH90 przekracza wartość
10% i tym samym silnik nie powinien wystartować. Przeprowadzone symulacje dynamiczne
(uwzględniające modele układów łagodnego rozruchu – soft-startów) pokazują jednak, że silnik
kompresora uruchomi się właściwie nawet w momencie 90 min. po powstaniu pożaru na trasie
kabla zasilającego kompresor.
Metoda doboru przekroju kabli pożarowych opisana w publikacji [4] może służyć jako źródło wiedzy
technicznej dla inżynierów elektryków projektujących instalacje elektryczne zawierające obwody
zasilające urządzenia ochrony przeciw pożarowej. Z kolei, symulacje oparte na szczegółowych modelach
dynamicznych elementów instalacji mogą pokazać bardziej szczegółowe wyniki
W niniejszym raporcie pokazano, że aby zapewnić właściwą pracę obwodów pożarowych w warunkach
pożaru na trasie kabli zasilających urządzenia pożarowe niezbędne jest:

14
zwiększyć przekrój żył kabla w celu zmniejszenia spadku napięcia,
zwiększyć przekrój żył czynnych i ochronnej kabla (lub zastosować miejscowe połączenia
wyrównawcze) w celu zmniejszenia impedancji pętli zwarcia i zapewnienia wymaganego czasu
samoczynnego wyłączenia zasilania,
ustawić prąd zabezpieczenia przeciążeniowego nie mniejszy niż 1,1 wartości prądu znamionowego
silnika w celu zmniejszenia prawdopodobieństwa wyłączenia przez zabezpieczenie w przypadku
nieznacznych przeciążeń silnika.
Po przeprowadzeniu badań, głównym wnioskiem, jaki się nasuwa jest pilna potrzeba opracowania
dokumentów normatywnych zawierających wytyczne dotyczące wymiarowania kabli zasilających odbiory
pożarowe w warunkach pożaru, jaki się oczekuje na trasie kabli zasilających urządzenia pożarowe.
ACKNOWLEDGEMENTS / PODZIĘKOWANIA
The authors would like to express their gratefulness to Mr. Waldemar Sikorski, Technical Director, TK
Development Polska and Mr. Tomasz Mądrzyk, Technical Manager, CBRE Property Management Poland
sp. z o.o., for agreeing to use and for making available all the design data of Galeria Tarnovia Shopping
Centre. The authors would also like to thank Mr. Michał Ramczykowski, CEO, Polish Copper Promotion
Centre, Mr. Roman Targosz, Director for Electrical Projects, Polish Copper Promotion Centre and Mr.
Hans De Keulenaer, Electricity & Energy Programme Manager, European Copper Institute, for sponsoring
this research project.
Autorzy skladają podziękowania Panu Waldemarowi Sikorskiemu, Dyrektorowi Technicznemu,
TK Development Polska I Panu Tomaszowi Mądrzykowi, Technical Manager, CBRE Property Management
Poland sp. z o.o., za zgodę na wykożystanie oraz za udostępnienie dokumentacji projektowej
powykonawczej centrum handlowego galeria „Tarnovia”. Autorzy chcieliby również złożyć podziękowania
dla Polskiego Centrum promocji Miedzi, Panu Michałowi Ramczykowskiemu, Prezesowi Zarządu i Panu
Romanowi Targoszowi, Kierownikowi ds. Projektów Elektrycznych, a także Panu Hans De Keulenaer,
Electricity & Energy Programme Manager, European Copper Institute, za współpracę i za finansowanie
tego projektu badawczego.

15
1. Introduction
The maintenance of safety services in a building during an emergency is vital if the situation is to be con-
tained and the lives of occupants and emergency workers reserved. The first objective is to provide, as
far as possible, safe evacuation routes from the affected areas and to ensure that all services required for
fighting the fire are maintained for the required time. Another objective may be to ensure the preserva-
tion of services in areas of the building not directly affected by fire to provide a safe environment for eva-
cuees from affected areas, pending complete evacuation, or the end of the incident. Finally, the preserva-
tion of property is important.
The time for which safety services are required to operate depends on the use of the building and the
number and type of occupants involved. While it may be feasible to evacuate a dwelling or a small office
building in a short time, larger or higher buildings, or those used by the public will take much longer. The
longer that safety services remain operational, the better the chance of bringing a fire under control and
reducing property damage.
Cables that are exposed to fire while being expected to retain their functionality and provide power to
essential equipment at another location must be appropriately selected and sized to take account of the
increased electrical resistance at elevated temperature. Manufacturers offer cables and accessories that
will survive a standard cellulose fire for 30, 60 or 90 minutes when correctly specified and installed.
Cables, including fire safety cables, are specified in terms that reflect their normal duty conditions; design
parameters under fire conditions are rarely, if ever, specified.
Insufficient cross-section of the cables in fire conditions may result in resistance becoming so large that
the voltage drop across it will hamper proper operation of the electrical equipment that must be function-
al during fire. Moreover, the increased resistance of the safety circuit will cause fault currents to have
lower amplitudes, which will result in delayed disconnection of the circuit by a circuit breaker. This delay
will pose an increased electric shock risk for the people still present in the building and may also ignite
fires in other, previously safe areas of the building. 1
The overall aim of this study titled “Dynamic modeling of the performance of safety electrical circuits
during a fire - study case based on the data of an actual building”, is to demonstrate and emphasize the
importance of selection of proper cross-section of conductors for fire safety circuits that ensure proper
start-up and steady-state operation of fire safety equipment when the electrical resistance of the wires
becomes high due to a fire. The study is based on the design data of an actual large building, which is a
newly build shopping Centre Galeria Tarnovia, located in Southern Poland, in the city of Tarnow.
Detailed aims of this study are:
to verify whether or not the selected existing safety circuits in the building have been designed
correctly according to the existing regulations,
to verify whether or not the existing regulations ensure that the cross-section of fire safety cir-
cuits guarantee desired operation of electrical equipment and circuit protection during fire,
If not, to determine the cross-section of conductors that will allow a satisfactory performance of
the electrical and protection equipment during fire,
1
Introduction text is based on [4]

16
to perform the study for safety circuits for which methodology described in [4] for calculation of
circuit parameters has been applied and compare results,
to point out the most sensitive type of equipment to voltage drop.
2. Selected object
The object selected for the study is a Shopping Mall "Galeria Tarnovia":
Galeria Tarnovia
ul. Krakowska 149
33-100 Tarnów
This description has been made based on:
Site visit, which took place on the 12 December 2011,
Hard copy of essential documents, plans and diagrams (as-built Docs. No 08022/36, Nov 2009),
As-built AutoCAD files received on the 21 December 2011 (as-built Docs. No 08022/36, Nov
2009).
Following is a list of the most important documents obtained and / or copied:
General description of the electrical installations,
Fire development scenarios (developed Sept. 2009)
Tables of devices with rated power,
Schematic diagrams and plans
2.1 Main building parameters
Area: 36.500 m2 (2 levels)
Cark park area: 11.500 m2
Number of shops: 115
Year of opening: 2009
3 fire zones separated by fire resistant walls REI60 and doors EI30:
I - 11.092 m2 ground floor
- 2.004 m2 first floor
II - 6.526 m2 ground floor
- 3.820 m2 first floor
III - 11.511 m2 car park
As the building has three levels above ground, it is classified it as of Average Height (budynek
średniowysoki SW).

