Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 144 (2016) 1356 – 1363
12th International Conference on Vibration Problems, ICOVP 2015
Seismic Retrofitting of a Process Column using Friction Dampers A. Ravi Kirana*, M. K. Agrawala and G. R. Reddya a
Reactor Safety Division, BARC, Tombay, Mumbai, 400085, India
Abstract During seismic reevaluation of a process column, it was found that anchor bolts are not adequate for the present seismic load. A retrofitting scheme using friction dampers of 30 kN capacity has been worked out for its qualification. Double Sliding Friction Dampers (DSFD) have been designed and fabricated. Friction dampers absorb vibration energy by friction forces between contacting plates. Normal force required to induce desired slip load is applied through pre-tensioning of the bolts. The pretensioning of bolts is done by tightening bolts to a specified value of torque using wrench. The bolt torque required to obtain desired slip load of damper is obtained from the damper characteristics. The characterization of friction dampers has been carried out by means of cyclic tests using hydraulic actuator. In the present paper, the details of friction damper and characterization are provided. The details of retrofitting of process column using these dampers are also presented. © Published by Elsevier Ltd.Ltd. This is an open access article under the CC BY-NC-ND license © 2016 2016The TheAuthors. Authors. Published by Elsevier (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICOVP 2015. Peer-review under responsibility of the organizing committee of ICOVP 2015 Keywords: Fricton Damper; Retrofitting; Response reduction; Iterative Response Spectrum method
1. Introduction A process column is a tall cylindrical tower and is widely used in process industries. Structural integrity of a process column was assessed for present seismic requirements and found that the anchor bolts are not meeting the codal requirements. It is proposed to qualify the tower by strengthening the foundation by increasing the number of anchor bolts. This scheme requires digging around the existing foundation and adding extra anchor bolts etc. It was found that due to space constraints this scheme is not suitable. Hence, it is found that retrofitting using passive control devices may be the suitable option for seismic requalification of process columns. Various passive control devices such as viscoelastic damper, elasto-plastic damper, tuned mass damper, tuned liquid damper and friction
* Corresponding author. Tel.: +91-22-25593546; fax: +91-22-25505151. E-mail address:
[email protected]
1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of ICOVP 2015
doi:10.1016/j.proeng.2016.05.165
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
1357
damper may be used for response reduction. These devices control the motion of the system by significantly increasing the equivalent damping due to energy dissipation in the devices. These devices can be easily replaced after an earthquake if required due to unexpected damage. The performance of viscoelastic dampers is affected by temperature and stiffness degradation occurs due to ageing. The limitation of tuned dampers is the requirement of accurate tuning to the natural frequency of the process column. On the other hand, friction dampers do not possess many of these limitations. The friction brake is widely used to reduce the kinetic energy of a moving body and it is the most effective, reliable and economical means to dissipate the energy. For centuries, engineers have successfully used this concept to control the motion of machinery and automobiles. Based on past studies, it is evident that friction dampers can be effectively used in reducing the seismic response of structures [1-3], equipment and piping systems [4-6]. From the tests carried out by researchers [1], it is observed that the performance of friction damping devices is reliable, repeatable and possess large rectangular loops with negligible fade over several cycles of reversals. Hence, friction dampers have been selected for seismic response reduction of the process column. The friction damper is a passive energy dissipation device which absorbs vibration energy by friction forces between contacting plates. Double Sliding Friction Dampers (DSFD) have been used for the present work. The dampers have been designed, fabricated and their characterization has been done by means of cyclic tests using hydraulic actuator. The present paper provides the details of experimental characterization of friction damper and retrofitting of the process column. Nomenclature ΔE ESE ξs ξd F σca σta τa σy µp µb Fb Db σ τ E k l r
Energy dissipated in hysteretic deformation Strain energy Structural damping Damping due to Friction damper Capacity of friction damper Allowable compressive stress Allowable tensile stress Allowable shear stress Yield stress Coefficient of friction between liner and plate Coefficient of friction for steel threads Normal force in each bolt Nominal diameter of bolt Axial stress (tensile/ compressive) in the plate Shear stress in the plate Young’s modulus Effective length factor Length of support Radius of gyration
2. Details of friction damper and its characterization The schematic of friction damper is shown in Fig. 1. It comprises of one inner (Plate I) and two outer steel plates (Plate II). Teflon liners are provided between the contacting plates. Friction forces are produced between the steel plates and Teflon liners. Normal force required to induce desired friction force between the contacting plates is applied through pre-tensioning of the bolts between Plates I and II. Oblong holes are provided in the inner plate to facilitate sliding. The pre-tensioning of bolts is provided by tightening bolts to a specified value of torque using wrench. The photograph of two friction dampers is given in Fig. 2. The test setup used for characterization of damper is shown in Fig. 3. One end of the damper is fixed rigidly through bracket and other end is connected to hydraulic actuator. Initially all bolts between plates-II and III are tightened rigidly and a torque of 40 lb-ft is applied
