1.2 Assessment of slender external steel columns
2
1.2 Assessment of slender external steel columns Block F., United Kingdom Buro Happold FEDRA COST TU904 – WG1 – Barcelona – 5 th of July 2010 Assessment of slender external steel columns Dr Florian Block Introduction • Large university building with internal court yard in the US • External intumescent paint expensive and high maintenance • IBC 2003 requires 60 minutes fire resistance to columns supporting roof only and 120 minutes if columns supporting floor 1 2 Structural overview • Column section: HSS18x0.5 • Length: 21m • Slenderness λ = 127 • Estimated cost of fire protection per column: $20,000 (2 hours) • Distance from column to façade: 0.5m – 7.1m 2 hours Fire Resistance 1 hour Fire Resistance Approach • Agree approach with stakeholders • Define the design fires • Represent the fire with a simple model • Perform a heat transfer analysis from the fire to the columns • Calculate the time depended steel temperature of the column • Definition of Fire Limit State loading • Perform a non-linear structural analysis of columns • Present findings to Code Officials and Fire Marshals Royal Ascot Racecourse, Ascot Emirate Football Stadium, London 3 4 Defining the design fires Three types of fires are likely in the university building and the court yard 1. Localised sprinkler controlled fire 2. High ventilation sprinkler controlled compartment fire 3. External fire in court yard 0 1000 2000 3000 4000 5000 6000 7000 8000 0 300 600 900 1200 1500 1800 2100 2400 Time [s] Heat Release Rate [kW] Office Workstation - Experimental data Office - Localised Design Fire Assumptions: • Sprinklers control spread of fire to adjacent objects but do not influence localised fire • Localised fire is a 3-panle office workstation • Temperature of continuous part of flame: 1000°C • Temperature of intermittent part of flame: 1000°C -> 600°C • Column is not engulfed 7MW Steady State Fire was used for design Represent the fire with a simple model • Select HRR and fire base size • Calculate flame height using Hekastead equation for rectangular fires • Assume flame shape • Split flame into individual slices • Calculate emissivity of flame slice • Select a receiver (height, front or side) • Calculate heat flux to receiver from each flame slice • Sum up the total heat flux for selected receiver • Select next receiver • Select next HRR and fire base size Fire base width w Max (w/2, w- hf) Fire Base Width w Fire base width w Flame height h f Base distance to column d Receiver height z Side receiver Side receiver Front receive r 5 6
Transcript of 1.2 Assessment of slender external steel columns
1.2 Assessment of slender external steel columns
Block F., United Kingdom
Buro Happold FEDRA
COST TU904 – WG1 – Barcelona – 5th of July 2010
Assessment of slender external steel columns
Dr Florian Block
Introduction• Large university building with internal court yard in
the US
• External intumescent paint expensive and high maintenance
• IBC 2003 requires 60 minutes fire resistance to columns supporting roof only and 120 minutes if columns supporting floor
1 2
Structural overview• Column section: HSS18x0.5• Length: 21m• Slenderness λ = 127• Estimated cost of fire protection per
column: $20,000 (2 hours)• Distance from column to façade:
0.5m – 7.1m
2 hours Fire Resistance
1 hour Fire Resistance
Approach• Agree approach with stakeholders• Define the design fires• Represent the fire with a simple model• Perform a heat transfer analysis from the fire to the columns• Calculate the time depended steel temperature of the column• Definition of Fire Limit State loading• Perform a non-linear structural analysis of columns
• Present findings to Code Officials and Fire Marshals
Royal Ascot Racecourse, AscotEmirate Football Stadium, London
3 4
Defining the design firesThree types of fires are likely in the university building and the court yard
1. Localised sprinkler controlled fire2. High ventilation sprinkler controlled compartment fire3. External fire in court yard
0
1000
2000
3000
4000
5000
6000
7000
8000
0 300 600 900 1200 1500 1800 2100 2400Time [s]
Hea
t Rel
ease
Rat
e [k
W]
Office Workstation - Experimental data
Office - Localised Design Fire
Assumptions:• Sprinklers control spread of fire to
adjacent objects but do not influence localised fire
• Localised fire is a 3-panle office workstation
• Temperature of continuous part of flame: 1000°C