17
A plan of the ground floor of the centre is shown in Figure 2
Figure 2 Plan of the Shopping Centre Galeria Tarnovia.
2.2 Brief description of the building in terms of fire safety issues
In terms of fire there are mostly solid materials in the building. No dangerous materials will be stored
there except small amounts of chemicals, cosmetics and alcohols, all in small hermetic containers.
Fire density does not exceed 500 MJ/m2 and except some utility rooms and storage where it reaches
1000 MJ/m2.
Due to its purpose (shopping mall) Galeria Tarnovia is classified as ZL1 category for danger for humans.
(ZL1 - defined as a place that contains facilities for the simultaneous presence of more than 50 people,
which are not their regular users, and are not intended primarily for use by people with limited mobility).
Due to existence of large quantities of flammable materials (clothes, etc) that can result in rapid devel-
opment of fire and thick smoke the fire hazard for the building is characterized as LARGE.

18
Number of people simultaneously present in fire safety zones:
Zone I - 2210 people (ground floor 1980 and 230 first floor)
Zone II - 1715 people (ground floor 1205 and 510 first floor)
Zone III - 880 people (car park)
Car park is made in fire resistant class B and the rest of the mall in class C1. Roof of the car park is fire
resistant REI120.
Fire resistance of building elements above the car park is:
Main load-bearing structure - R60
Ceiling - REI60
Roof construction - R152
Roof covering - E15
Internal walls - EI15
External walls - EI30
The building is equipped with the following fire safety installations:
Water sprinklers
Fire water supply (for the hydrant hoses)
Fire alarm system
Emergency evacuation lighting
automatic smoke exhaust devices
acoustic warning system
Jet ventilation system in the car park
Water sprinklers are located in the entire building except:
All electrical equipment rooms (with transformers, switchboards, etc)
Storage rooms in cark park level
Staircases
Toilets

19
3. Electric power supply in the building
Electrical power supply for the Centre is provided by two 15 kV cable lines of the type 3xXRUHAKXS
1X240mm2, where one of the lines is the main supply designed for active power of 750 kW and a redun-
dant power supply designed for active power of 350 kW. The two MV lines are lead via different routes to
own MV/LV substation of the Centre ZE-1 where two transformers are installed (Ratio 15/0,4 kV; Dyn5,
Power 1 MVA, Short-circuit voltage Usc = 6%).
The MV part of substation is split into the main and the redundant power supply parts. Both parts are
composed of the incoming MV line field, Metering field and transformer supply field (equipped with trans-
former circuit breaker CB).
The LV part of the substation is split into 3 sections, as shown in Figure 3. Sections 1 and 2 are supplied
each by one of the 1 MV 15/0,4 kV transformers (with main circuit breakers protecting both sections) and
through cables YKY 4x(4x1x240 mm2). Section 3 is for supply of the fire safety equipment and it is sup-
plied directly from 0,4 kV side of both transformers (before main circuit breakers for sections 1 and 2).
All circuit breakers and switchboards have been manufactured by Schneider Electric. All circuits have
been made as TN-S (3L+N+PE). Protected with over-current and short-circuit protection, differential pro-
tection (0,03 A), overvoltage protection B and C type. Assumed maximal voltage drop from transformer
LV nodes are 5% for power circuits and 3 % for lighting. Minimal cross-section for wires are 1,5 mm2 for
lighting and 2,5 mm2 for other circuits.

20
Figure 3. Schematic diagram of the main switchboard RGINN with power supply for the fire safety switch-
board RGPOŻ taken directly from the LV side of the MV/LV transformers (source: AutoCAD files provided
by building administrator).

21
3.1 Fire safety switchboard RGPOŻ
One of main distribution switchboard sections is for supply of the fire safety equipment and it is supplied
directly from 0,4 kV side of both transformers. This section is called RGPOŻ (Rozdzielnica Główna
Pożarowa, which can be translated to Main Fire Switchboard). A schematic diagram of the RGPOŻ is
shown in Figure 4.
Supply of the RGPOŻ switchboard has been made using fire resistant cables E90. All other fire safety
equipment has been made using EI30 (if along water sprinklers), EI90 in other places. Walls of cable
ducts are fire resistant EI120 with doors EI60.
Most of the circuits have been laid in cable trays. Calculation and selection of circuits cross-section was
based on IEC 364-5-523.
All cables for supply of fire safety circuits should ensure continuity of supply during fire for the required
time of operation of particular piece of equipment but no shorter than 90 minutes. Exception are circuits
for smoke dampers and laid in zones protected by water sprinklers.
For RGPOŻ switchboard the parallel supply redundancy is applied using automatic transfer switching
equipment. RGPOŻ is protected in each power circuit with the use of circuit breaker of NT16H1 3p Micro-
logic 5.0A type (Merlin Gerin). The 180 kvar shunt capacitor is installed for power factor correction pur-
pose. The overvoltage protection with the use of varistor technique is also applied.
The Main Fire Switchboard RGPOŻ supplies all fire safety circuits in the Centre, i.e.:
Outdoor lighting fittings of 10 kW total installed power for fire asses roads; indirectly supplied
through SB32a switchboard section; the feeder is protected with the use of NG125N C32A circuit
breakers (Merlin Gerin); the longest line section between RGPOŻ and the farthest lighting fitting
is about 100 m.
Four smoke fans of 45 kW rated power; indirectly supplied through RODG1, RODG2, RODG3 and
ODG4 switchboard respectively located in the ventilation control rooms; the feeders are protected
with the use of NS630N 630A STR23SE circuit breakers (Merlin Gerin); the longest line section
between RGPOŻ and the farthest smoke fan is above 290 m.
Sprinkler water pump of 200 kW rated power and its control equipment; indirectly supplied
through RPT1 and RT switchboards (control and protection equipment) located in the sprinkler
room; the feeders are protected with the use of NS160N 160A STR22SE and NG125N C32A circuit
breakers (Merlin Gerin) respectively; the longest line section between RGPOŻ and the farthest
device (the sprinkler pump) is above 280 m.
Central storage battery of 20 kW rated power (for crucial control equipment supply) located near
(a few meters) the RGPOŻ switchboard; the feeder is protected with the use of NG125N C63A cir-
cuit breaker (Merlin Gerin);
Compressor of 2,2 kW rated power as an element of the hydrophore reservoir; indirectly supplied
through the separated section of the hydrophore switchboard RH located in the hydrophore room;
the feeder is protected with the use of NG125N C32A circuit breaker (Merlin Gerin). The line
length between RGPOŻ and the compressor is above 90 m.

