1. QUAKE SUMMIT 2012, Boston, July 12, 2012
Seismic Performance Evaluation of Energy Efficient Structural Insulated Panels (SIPs) Using Hybrid Simulation and Cyclic Testing
Selim Günay, PostDoctoral Researcher
KHALID MOSALAM, PROFESSOR, PROJECT PI
SHAKHZOD TAKHIROV, SITE OPERATIONS MANAGER
QUAKE SUMMIT 2012, Boston, July 12, 2012

2. QUAKE SUMMIT 2012, Boston, July 12, 2012
Introduction
Structural Insulated Panels (SIPs) are composite panels for energy efficient construction
Composed of an energy-efficient core placed in between facing materials
Their application in seismically hazardous regions is limited due to unacceptable performance as demonstrated by cyclic testing
Limited number of tests with more realistic dynamic loading regimes
Structural insulated panels are composite panels mainly employed for energy efficient construction
Hybrid simulation is ideal to test SIPs with a variety of structural configurations and ground motion excitations
QUAKE SUMMIT 2012, Boston, July 12, 2012

3. Test Setup Loading Steel Tube Reconfigurable Reaction Wall Actuator
Specimen
Gravity Loading
Support beam
Structural insulated panels are composite panels mainly employed for energy efficient construction
QUAKE SUMMIT 2012, Boston, July 12, 2012

4. QUAKE SUMMIT 2012, Boston, July 12, 2012
Test Setup
QUAKE SUMMIT 2012, Boston, July 12, 2012

5. Test Setup and Specimen
Two 4 ft x 8 ft panels connected with panel to panel thermal spline
QUAKE SUMMIT 2012, Boston, July 12, 2012

6. Test Specimen 7/16” OSB Skins 3-5/8” EPS Insulating Foam
QUAKE SUMMIT 2012, Boston, July 12, 2012

7. Instrumentation Tube sliding Top gap opening Top vertical sliding
Bottom gap opening
Bottom vertical sliding
Two 4 ft x 8 ft panels connected with panel to panel thermal spline
Left Uplift
Right Uplift
QUAKE SUMMIT 2012, Boston, July 12, 2012

8. Test Matrix Compare the responses of conventional wood panel vs SIPs
Specimen
Protocol
Gravity
Nail spacing [in]
Remarks S1
CUREE No 6
Conventional wood panel S2 - S3
Yes S4 HS
Near-fault pulse-type GM S5 3 S6 S7
Long duration, harmonic GM S8
Near-fault GM; 3 stories computational substructure
Compare the responses of conventional wood panel vs SIPs
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Investigate the effects of
A parameter related to the design and construction of panels: Nail spacing
Parameters related to loading
Presence of gravity loading
Lateral loading: CUREE protocol vs HS
Type of ground motion (Pulse type vs Long duration, harmonic)
A parameter related to HS: presence of an analytical substructure
QUAKE SUMMIT 2012, Boston, July 12, 2012

9. QUAKE SUMMIT 2012, Boston, July 12, 2012
Hybrid Simulation
Specimens S4, S5, S7 c m
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen
m (kip-sec2/in) ξ
k (kip/in)
c (kip-sec/in)
T (sec) S4
0.0325
0.05 18
0.0076
0.27 S5 32
0.0102
0.20 S7
QUAKE SUMMIT 2012, Boston, July 12, 2012

10. QUAKE SUMMIT 2012, Boston, July 12, 2012
Hybrid Simulation
Specimen S8 m
c=αm
Analytical DOF
c=αm m u3
force-displacement relation from previous tests
c=αm m u2 u1 m
c=αm
Experimental DOF
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

11. Hybrid Simulation: Numerical Integration
Explicit Newmark Integration with γ=0.5
Does not require iterations
Does not require knowledge of initial experimental stiffness
Specimen m k
T (sec)
dt (sec)
dt/T S4
0.0325 18
0.27
0.005
0.018 ≤ 1/π S5 32
0.20
0.025 ≤ 1/π S7
0.0125
≤ 1/π S8 -
T4=0.10
0.05 ≤ 1/π
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

12. Hybrid Simulation: Ground Motions
Near fault, pulse-type GM
Long duration, harmonic GM
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

13. Test Results: Global Parameters
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Initial stiffness =fi /di
Force capacity = fc
Ductility =du/dy
Hysteretic energy =
Positive peak displacement = dp
Negative peak displacement = dn
Residual displacement
QUAKE SUMMIT 2012, Boston, July 12, 2012