1358
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
to the bolts between plates-I and II. A quasi-static cyclic displacement of 5 mm amplitude has been applied to the damper. The cyclic load has been repeated for ten cycles. The test has been carried out for various torque values ranging between 40 lb-ft to 100 lb-ft in the intervals of 10 lb-ft. The resulting cyclic load-displacement curves for various torque values in lb-ft (T40 to T100) are shown in Fig. 4. It is observed that a stable hysteretic behavior was obtained over the repeated loading cycles. Then slip load for each applied torque has been obtained and plotted in Fig. 5. It is observed that variation of slip load is linear with respect to applied torque. A torque of 60 lb-ft is applied to the bolts of two friction dampers shown in Fig. 2 to obtain the slip load (F) of 30 kN. The coefficient of friciton between the liner and steel plate is calculated as follows: Torque applied for each bolt, T = µ b Fb Db = 60 lb-ft = 81.4 N-m (µ b= 0.2, Db= 0.02m) Normal force in each bolt, Fb = T/( µ b Db) = 20.35 kN Coefficient of friction between the liner and plate, µ p = F/ (No. of contact points x Fb) = 0.12 M 20
t 18
PLATE- II
t 15
t 20
Section Y-Y
PLATE-III
PLATE- I 700
PLATE-IV
75
TEFLON LINER
M 20 Φ 52
200
t 15
325 75 125 75 25 7
Φ 52
75
100
t 20
100
125
125
100
50
125 100
1100 60
Section X-X
22
t 20 Y 200 X
Y
100
X
225
Fig.1 Schematic of Friction Damper
Damper Plate-III
Bracket
Plate-II Plate-I Hydraulic actuator
Fig. 2 Photograph of friction dampers
Fig. 3 Photograph of test setup used for characterization of friction dampers
1359
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363 50
45 T40 T50 T60 T70 T80 T90 T100
30
Slip load (kN)
20 10 0 -10 -20
35 30 25 20
-30 -40 -50
Experiment Linear fit
40 Slip load (kN)
40
-5.0
-2.5
0.0
2.5
Displacement (mm)
Fig. 4 Cyclic load-displacement curves
5.0
15 40
50
60
70 80 Torque (lb-ft)
90
100
Fig. 5 Variation of slip load with applied torque
3. Details of the retrofitting scheme for process column using friction dampers The scheme of retrofitting for process column using friction dampers is shown in Fig. 6. The process column has a height of 22.549 m and inner diameter of 4.25m and is provided with cross flow sieve trays. The tower is made of carbon steel, SA516 Gr 70 and is supported on skirt with base plate [7]. The upper cylindrical portion has 40 mm and the lower portion has 42 mm thicknesses. The top and bottom hemi-spherical portions have thickness of 25 and 28 mm respectively. The natural frequencies and corresponding mass participation of the process column are given in Table 1. The tower has eight number of M24 foundation bolts and it is found that these bolts are not sufficient to meet present seismic load requirements of Maximum Considered Earthquake (MCE) condition [8]. Two friction dampers are provided in orthogonal directions at 19.5m / 19.2 m elevations and the slip load for each damper is set to 30 kN. Design checks for various parts of Friction damper are provided in Appendix A. One end of the damper is attached to process column while the other end is connected to the adjacent supporting structure. The input seismic load is conservatively taken by considering the envelope of excitation at base and damper location. The analysis is carried out by using the envelope of ground spectrum (MCE) for 4% damping and Floor Response Spectrum (FRS) of supporting structure at 19.5m EL. Fig. 7 shows ground response spectrum and FRS at 19.5m EL of supporting structure (±15% peak broadened). The envelope of these two spectra is computed and is given in Fig. 8. The base moment and displacement at 19.5m of the process column before retrofitting are 4.886 MN-m and 3.108 mm respectively. Iterative Response Spectrum method is used to evaluate the seismic response of the process column with friction dampers. Flow chart for Iterative Response Spectrum method is given in Fig. 9. Initially, process column with structural damping (ξs) of 4% (without dampers) is considered and the response is obtained from the response spectrum analysis [9]. The displacement obtained at the damper location (δi) is used in the first iteration. The equivalent damping of the friction damper is obtained using the ratio of energy dissipated in hysteretic deformation ΔE and strain energy ESE. The equivalent viscous damping ratio, ξd is given by, ߦௗ ൌ
ଵ
ቀ
οா
ସగ ாೄಶ
ቁ
(1)
Total damping (ξs+ξd) is used in the next iteration. Using this total damping the seismic response of the process column is again obtained using response spectrum analysis. The displacement obtained at the damper location (δi+1) is used in this iteration to modify the total damping. This process is repeated until the difference of displacements in the successive iteration is less than 5%. Variation of displacement and damping from friction damper (ξd) in various iterations are given in Figs. 10 and 11 respectively. It is observed that an additional damping of 33% is obtained from friction damper and displacement at 19.5 m elevation is reduced from 3.108 to 0.7 mm. The base moment has been reduced to 1.168 MN-m for one directional spectral loading. The resultant base moment for tri-axial excitation is 1.65 MN-m.