22
Telecommunications equipment of 6 kW installed power located in the telecommunications room;
the feeder is protected with the use of NG125N C32A circuit breaker (Merlin Gerin). The line
length between RGPOŻ and the telecommunications box is above 60 m.
Five air admission fans of 7,5 15 kW rated power and two smoke exhausting fans of 10,5 kW; in-
directly supplied through the fire ventilation switchboard RWN located in the fan control rooms;
the feeder is protected with the use of NS400N 400A STR23SE circuit breaker (Merlin Gerin); the
longest line section between RGPOŻ and the farthest fan is above 220 m.
Four 0,5 kW electric actuators of the fire gateways located in the car park; the feeder is protected
with the use of C60L C20A circuit breaker (Merlin Gerin); the longest line section between RGPOŻ
and the farthest gateway actuator is above 70 m.
Smoke flaps and fire signaling and monitoring equipment of 6 kW total installed power; indirectly
supplied through RM switchboard located in the monitoring room; the feeder is protected with the
use of NG125N C32A circuit breaker (Merlin Gerin); the longest line section between RGPOŻ and
the farthest smoke flaps actuator is about 100 m.
Six spare switch bays.

23
Figure 4. Main fire switchboard RGPOŻ (source: AutoCAD files provided by building administrator).
NT16H
1
1600
A
1250
A
3
P
1
05x2
5NHX
HRGPO¯/
1
NG125
N
C32
A3
P
SB32
a
6
74x1x70/1x3
5
RGPO¯/
2
NHX
HRGPO¯/
3
RGPO¯/
4
5
64x1x95/1x5
0 NHX
H
5
64x1x70/1x3
5 NHX
H
NS160
N
160
A
STR22S
E
3
P
RODG
1
RODG
2
RODG
3
2
25x1
6
RGPO¯/
5
NHX
HRGPO¯/
6
RGPO¯/
7
20
04x(2x1x120)/1x12
0 NHX
H
4
95x3
5NHX
H
NS630
N
630
A
STR23S
E
3
P
NG125
N
C32
A3
P
RODG
4
RPT
1
ROZDZIELNI
A
4
5x
6NHX
HRGPO¯/
9
RGPO¯/1
0
2
05x3
5NHX
H
--
---
---
-
NG125
N
C63
A3
P
NG125
N
C32
A3
P
REZERW
A
CENTRALN
A
KOMPRESO
R
RGPO¯/1
1
RGPO¯/1
2
RGPO¯/1
3
--
---
---
-
C60
L
C10
A3
P
REZERW
A
RGPO¯/1
4
RGPO¯/1
5
RGPO¯/1
6
--
---
---
-
6
5x
4NHX
H
TPS
A
REZERW
A
RGPO¯/1
7
RGPO¯/1
8
RGPO¯/1
9
79,
34x1x95/1x5
0 NHX
H
NS400
N
Ir=1In=400
A
STR23S
E
3
P
C60
L
C20
A3
P
NG125
N
C32
A3
P
RW
N
BRAMY
P.PO¯AROWE
ROZDZIELNIC
A
TYP
B
180kva
r
1500/5
A
NR
OBWODU:
KABEL
mm2:
MOC
kW:
KABEL
TYP:
ODBIORNI
K:
NS160
N
160
A
STR22S
E
3
P
NS160
N
160
A
STR22S
E
3
P
NS160
N
160
A
STR22S
E
3
P
C60
L
C10
A3
P
C60
L
C10
A3
P
C60
L
C20
A3
P
C60
L
C16
A3
P
C60
L
C10
A3
P
NT16H
1
1600
A
1250
A
3
P
NS400
N
400
A
STR23S
E
3
P
315
A
ISFT40
0
3
P
FUPAC
T
2
3x2,
5NHX
H
6
5x1
6NHX
H
NG125
N
C10
A3
P
YKY
4x(3x1x240)
YKY
4x(3x1x240)
--
---
---
-
REZERW
A
--
---
---
-
REZERW
A
--
---
---
-
REZERW
A R
M
BATERI
A
R
T
RGPO¯/
8

24
3.2 Criteria for selection of fire safety circuits to be used in further investigations
Main criteria for selection of fire safety circuits to be used in investigations are:
circuit significance – fire protection devices which are important for fire safety of the building
relative high rated power,
probable fire occurrence along the cable route supplying the device – as the aim is to analyze the
effect of the increase of conductor resistance under fire conditions on the performance of fire
safety device,
longest circuit lengths across zones where fire can occur.
3.3 Description of the selected circuits
On the basis of the aforementioned criteria, the following fire safety circuits have been selected:
sprinkler pump;
smoke fan;
compressor of hydrophore reservoir
First modeled fire protection device is a 200 kW sprinkler pump (see Figure 5). It is a part of a fire
sprinkler system's water supply and is driven by electric motor (also diesel engine for stand-by unit). In
the investigated building the pump intake is connected to the public underground water supply piping, or
a static water source which is a tank. The pump provides water flow at a higher pressure to the sprinkler
system risers and hose standpipes. The total cable length supplying the sprinkler pump is above 280 m.
Four smoke fans of 45 kW rated power (see Figure 6) are a part of a smoke exhaust ventilation and air
admission systems installed in the building. The burning building creates a lot of heat, smoke and toxic
fire gases. They all reduce visibility, make exhaust and fire fighting more difficult, and increase the risk of
combustion and fire stress to the building structures. In order to secure personal safety, effective fire
fighting and minimal damages to structures and property, the developing smoke must be extracted dur-
ing the early stages of the fire. The key tasks for the smoke exhaust system are ensuring that people find
their evacuation routes and facilitating the evacuation itself by keeping the routes free from smoke. It will
also support fire fighting and rescue operations by forming a smokeless zone of make-up air with good
visibility, lower temperature and toxin-free air. For the analysis purposes the 45 kW smoke fan located on
the roof is selected. The total cable length supplying the smoke fan is above 250 m.