• Temperature of intermittent part of flame: 1000°C -> 600°C
• Column is not engulfed7MW Steady State Fire was used for design
Represent the fire with a simple model
• Select HRR and fire base size• Calculate flame height using Hekastead
equation for rectangular fires• Assume flame shape• Split flame into individual slices• Calculate emissivity of flame slice• Select a receiver (height, front or side)
• Calculate heat flux to receiver from each flame slice
• Sum up the total heat flux for selected receiver
• Select next receiver
• Select next HRR and fire base size
Fire base width w
Max (w/2, w-hf)
Fire Base Width w
Fire
bas
e w
idth
wF
lam
e he
ight
hf
Base distance to column d
Receiver
height z
Side receiver
Side receiver
Fro
nt
rece
ive
r
5 6
Perform a heat transfer analysis from the fire to column
• The heat flux is calculated for each side of the column separately• View factors are calculated for each slice of the flame individually• Circular columns are modelled as square• Only radiation is considered
0
0.5
1
1.5
2
2.5
3
3.5
0 2 4 6 8 10 12 14 16Radiative Heat Flux [kW/m 2]
Hei
ght [
m]
Calculated heat flux to front of Column - 2m distanceDesign heat flux to front of Column - 2m distanceCalculated heat flux to side of Column - 2m distanceDesign heat flux to side of Column - 2m distance
• Different distances between the base of the fire and the columns are considered depending on the situation
• Maximum heat flux along the height of the column is used as design value
• Note: EC1 considers an uniform heat flux to an external column based on the weighted average of the exposed sides
Fire size: 7MWFire base: 2m x 2m
Calculate the time depended steel temperature of the column• The steel temperature is
calculated using the heat transfer software TASEF
• Heat loss of the column to the surrounding air is considered
0
20
40
60
80
100
120
0 300 600 900 1200 1500 1800 2100 2400Time [s]
Ste
el T
empe
ratu
re [°
C]
0.00
2.00
4.00
6.00
8.00
10.00
12.00
Hea
t Flu
x [k
W/m
2 ]
Temperature of exposed side
Temperature of un-exposed side
Heat Flux from D2F1.1 - Exp
67 75 88 94 97 97 94 88 75 67
59 59
46 46
39 39
35 35
33 33
32 32
31 31
29 29
28 28 27 26 26 26 26 27 28 28
0
50
100
150
200
250
300
350
400
0 600 1200 1800 2400 3000 3600 4200 4800 5400Time [s]
Ste
el T
empe
ratu
re [°
C]
0.0
2.5
5.0
7.5
10.0
12.5
15.0
17.5
20.0
Hea
t Flu
x [k
W/m
2 ]
Temperature of exposed side
Temperature of un-exposed side
Constant heat flux from 3m x 3m fire
Fire size: ExperimentalFire base: 2m x 2mDistance: 2m
Fire size: 7MW Steady StateFire base: 3m x 3mDistance: 2m
7 8
Perform a non-linear structural analysis of columns• Vulcan was used for the analyses• Fire Limit State Loading was used
-10
0
10
20
30
40
50
60
70
0 600 1200 1800 2400 3000 3600 4200 4800 5400
Time [s]
Dis
plac
emen
t [m
m]
L6D2F1.4 - Vertical displacement at top of column
L7D2F1.4 - Vertical displacement at top of column
0
50
100
150
200
250
300
350
0 600 1200 1800 2400 3000 3600 4200 4800 5400Time [s]
Dis
plac
emen
t [m
m]
L6D2F1.4 - Horizontal displacement at mid-height of column
L7D2F1.4 - Horizontal displacement at mid-height of column
Vertical displacement
Horizontal displacement
0
50
100
150
200
250
300
350
400
450
0 10 20 30 40 50 60Time [minutes]
Dis
plac
emen
t [m
m]
Average heating
Non-uniform heating
-5
0
5
10
15
20
25
30
35
0 10 20 30 40 50 60Time [minutes]
Dis
plac
emen
t [m
m]
Average heating
Non-uniform heating
615 641 666 670 671 671 670 666 641 615
582 582
544 544
531 531
525 525
520 520
515 515
508 508
491 491
478 466 452 449 448 448 449 452 466 478
Exposed face
Unexposed face
0
100
200
300
400
500
600
700
0 10 20 30 40 50 60Time [minutes]
Ste
el T
empe
ratu
re [°
C]
Temp. of exposed face
Temp. of un-exposed face
Average temperature
Comparison with the Eurocode approach
Eurocodes should consider non-uniform heating of external steelwork or propose boundaries of applicability.
9 10
Conclusion
• Buro Happold FEDRA has been successfully designing buildings for fire for
the last 20 years
• The majority of the external columns will have no applied fire protection
• Significant cost savings
• Less maintenance and better architectural finish
• For slender columns and unrestrained beams the non-uniform heating must
be included
• The external steelwork approach in the Eurocode does not include this or
highlight the need for itBuro Happold FEDRA
Thank you for your attention!
11 12