1360
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
Friction Damper El 19.5 m F Friction Damper El 19.2 m Y X Z
Support Supp port structure
El - 0
FD-1 El 19.2 m
Foundation bolts FD-2 El 19.5 m
Fig. 6 Scheme of retrofitting for process column using Friction Dampers
Table 1. Natural frequencies and mass participation of the process column Freq. No.
Natural frequency (Hz)
X- mass participation (% mass)
Y- mass participation (% mass)
Z- mass participation (% mass)
1.
6.54
69.4
1.7e-17
69.4
2.
33.29
21.5
2.5e-16
21.5
3.
45.64
2e-17
89.6
2e-17
1.25
Ground spectrum FRS at 19.5 m
1.00
Acceleration (g)
Acceleration (g)
1.25
0.75 0.50 0.25 0.00
1.00 0.75 0.50 0.25 0.00
0
5
10
15
20
25
30
35
Frequency (Hz) Fig. 7 Ground response spectrum (MCE) and FRS at 19.5 m of Elevator structure
0
5
10
15
20
25
30
Frequency (Hz) Fig. 8 Envelope response spectrum
35
1361
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
Evaluate the fundamental model properties of tower
Using the structural damping (ξs) alone obtain the relative displacement (δ =δi) at friction damper location from Response Spectrum analysis Evaluate energy dissipated in hysteretic deformation ΔE and strain energy ESE for structural configuration with relative displacement at friction damper location as δ Calculate friction damper damping (ξd) using Eq.(1) and then total damping (ξs+ξd) δ = δi+1 Using the total damping (ξs), obtain the relative displacement (δi+1) at friction damper location from Response Spectrum analysis
If (δi+1- δi) / δi < 0.05
No
Yes Relative displacement at damper location is δi+1
35
3.0
30
Damping ratio (%)
Displacement (mm)
Fig. 9 Flow chart for Iterative Response Spectrum method 3.5
2.5 2.0 1.5 1.0 0.5
25 20 15 10 5 0
0.0 0
1
2
3
4
5
6
7
8
0
9
1
2
3
Iteration
4
5
6
7
8
9
Iteration
Fig. 10 Variation of displacement in various iterations
Fig. 11 Variation of damping (ξd) in various iterations
The comparison of seismic demand and capacity of foundation loads after retrofitting are given in Table 2. It is observed that the seismic demand is reduced to 1.65MN-m which is less than the capacity of existing foundation bolts of 1.84MN-m. Hence, existing foundation bolts of the process column are qualified for MCE level of earthquake after retrofitting with 30 kN friction dampers. Table 2. Comparison of seismic demand and capacity of foundation bolts after retrofitting S. No.
Description
Demand before retrofitting
Demand after retrofitting
Capacity of foundation bolts
1.
Base moment (MN-m)
3.312
1.65
1.84
2.
Shear force at base (kN)
290
103.2
513
1362
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
4. Conclusion During seismic re-evaluation of a process column, it is found that existing foundation bolts are not sufficient to meet the present seismic load requirements. Hence, seismic retrofitting has been worked out for the process column using friction dampers of 30 kN capacity. Double Sliding Friction Dampers (DSFD) have been designed and fabricated. The characterization of these dampers has been carried out by cyclic tests. From the damper characteristics, torque of 60 lb-ft is applied to the bolts of two dampers to obtain a slip load of 30 kN. Iterative Response Spectrum method has been used for the seismic analysis. It is observed that existing foundation bolts of the column are qualified for MCE condition after providing Friction dampers in two orthogonal directions.