25
Figure 5. Sprinkler pump of rated power 200 kW.
Figure 6. One of the main smoke fans of rated power 45 kW.

26
Third investigated fire protection device is a compressor of a hydrophore reservoir. It is a part of indoor
hydrant system installed in the Shopping Centre Galeria Tarnovia. The fire hydrant system is an active
fire protection measure, and a source of water provided with municipal water service to enable firefight-
ers to tap into the municipal water supply to assist in extinguishing a fire. The total cable length supply-
ing the compressor is about 90 m.
One can distinguish both smoke flap actuators and outdoor lighting fittings for fire service roads as the
important fire protection devices. These devices have not been selected for further investigations be-
cause:
The smoke flaps are supplied from the flaps control switchboards where the batteries have been
installed for the emergency power supply of the smoke flaps. Therefore a fire occurring on the
way of a flap power supply path does not affect the reliability of smoke flap actuators operation
(the actuator will be supplied form a battery in case of too low voltage in the primary supply
line);
The cables supplying the outdoor lighting fittings are laid outside the building where occurrence
of the fire is unlikely.
The diagram of selected fire protection circuits is presented in Figure 7. The 15 kV distribution network
which supplies the Shopping Centre Galeria Tarnovia can be modeled as a voltage source with equivalent
impedance calculated from the short-circuit apparent power value given by the electric distribution utility
PGE Dystrybucja S.A.
Fire safety switchboard is supplied by one of the MV feeders at a time, never in parallel. The feeder con-
sists of MV cable XUHAKXS 3x(1x70) 12/20 kV of 5 m length, MV/LV transformer, LV cable YKY
4x(4x1x240) of 6 m supplying LV main switchboard of the building and LV cable YKY 4x(3x1x240) of 5 m
length from the main switchboard terminal (from before main circuit-breaker) to the LV fire protection
switchboard (RGPOŻ). The cables and other components supplying the fire protection switchboard are
located in the MV distribution room, transformer chamber and main distribution room. These rooms are
separated by fire resistant walls REI 120. All cables for supply of fire safety circuits should ensure conti-
nuity of supply during fire for the required time of operation of particular piece of equipment but no
shorter than 90 minutes.
The modeled MV/LV transformer is protected with the use of digital unit VIP 35 Schneider Electric coope-
rating with the MV circuit-breaker located in the transformer bay of MV switchboard. The most important
rated data of the MV/LV transformer are presented in Table A1.1 (see Appendix 1).

27
Figure 7. Diagram of the selected fire safety circuits.
External
power
system
MV switchgear 15 kV
Terminal of LV main
switchgear 0,4 kV
(before main circuit
breaker)
630 A1600 A
M
~
Sprinkler
pump
(PT1)
l = 9 m
NHXH 4x(2x1x120)/1x120
main distribution
room
l = 244 m
Fire protection switchgear 0,4 kV (RGPOŻ)
Transformer
chamber
l = 24 m
sprinkler room;
level +0
NHXH
3x(2x1x120)/
1x120
l = 7 m
700 A
330 A
200 kW
YKY 4x(3x1x240)
l = 5 m
YKY 4x(4x1x240)
REI120
REI120
Transformer
15,75/0,4 kV
1000 kVA
REI120
l = 6 m
XUHAKXS 3x(1x70)
12/20 kV
l = 5 m
160 A
M
~
Smoke
fan
(ODG2)
garage;
level -1
l = 9 m
NHXH 4x1x95/1x50
smoke ventilation
control room; level
-1
160 A
45 kW
REI120
NHXH
3x1x95/1x50
cable well;
level -1,+0,+1
REI120REI120
l = 10 m
REI120
l = 11 m
REI120
roofl = 14 m
C32 A
l = 9 m
NHXH 5x6
hydrophore room;
level -1
REI120
REI120
l = 78 m
M
~
Compressor
(K)
l = 4 m
2,2 kW
l = 240 m
NHXH 4x6
l = 8 m
l = 3 m
80 A
4 A
equivalent
loadequivalent
load
sprinkler switchbord
0,4 kV (RPT1)
smoke ventilation
switchbord 0,4 kV
(RODG2)
hydrophore
switchbord 0,4 kV
(RH)
equivalent
load
MV distribution
room

28
All of the electric appliances and switchboards supplied from RGPOŻ except the investigated devices are
modeled as an aggregated equivalent load (see Figure 7). The equivalent load located in RGPOŻ has
been calculated as the difference between estimated peak active power value of the RGPOŻ switchboard
and the sum of rated power values of modeled fire protection appliances, i.e.: PT1, ODG2 and K. The
equivalent load power is calculated as follows:
p_Kp_PT1p_PT1p_RGPOŻEL_RGPOŻ PPPPP (1)
kW6,2662,748,5208,8-6,526EL_RGPOŻP
Circuit supplying the sprinkler pump
The sprinkler pump is used only for the sprinkler system supply. The pump is not a source of water for
fire external and internal hydrants. The sprinkler pump starts pumping when the building service con-
firms a fire. Stopping of the sprinkler pump must be manually after extinction of the fire is confirmed.
The most important rated data of the modeled sprinkler pump and the electric drive motor are presented
in Tables A1.2 and A1.3 (see Appendix 1).
The circuit supplying the sprinkler switchboard RPT1 (see Figure 7) is protected with the use of overcur-
rent protection installed in the circuit-breaker NS630N Merlin Gerin (RGPOŻ). The most important data of
the circuit breaker are presented in Table A1.4 (see Appendix 1).
The circuit supplying the sprinkler switchboard RPT1 is made of XLPE cables 4x(2x1x120)/1x120 of the
total length of 277 m which are laid in three fire zones. The cable section of 9 m length is located in the
main distribution room, the cable section of 244 m length is located in the car park and the cable section
of 24 m length is located in the sprinkler room. Cables NHXH 4x(2x1x120)/1x120 are laid in fire resis-
tance E90 cable trays. The structure of cable bundle is shown in
Figure 8. NHXH - halogen-free, flame-resistant cable. The cable bundle 4x(2x1x120)/1x120 denotation
means:
4x(…). – phase 1, phase 2, phase 3 and neutral points used in LV network,
(2x1x120) – two single-core cables (120 mm2
cross-section) as a protective conductor.
The most important rated data of the modeled cable supplying the sprinkler pump are presented in Table
A1.5 (see Appendix 1).
Figure 8. A cable bundle supplying the RPT1 switchboard.
ProtectivePhase 3
Phase 2
Phase 1
Neutral

29
The electric motor driving the sprinkler pump is supplied from the sprinkler switchboard RPT1 using the
cables NHXH 3x(2x1x120)/1x120 of the length of 7 m (neutral point of LV network is not used – NHXH
3x…). The fire resistance E90 for this cable section is applied as well. The induction machine which driving
the sprinkler pump is supplied from a softstart unit PST250 ABB. The most important rated data have
been presented in Table A1.6 (see Appendix 1).
The softstart unit is protected by the fuses Square Body Bussmann 700 A. The softstart unit is also pro-
tected by overcurrent protection with the use of a thermal relay.
Circuit supplying the smoke fan
The smoke fan ODG2 is used both for air admission and smoke exhausting. It is an integral part of smoke
exhausting system of the building. This system ensures appropriate evacuation conditions for the building
users (consumers and service). The fan ODG2 operates in two smoke exhausting zones WD-2 i WD-3. It
operates in a reverse system (changing of motor rotation direction). The smoke fan ODG2 is started ma-
nually after service crew confirmation. Stopping of the ODG2 fan is possible only in a manual way after
fire extinguish confirmation. The most important rated data of the modeled smoke fan and the electric
drive motor are presented in Table A1.7 and A1.8 (see Appendix 1).
The circuit supplying the smoke ventilation switchboard RODG2 is protected with the use of overcurrent
protection relay installed in the circuit-breaker NS160N Merlin Gerin (RGPOŻ). The most important rated
data have of the circuit breaker are presented in Table A1.9 (see Appendix 1).
The circuit supplying the smoke ventilation switchboard RODG2 (see Figure 7) is made using XLPE
cables NHXH 4x(1x95)/1x50 of the total length of 257 m which are laid in three fire zones. The cable
section of 9 m length is located in the main distribution room, the cable section of 240 m length is lo-
cated in the car park and the cable section of 8 m length is located in the smoke ventilation control room.
Cables NHXH 4x(1x95)/1x50 are laid in the fire resistance E90 cable trays. The structure of cable bundle
is shown in Figure 9. The cable bundle 4x(1x95)/1x50 denotation means:
4x(…). – phase 1, phase 2, phase 3 and neutral points used in LV network,
(1x95) – one single-core cable (95 mm2
cross-section) as a protective conductor
Figure 9. A cable bundle supplying the RODG2.
ProtectivePhase 3
Phase 2
Phase 1
Neutral