14. Test Results: Local Parameters
Peaks of local responses
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

15. Conventional Wood Frame (S1)
Test Results: Comparison of Conventional Wood Panel and SIPs (S1 vs S2)
Conventional Wood Frame (S1)
SIPs (S2)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
7/16’’ OSB Skin on both sides
3-5/8” EPS Insulating Foam
Panel to panel thermal connections
Double 2x4’’ 96’’
6’’ nail spacing
7/16” OSB Skin on both sides
2x4’’ 16’’
Double 2x4’’ the ends
6’’ nail spacing
Cyclic Testing with CUREE protocol
QUAKE SUMMIT 2012, Boston, July 12, 2012

16. Initial Stiffness [kip/in] Hysteretic Energy [kip-in]
Test Results: Comparison of Conventional Wood Panel and SIPs (S1 vs S2)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen S1 S2
Initial Stiffness [kip/in]
46.2
12.2
Force Capacity [kip]
11.4
Ductility
7.0
3.6
Hysteretic Energy [kip-in]
201.8
193.1
QUAKE SUMMIT 2012, Boston, July 12, 2012

17. QUAKE SUMMIT 2012, Boston, July 12, 2012
Test Results: Comparison of Conventional Wood Panel and SIPs (S1 vs S2)
Heat transfer analysis using THERM 6.3:
A software developed at Lawrence Berkeley National Laboratory for modeling and analyzing heat-transfer effects in building components S1
(Conventional wood) S2
(SIPs) S1 S2
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

18. Test Results: Effect of Gravity Loading (S2 vs S3)
No gravity loading (S2)
Gravity loading (S3)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Cyclic Testing with CUREE protocol
QUAKE SUMMIT 2012, Boston, July 12, 2012

19. Test Results: Effect of Gravity Loading (S2 vs S3)
Specimen S2 S3
Initial Stiffness [kip/in]
12.2
23.4
Force Capacity [kip]
11.4
9.5
Ductility
3.6
3.5
Hysteretic Energy [kip-in]
193.1
189.2
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen
Bottom ver. sliding
Bottom gap opening
Top ver. Sliding
Top gap opening
Uplift right
Uplift left
Tube sliding S2
0.71
0.04
0.73
0.27
0.02 S3
0.49
0.01
0.50
0.14
0.03
* All units in inches
QUAKE SUMMIT 2012, Boston, July 12, 2012

20. Test Results: Effect of Nail Spacing (S4 vs S5)3” 6”
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Hybrid Simulation with Pulse-type GM
QUAKE SUMMIT 2012, Boston, July 12, 2012

21. Test Results: Effect of Nail Spacing (S4 vs S5)
Specimen S4 S5
Initial Stiffness [kip/in]
22.9
35.5
Force Capacity [kip]
8.6
15.6
Ductility
2.5
3.7
Hysteretic Energy [kip-in]
152.7
363.1
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen DE
MCE
1.5MCE S4 S5
Peak Disp. (+)
2.7
1.3
4.7
3.5 -
5.8
Peak Disp. (-)
-2.8
-1.0
-3.2
Residual Disp.
1.5
0.1
0.8
QUAKE SUMMIT 2012, Boston, July 12, 2012

22. Test Results: Effect of Nail Spacing (S3 vs S6)3” 6”
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Cyclic Testing with CUREE protocol
QUAKE SUMMIT 2012, Boston, July 12, 2012

23. Test Results: Effect of Nail Spacing (S3 vs S6)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen S3 S6
Initial Stiffness [kip/in]
23.4
32.7
Force Capacity [kip]
9.5
16.2
Ductility
3.5
4.8
Hysteretic Energy [kip-in]
189.2
309.9
QUAKE SUMMIT 2012, Boston, July 12, 2012

24. Test Results: Effect of Lateral Loading (S6 vs S7)
Cyclic Testing with CUREE Protocol for Ordinary GM (S6)
Hybrid Simulation with Long Duration, Harmonic GM (S7)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Nail spacing: 3”
QUAKE SUMMIT 2012, Boston, July 12, 2012