30
The most important rated data of the modeled cable supplying the smoke fan is presented in Table
The electric motor driving the smoke fan is supplied from the smoke ventilation switchboard RODG2 us-
ing the cables NHXH 3x(1x95)/1x50 of 35 m length (neutral point of LV network is not used – NHXH 3x…)
which are laid in three fire zones The cable section of 10 m length is located in the smoke ventilation
control room, the cable section of 11 m length is located in the cable well and the cable section of 14 m
length is located on the roof. The fire resistance E90 for this cable section is applied as well. The asyn-
chronous motor which as a drive of the smoke fan is supplied from a softstart unit PST50 ABB. Its the
most important rated data have been presented below in Table A1.11 (see Appendix 1).
The softstart unit is protected by the fuses Square Body Bussmann 160 A. The softstart unit is also pro-
tected by overcurrent protection with the use of a thermal relay.
Circuit supplying the compressor of hydrophore reservoir
The compressor is an element of the hydrophore reservoir which is a source of water for hydrant system
among other things. The compressor operates automatically and its operation exactly depends on the
hydrant system load during a fire as well as determined water pressure in the hydrophore. The most im-
portant rated data of the modeled compressor and the electric drive motor are presented in Table A1.12
and A1.13 (see Appendix 1).
The circuit supplying the separated section of hydrophore switchboard RH (see Figure 7) is protected
with the use of overcurrent releases installed in the circuit-breaker NG125N C32A Merlin Gerin (RGPOŻ).
Its the most important rated data have been presented in Table A1.14 (see Appendix 1).
The circuit supplying the separated section of hydrophore switchboard RH (see Figure 3) is made as a
cable NHXH (5x6) (fifth-core cable of 6 mm2 cross-section; five wires are used for three phases, neutral
and protective point of LV network) of 90 m total length which is laid in three fire zones. The cable sec-
tion of 9 m length is located in the main distribution room, the cable section of 78 m length is located in
the car park and the cable section of 3 m length is located in the hydrophore room. Cable NHXH (5x6) is
laid in the fire resistance E90 cable trays.
The most important rated data of the modeled cable supplying the compressor is presented in Table
The electric motor driving the compressor is supplied from the separated section of hydrophore switch-
board RH using the cables NHXH (4x6) (without the neutral point of LV network) of 4 m length. The fire
resistance E90 for this cable section is applied as well. The electric motor as a drive of the compressor is
supplied directly (direct motor start-up). The overload protection with the use of thermal relay installed in
RH is used.

31
4. Verification of cross section of cables based on IEC 60364
The sizing of conductors supplying electrical equipment that must remain functional during a fire has
been made by the Galeria Tarnovia electrical installation designers using common cable selection criteria
described in IEC 60364 for 90 C maximum permissible operating and 25 C ambient temperatures.
The increased resistance must be considered in calculating the appropriate size of conductor to maintain
voltage within required limits to ensure that the protection devices can operate effectively [4].
Below a process for cable sizing is presented both for normal and fire conditions using IEC recommended
criteria and their modifications given by [4]. Main goals are to show the problems of fire safety devices
which may occur during a fire as well as indicate on the necessity of safety cable selection modification if
necessary.
4.1 Criterion 1: Current rating & overload protection
IEC 60364-5-523 standard suggests the base current ratings of different types of cables in tables. Each
of these tables pertain to a specific type of cable construction (e.g. copper conductor, PVC insulated,
0.6/1kV nominal voltage, etc) and a set of installation conditions (e.g. ambient temperature, installation
method, etc). It is important to note that the current ratings are only valid for the specific types of cables
and base installation conditions [2].
When the proposed installation conditions differ from the base conditions, derating (or correction) factors
can be applied to the base current ratings to obtain the actual installed current ratings [2].
IEC 60364-5-523 standard provides derating factors for a range of installation conditions, for example
ambient / soil temperature, grouping or bunching of cables, soil thermal resistivity, etc. The installed
current rating is calculated by multiplying the base current rating with each of the derating factors, i.e.:
'zdz IkI (2)
where Iz the installed current rating, Iz the base current rating, kd the product of all the derating
factors
The protective device must therefore be selected to exceed the full load current, but not exceed the ca-
ble's installed current rating, i.e. this inequality must be met:
z
zrFset_OLPB
,
)(
II
IIII
4512
(3)
where IB full load current, Iset_OLP – current set on the overload protection, IrF – protective device rating if
current set on the overload protection is unavailable, Iz – installed cable current rating, I2 – overcurrent
protection trip current.

32
4.2 Criterion 2: Permissible voltage drop
In normal electrical applications, the resistance of a copper conductor can be calculated by the following
formula, which is valid up to about 200°C (a conductor resistance linearly depends upon the conductor
temperature) [4]:
)( 2020 1RR (4)
where R – the resistance of a copper conductor at high temperature, R20 – the resistance at 20 C (293,15
K), – temperature growth.
At temperatures higher than +200°C, the relation describing the conductor’s resistance becomes non-
linear and is given by the Wiedemann-Franz formula [4]:
161
20
20
,
RR (5)
where R – the resistance of a copper conductor at high temperature, R20 – the resistance at 20 C (293,15
K), is the considered temperature in K.
Because the considered building is compartmentalized into fire zones to reduce fire spread, cables feed-
ing protection equipment are rarely exposed to fire temperatures over their entire length. The part of the
cable not affected by the fire will operate at the normal temperature appropriate to the loading, while
that exposed to fire has increased resistance. It is necessary to assess which areas may be simultaneous-
ly affected by fire in the worst case and assess the proportion of cable length that may be affected. The
total conductor resistance is then calculated by assuming normal resistance for the length unaffected by
fire and applying a multiplication factor to the length that is affected [4].
100
100 N
N
)( R
R
R
y
RT
(6)
where: RT the resistance of the conductors, RN the resistance of the conductor under normal condi-
tions, y is the percentage of the cable length estimated to be affected by fire.
Because of the importance of voltage stability to the proper working of any electrical or electronic device,
the voltage drop between the incoming supply at the point of common coupling and the terminals of the
end-use equipment should be limited to 5% under normal conditions and 10% under emergency condi-
tions [4].
The voltage drop can be calculated as follows:
permB
N
%)sincos(
1003
% UXRI
U
U (7)

33
where: U% perm – permissible voltage drop for given conditions, UN – nominal voltage in V, IB full load
current in A, R – is conductor resistance (depending on the considered temperature) in , X – reactance
in .
4.3 Criterion 3: Short-circuit strength
The minimum cable size due to short circuit temperature rise is typically calculated with an equation of
the form:
tiks 22
)( (8)
where s – the cross-sectional area of the cable in mm2
, ti2 clearing Joule’s integral of a short-circuit
protective device in A2
s, k – a short circuit temperature rise constant.
The temperature rise constant is calculated based on the material properties of the conductor and the
initial and final conductor temperatures [3]. IEC 60364-5-54 standard treats the temperature rise con-
stant for copper conductors in the following way:
i
if
,
ln
T
TT
k
5234
1226 (9)
where Ti – the initial conductor temperature in C, Tf – the final conductor temperature in C.
4.4 Criterion 4: Automatic disconnection of supply
In case of earth fault the upstream protective device (i.e. fuse or circuit breaker) must trip within the
maximum disconnection time. In order for the protective device to trip, the fault current due to a bolted
short circuit must exceed the value that will cause the protective device to act within the maximum dis-
connection time (5 s for distribution circuits and 0,4 s for final circuit in TN earthling systems 400/230 V)
[5]
a
s
I
Z
U0 (10)
where U0 – nominal phase to ground voltage in V, Zs – the impedance of the earth fault loop in Ω (calcu-
lated for normal and emergency condition), Ia – the earth fault current required to trip the protective de-
vice within the minimum disconnection time in A.

34
4.5 Cable size verification for analyzed fire protection circuits
The cable size verification for all considered sections have been presented in Tables 1 6. All the calcula-
tions have been carried out based on formulas (4,8,9,11).
Table 3. Cable size verification data for the line section between RPT1 and PT1 (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH 3x(2x1x120)/1x120
Cable length: 7 m
Cable installation: horizontal no perforated tray (flat way)
Distance between cables: 0 mm (flat, zero spacing)
Number of bundles: 2
Ambient temperature: 25 C
Maximum conductor operating temperature: 90 C
Maximum short-circuit conductor temperature: 250 C
Required time of disconnection: 0,4 s
CRITERION 1
IB 327 A
Iset_OLP(IrF) 330 A
Iz (IEC 60364-5-523) 800 A
I2 396 A
Derating factors 0,91 1,06
Criterion (equation 4) fulfilled? YES
CRITERION 2 – normal conditions (no fire at the entire cable route)
U% 4,6%
U% perm 5%
CRITERION 2 – fire conditions (90 min ambient temperature for garage cable section is 986 C)
U% 14,9%
U% perm 10%
Criterion (equation 8) fulfilled? NO
CRITERION 3
k 135
s 120 mm2
ti2
465000 A2
s
U0 230 V
Zs 0,074
Ia 2646 A
U0 230 V
Zs 0,220
Ia 2646 A

35
Table 4. Cable size verification data for the line section between RGPOŻ and the RPT1 (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH 4x(2x1x120)/1x120
Cable length: 277 m
Required time of disconnection: 5 s
CRITERION 1
IB 327 A
Iset_OLP(IrF) 330 A
Iz (IEC 60364-5-523) 800 A
I2 396 A
U% 4,6%
U% perm 5%
U% 14,9%
U% perm 10%
CRITERION 3
k 135
s 120 mm2
ti2
4000000 A2
s
U0 230 V
Zs 0,073
Ia 2646 A
U0 230 V
Zs 0,219
Ia 2646 A

36
Table 5. Cable size verification data for the line section between RODG2 and ODG2 (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH 3x(1x95)/1x50
Cable length: 35 m
Cable installation:
horizontal and vertical no perforated tray (flat
way)
CRITERION 1
IB 80 A
Iset_OLP(IrF) 80 A
Iz (IEC 60364-5-523) 342 A
I2 96 A
U% 2,4%
U% perm 5%
U% 8,3%
U% perm 10%
CRITERION 3
k 135
s 95 mm2
ti2
7500 A2
s
U0 230 V
Zs 0,182
Ia 1248 A
U0 230 V
Zs 0,796
Ia 1248 A

37
Table 6. Cable size verification data for the line section between RGPOŻ and the RODG2 (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH 4x(1x95)/1x50
Cable length: 257 m
CRITERION 1
IB 80 A
Iset_OLP(IrF) 80 A
Iz (IEC 60364-5-523) 342 A
I2 96 A
U% 2,4%
U% perm 5%
U% 8,3%
U% perm 10%
CRITERION 3
k 135
s 95 mm2
ti2
600000 A2
s
U0 230 V
Zs 0,161
Ia 1248 A
U0 230 V
Zs 0,771
Ia 1248 A

38
Table 7. Cable size verification data for the line section between RH and K (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH (4x6)
Cable length: 4 m
CRITERION 1
IB 3,8 A
Iset_OLP(IrF) 32,0 A
Iz (IEC 60364-5-523) 54 A
I2 38,4 A
U% 0,48%
U% perm 5%
U% 1,8%
U% perm 10%
CRITERION 3
k 135
s 6 mm2
ti2
50000 A2
s
U0 230 V
Zs 0,607
Ia 320 A
U0 230 V
Zs 2,757
Ia 320 A

39
Table 8. Cable size verification data for the line section between RGPOŻ and RH (see Figure 7).
CABLE SECTION DATA
Cable type: NHXH (5x6)
Cable length: 90 m
CRITERION 1
IB 3,8 A
Iset_OLP(IrF) 32,0 A
Iz (IEC 60364-5-523) 54 A
I2 38,4 A
U% 0,48%
U% perm 5%
U% 1,8%
U% perm 10%
CRITERION 3
k 135
s 6 mm2
ti2
50000 A2
s
U0 230 V
Zs 0,551
Ia 320 A
U0 230 V
Zs 2,690
Ia 320 A