25. Test Results: Effect of Lateral Loading (S6 vs S7)
Specimen S6 S7
Initial Stiffness [kip/in]
32.7
33.2
Force Capacity [kip]
16.2
15.5
Ductility
4.8
3.4
Hysteretic Energy [kip-in]
309.9
1077.8
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen S6 S7
Peak Disp. (+)
4.7
3.3
Peak Disp. (-)
-4.7
-4.2
Residual Disp.
0.0
0.3
QUAKE SUMMIT 2012, Boston, July 12, 2012

26. Test Results: Effect of Ground Motion Type (S5 vs S7)
Hybrid Simulation with Pulse-Type GM (S5)
Hybrid Simulation with Long Duration, Harmonic GM (S7)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Nail spacing: 3”
QUAKE SUMMIT 2012, Boston, July 12, 2012

27. Test Results: Effect of Ground Motion Type (S5 vs S7)
Specimen S5 S7
Initial Stiffness [kip/in]
35.5
33.2
Force Capacity [kip]
15.6
15.5
Ductility
3.7
3.4
Hysteretic Energy [kip-in]
363.1
1077.8
Specimen DE
MCE
1.5MCE S5 S7
Peak Disp. (+)
1.3
1.1
3.5
2.2
5.8
3.3
Peak Disp. (-)
-1.0
-3.2
-2.0 -
-4.2
Residual Disp.
0.1
0.0
0.8
0.3
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

28. Test Results: Effect of Ground Motion Type (S5 vs S7)
Specimen DE
MCE
1.5MCE S5 S7
Peak Disp. (+)
1.3
1.1
3.5
2.2
5.8
3.3
Peak Disp. (-)
-1.0
-3.2
-2.0 -
-4.2
Residual Disp.
0.1
0.0
0.8
0.3
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen
Bottom ver. sliding
Bottom gap opening
Top ver. sliding
Top gap opening
Uplift right
Uplift left
Tube sliding DE S5
0.26
0.02
0.27
0.03
0.08
0.07
0.18 S7
0.23
0.21
0.15
0.04
MCE
0.63
0.05
0.64
0.09
0.14
0.12
0.19
0.45
0.43
0.53
0.06
QUAKE SUMMIT 2012, Boston, July 12, 2012

29. Test Results: Effect of Analytical Substructuring (S5 vs S8)
Hybrid Simulation with no Analytical Substructure (S5)
Hybrid Simulation with Analytical Substructure (S8)
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Pulse-type GM
QUAKE SUMMIT 2012, Boston, July 12, 2012

30. Test Results: Effect of Analytical Substructuring (S5 vs S8)
Specimen S5 S8
Initial Stiffness [kip/in]
35.5
38.3
Force Capacity [kip]
15.6
16.0
Ductility
3.7
4.0
Specimen DE
MCE S5 S8
Peak Disp. (+)
1.3
1.2
3.5
2.4
Peak Disp. (-)
-1.0
-1.7
-3.2
-3.1
Residual Disp.
0.1
0.0
0.8
0.4
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
Specimen
Bottom ver. sliding
Bottom gap opening
Top ver. sliding
Top gap opening
Uplift right
Uplift left
Tube sliding DE S5
0.26
0.02
0.27
0.03
0.08
0.07
0.18 S8
0.37
0.04
0.09
0.11
0.13
MCE
0.63
0.05
0.64
0.14
0.12
0.19
0.65
0.55
0.16
QUAKE SUMMIT 2012, Boston, July 12, 2012

31. QUAKE SUMMIT 2012, Boston, July 12, 2012
Concluding Remarks
Finite element heat transfer analyses quantitatively show the thermal insulation efficiency of SIPs compared to conventional wood panels.
Effect of nail spacing is significant on the structural performance of SIPs.
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

32. QUAKE SUMMIT 2012, Boston, July 12, 2012
Concluding Remarks
Hybrid simulation provides the force-deformation envelope that can also be gathered from a cyclic test. But it also provides response values, where the cyclic test would require complimentary analytical simulations to get the response values.
Although the global and local responses of SIPs with and without analytical substructuring are not dramatically different, there is a need for analytical substructuring for a more realistic representation.
-Panels are connected with thermal spline connections.
-Framing timber consists of 2×4 lumber placed at the top and bottom and on the sides
and they are bearing on the top and bottom plates which are 2×6 lumber
QUAKE SUMMIT 2012, Boston, July 12, 2012

33. QUAKE SUMMIT 2012, Boston, July 12, 2012
Thank you
QUAKE SUMMIT 2012, Boston, July 12, 2012