40
Verification summary
The aforementioned preliminary calculations show:
correct cable size selection on the electrical installation design stage based on IEC 60364-4-41
problems with the time of automatic disconnection of supply only under fire conditions
problems with the voltage drop under fire conditions as well
For the considered circuits operating under normal conditions (IEC 60364-4-41) the criterion of automatic
disconnection of supply is fulfilled. One can state that the cable size verification made on the electrical
installation design stage is correct. Besides for the circuits supplying the smoke fun and compressor un-
der fire conditions the disconnection time considerable exceeds in comparison to the required time. This
problem should be particularly verified by detailed dynamic simulations.
For the circuit supplying the sprinkler pump, the voltage drop is above the required percentage value
(10%). This problem should be verified by detailed dynamic simulations as well.
5. Dynamic simulations
Next aim of this project is to verify the performance of fire protection devices and protective relays in
dynamic simulations taking into account increased cable resistances during fire. Modeling different dy-
namic phenomena such as:
current and flux transients in a motor stator and rotor,
speed-torque interaction between a motor and a driven machine
thermal and electromagnetic response (modeled by Time-Current characteristic) to a fault cur-
rent,
which may occur during a fire in the building or its part may confirm the identified problems with the vol-
tage drop and automatic disconnection of supply. The planned dynamic simulations are likely to enable to
discover other problems, such as steady-state overvoltages or voltage dips, which can not be identified
using simplified calculations. Therefore electromechanical transients (RMS) studies should be performed
in order to show such actions as electric motor starting, fault clearing by protection devices, etc. All of
these studies must consider both normal and fire conditions in the investigated building.
5.1 Models of selected fire safety circuits
All the component models are based on the DIgSILENT PowerFactory references recommending appropri-
ate modeling for dynamic RMS simulations [6, 7]. The investigated fire safety circuits modeled in DIgSI-
LENT PowerFactory software are presented in Figure 10. Detailed description of all the models of the
considered fire system and all the components are presented in Appendix 2.

41
Figure 10. PowerFactory screenshot presenting a diagram of modeled fire safety cicuits.

42
5.2 Analyzed scenarios
Simulation scenarios take into account both normal conditions (no fire) and the most unfavorable fire
conditions under which the equipment is still required to operate (90 minutes after fire breaks out). It is
assumed that ambient temperature will rise due to the fire only in the car park, because the cables sup-
plying the fire safety switchboards (RPT1, RODG2 and RH see Figure 7) are the longest in the car park
area, i.e. 244, 240 and 70 m respectively.
Table 9. Scenarios considered in the simulation analysis.
Scenario
code
Scenario description
Quasi steady-state
QSS During steady state operation of fire safety devices a fire breaks out and develops in the
car park (total simulation time 90 min)
Motor starting
MS1/0 Starting of a sprinkler pump; normal conditions (no fire) (the rest of fire safety devices
operate in the steady-state)
MS1/90 Starting of a sprinkler pump 90 min after fire breaks out (the rest of fire safety devices
operate in the steady-state)*
MS2/0 Starting of a smoke fan; normal conditions (no fire) (the rest of fire safety devices operate
in the steady-state)
MS2/90 Starting of a smoke fan 90 min after fire breaks out (the rest of fire safety devices operate
in the steady-state)*
MS3/0 Starting of a compressor; normal conditions (no fire) (the rest of fire safety devices oper-
ate in the steady-state)
MS3/60 Starting of a compressor 90 min after fire breaks out (the rest of fire safety devices oper-
ate in the steady-state)*
Single-phase fault clearing
SC1.1/0 Phase-to-ground short-circuit at the sprinkler switchboard busbar under no fire conditions
SC1.2/0 Phase-to-ground short-circuit at the sprinkler pump motor terminal under no fire condi-
tions
SC2.1/0 Phase-to-ground short-circuit at the smoke ventilation switchboard busbar under no fire
conditions
SC2.2/0 Phase-to-ground short-circuit at the smoke fan motor terminal under no fire conditions
SC3.1/0 Phase-to-ground short-circuit at the hydrophore switchboard busbar under no fire condi-
tions
SC3.2/0 Phase-to-ground short-circuit at the compressor motor terminal under no fire conditions
SC1.1/90 Phase-to-ground short-circuit at the sprinkler switchboard busbar at fire time of 90 min*
SC1.2/90 Phase-to-ground short-circuit at the sprinkler pump motor terminal at fire time of 90 min*
SC2.1/90 Phase-to-ground short-circuit at the smoke ventilation switchboard busbar at fire time of
90 min*
SC2.2/90 Phase-to-ground short-circuit at the smoke fan motor terminal at fire time of 90 min*
SC3.1/90 Phase-to-ground short-circuit at the hydrophore switchboard busbar at fire time of 90
min*
SC3.2/90 Phase-to-ground short-circuit at the compressor motor terminal at fire time of 90 min*
* An optional scenario is allowed if the expected problems will be confirmed (too large voltage drop, too
long time of supply disconnection). In such case an optional scenario will consider an increased cable
cross-section.

43
5.3 Simulations
For all the considered simulation scenarios, the plots including motors voltage, current, electric active and
reactive power as well as speed and mechanical torque are presented in figures found in Appendix 3.
The scenarios including existing cables cross-section as well as scenarios including cables cross-sections
sufficient to ensure proper operation of equipment have been considered.
Below only the most interesting cases are described and plots shown.
5.3.1 Quasi-steady-state
First considered scenario is a quasi-steady-state of the fire safety system (QSS). The fire in the car park
breaks out when all the safety equipment is in operation and develops during time of 90 min (ambient
temperature plot for standard temperature – time curve [4] is shown in Figure 11). The word "quasi"
means that the investigated operation state of a fire safety circuit is not strictly fixed but it changes slow-
ly over time as a result of a fire in the car park. Such scenario can happen in reality when fire breaks out
in other part of the building and reaches the fire safety circuits after fire safety devices had been acti-
vated.
Figure 11. Ambient temperature in the car park for standard temperature–time curve.
Figure 12 and Figure 13 show the plots of RMS values of voltage, current, active power and speed for
all the modeled motors (positive-sequence values). The simulation starts from time of -100 s and the fire
start develop from time of 0 s. Simulation lasts for 90 minutes. Looking at the left side of Figure 12 one
can see that the voltages at the terminals of the motors decrease while the fire develops due to the in-
creased resistance of cable sections laid in the car park. The calculated resistance values for the cable
sections located in the car park are shown in Table 10.
5.4E+34.3E+33.2E+32.1E+31.0E+3-1.0E+2 [s]
1250,00
1000,00
750,00
500,00
250,00
0,00
CommonModel: Ambient temperature[C deg]
DIgSILENT
Tarnovia ambient_temperature Date: 3/19/2012
Annex: /4

44
Figure 12. RMS Voltage and current plots for QSS scenario.
Figure 13. RMS motor speed and active power plots for QSS scenario.
5.4E+34.3E+33.2E+32.1E+31.0E+3-1.0E+2 [s]
1,05
1,00
0,95
0,90
0,85
0,80
AM_sprinkler_pump: Positive-Sequence-Voltage, Magnitude in p.u.
AM_smoke fan: Positive-Sequence-Voltage, Magnitude in p.u.
AM_compressor: Positive-Sequence-Voltage, Magnitude in p.u.
5.4E+34.3E+33.2E+32.1E+31.0E+3-1.0E+2 [s]
1,20
1,10
1,00
0,90
0,80
AM_sprinkler_pump: Positive-Sequence Current, Magnitude in p.u.
AM_smoke fan: Positive-Sequence Current, Magnitude in p.u.
AM_compressor: Positive-Sequence Current, Magnitude in p.u.
DIgSILENT
Tarnovia voltage and current Date: 4/2/2012
Annex: /4
5.4E+34.3E+33.2E+32.1E+31.0E+3-1.0E+2 [s]
1,0000
0,9875
0,9750
0,9625
0,9500
0,9375
AM_sprinkler_pump: Speed
AM_smoke fan: Speed
AM_compressor: Speed
5.4E+34.3E+33.2E+32.1E+31.0E+3-1.0E+2 [s]
1,02
1,00
0,98
0,96
0,94
0,92
0,90
AM_sprinkler_pump: Total Active Power in p.u. (base: 0,20 MW)
AM_smoke fan: Total Active Power in p.u. (base: 0,04 MW)
AM_compressor: Total Active Power in p.u. (base: 0,00 MW)
DIgSILENT
Tarnovia power&troque Date: 4/14/2012
Annex: /5

45
The largest voltage drop that is reached during the fire is observed on the terminals of the sprinkler
pump motor. The voltage approaches the value of 0,82 p.u.(left side of Figure 12). The voltage drop is
proportional to the increase of cable resistance during the fire.
Due to low value of supply voltage an unstable motor operation may occur. When the operating point of
the motor is below the knee of Speed-Torque characteristic (see Figure 17) the motor speed and torque
are much lower than their nominal values. Thereby the motor active power is reduced. In spite of de-
crease of supplying voltage due to increasing cable resistance the unstable motor operation is not ob-
served in the QSS scenario.
The largest decrease of active power does not exceed 9% and is observed in the circuit supplying the
smoke fan (right side of Figure 1). In case of the sprinkler pump motor an increase of current value is
observed (right side of Figure 12). The supplying voltage drops much faster than active power decrease
(the voltage drop is 14,9% and active power drop is 7,5%). Because the motor active power is propor-
tional to the voltage and current, the latter increases. Current in the sprinkler pump motor circuit is not
large enough to cause the thermal protection tripping. For the QSS scenario all the investigated machines
operate properly (ensure its fire safety functions) during 90 min of fire.
Assuming that in other buildings all the fire safety circuits have been designed and build according to
recommendations found in IEC 60364 standards, the general conclusion is that if the fire safety devices
operate when the fire reaches their supplying circuits they will stay in operation for the entire period of
90 minutes.
Table 10. Resistance of cable sections located in the car park.
Simulation time Resistance of cable sections located in the car park [Ohms] and [p.u]
Sprinkler pump circuit Smoke fan circuit Compressor circuit
Length: 244 m
Cross-section: 4x(2x1x120)/
1x120 mm2
Length: 240 m
Cross-section: 4x(1x95)/
1x50 mm2
Length: 78 m
Cross-section: (5x6) mm2
0 s 0,0190 (1,0 p.u.) 0,0236 (1,0 p.u.) 0,2449 (1,0 p.u.)
30 min 0,0860 (4,52 p.u.) 0,1067 (4,52 p.u.) 1,1075 (4,52 p.u.)
60 min 0,0956 (5,03 p.u.) 0,1186 (5,03 p.u.) 1,2307 (5,03 p.u.)
90 min 0,1013 (5,33 p.u.) 0,1257 (5,33 p.u.) 1,3042 (5,33 p.u.)
Table 10 shows the rated resistance of a cable conductor specified at 25°C is increased by a factor of
about 5,33 under PH90. The voltage drop roughly increases according to this factor value.
Next, motor starting situations are analyzed. For starting of the sprinkler pump (PT1) and smoke fan
(ODG2) motors soft-starters are used. The electric motor driving the compressor (K) starts in direct way.
Each motor starting is analyzed individually. The rest of motors operate in steady-state. The investigation
of starting of motors is done for two extreme situations, at no fire condition (MS(1,2,3)/0) and second
situation is for the conditions prevailing 90 min after the fire breaks out (MS(1,2,3)/90). All the fuses and
overcurrent protection are modeled and are activated in these simulations.
5.3.2 Motor starting at no-fire conditions
Figure 14 presents plots of RMS voltage and current during all the investigated motors starting at no fire
conditions. Looking at this figure no problems are observed. The investigated motors start properly. The
PT1 and ODG2 motors are supplied from softstarters. The voltage on terminals of these machines in-
creases linearly (left side of Figure 14) according to the softstarter settings (initial voltage is 0,3 p.u.

46
and the ramp time is 10 s). In case of PT1 motor starting, maximum current value approaches about 3,9
p.u. and is not large enough to cause the thermal protection tripping. The compressor (K) motor starts
properly as well.
Figure 14. Voltage and current plots for MS1/0, MS2/0 and MS3/0 scenarios.
5.3.3 Motor starting at 90 min. fire conditions
When PT1 motor starts at 90 min fire conditions (see Figure 16), the voltage value at RPT1 terminals
has the value of about 0,55 p.u., which is too low for the softstarter to properly control the output vol-
tage. The firing angles of soft-starter thyristors are controlled based on the input voltage, which is used
as a reference. Under such conditions, the operating point of the motor is below the knee of Speed-
Torque characteristic (see Figure 17). For this reason the motor current is high (about 3,3 p.u.) and
causes activation of thermal protection (located in RPT1 switchboard) and disconnection of the PT1 motor
circuit approx. 26 seconds after it was started. The delay is caused by the fact that the motor current
flows through relay bimetals which are indirectly heated. Under the effect of the heating, the bimetals
bend and cause a relay to trip. The effect of bimetal heating is determined by Time-Current characteristic
which is shown in Figure 15.
12,008,6005,2001,800-1,600-5,000 [s]
1,25
1,00
0,75
0,50
0,25
0,00
-0,25
AM_sprinkler_pump: Positive-Sequence-Voltage, Magnitude in p.u.
AM_smoke fan: Positive-Sequence-Voltage, Magnitude in p.u.
AM_compressor: Positive-Sequence-Voltage, Magnitude in p.u.
12,008,6005,2001,800-1,600-5,000 [s]
5,00
4,00
3,00
2,00
1,00
0,00
-1,00
AM_sprinkler_pump: Positive-Sequence Current, Magnitude in p.u.
AM_smoke fan: Positive-Sequence Current, Magnitude in p.u.
AM_compressor: Positive-Sequence Current, Magnitude in p.u.
DIgSILENT
Tarnovia SC-current Date: 4/3/2012
Annex: /5

Dynamic modeling of the performance of safety electrical circuits during a fire

Dynamic modeling of the performance of safety electrical circuits during a fire

Recommended

Recommended

More Related Content

Similar to Dynamic modeling of the performance of safety electrical circuits during a fire

Similar to Dynamic modeling of the performance of safety electrical circuits during a fire (20)

More from Leonardo ENERGY

More from Leonardo ENERGY (20)

Recently uploaded

Recently uploaded (20)

Dynamic modeling of the performance of safety electrical circuits during a fire