Acknowledgements The authors would like to thank the engineers of Heavy Water Board, Mumbai for useful discussions. The authors also wish to thank the engineers and technicians of Centre for Design and Manufacture (CDM) division of BARC, Mumbai for the support received during preparation of test setup.
Appendix A. Design checks for various parts of Friction damper Design checks have been carried out for various parts of the damper as per as per ASME Sec–III, Division 1; Subsection–NF–3320 [10]. The stresses in various parts are computed and compared with allowable stresses given in NF-3320.
A.1. Allowable Stresses as per NF-3320 The allowable stress in axial tension (σta) is 0.6 σy and the allowable stress in shear (σta) is 0.4 σy. The allowable stress in axial compression (σca) is calculated as: ª1 kl / r 2 / 2C 2 º ½V ®« c »¼ ¾ ¯¬ ¿ y ª ª 5 / 3 «3 kl / r /8C « kl / r 3 / 8C 3 º c ¬ c »¼ ¬
V ca
(For kl/r < Cc)
@
V ca
12S 2 E 2 23 kl / r
(For kl/r > Cc)
Where, yield stress, σy= 250 MPa and
C
c
2S 2 E / V y
125.66
A.2. Axial stresses in various parts of friction damper Capacity of the damper, F = 30 kN Area of plate-I, Aa1 = 18x (200-22x2) = 2808 mm2 Axial stress in plate-I, σ1 = F/ Aa1= 10.68 MPa Area of plate-II, Aa2 = 15x (200-22x2) = 2340 mm2 Axial stress in plate-II, σ2 = F/ 2Aa2= 6.41 MPa Area of plate-III, Aa3 = 18x (200-22x2) = 2808 mm2
A. Ravi Kiran et al. / Procedia Engineering 144 (2016) 1356 – 1363
1363
Axial stress in plate-III, σ2 = F/ Aa3= 10.68 MPa The allowable stress in axial tension (σta) is 0.6 σy i.e., 150 MPa (σy= 250 MPa). For plate-I, kl/r is 193.18 and σca is 27.6 MPa. For plate-II, kl/r is 161.66 and σca is 39.4 MPa. For plate-III, kl/r is 62.5 and σca is 119.2 MPa. Hence, the axial stresses in all plates are less than the permissible limits. A.3. Shear stresses in various parts of friction damper Area of plate-I, As1 = 2 x 100 x 20 = 4000 mm2 Shear stress in plate-I, τ1 = F/ As1= 7.5 MPa Area of plate-II, As2 = 4 x 75 x 15= 4500 mm2 Shear stress in plate-II, τ2 = F/ 2As2= 3.33 MPa Area of plate-III, As3 = 4 x 75 x 18= 5400 mm2 Shear stress in plate-III, τ3 = F/ As3= 5.55 MPa The allowable stress in shear (τ a) is 0.4 σy i.e., 100 MPa. Hence, the shear stresses in all plates are less than the permissible limit. A.4. Axial and shear Stresses in bolts Normal force in each bolt = Fb = F /(µ p N)=30 /(0.12x12) = 20.83 kN Axial stress in bolt, σb = Fb/ Ab= 66.3 MPa Shear stress in bolt, τb = µ p Fb / Ab= 7.96 MPa The allowable stress for bolts in axial tension (σta) is 0.6 σy i.e., 384 MPa (σy= 640 MPa). The allowable stress for bolts in shear (σta) is 0.4 σy i.e., 256 MPa. Hence the axial and shear stresses in bolts are less than allowable limits. References [1] A. S. Pall, C. Marsh, and P. Fazio, Friction Joints for Seismic control of Large Panel Structures, Journal of Prestressed Concrete Institute, 256 (1980) 38-61. [2] R. Chandra, M. Masand, S. K. Nandi, C. P. Tripathi, R. Pall and A. Pall, Friction-dampers for seismic control of la gardenia towers south city, gurgaon, india, Paper No. 2008, Proceedings of the 12th World Conference on Earthquake Engineering (12WCEE), New Zealand, 2000. [3] I. H. Mualla, B. Belev, Performance of steel frames with a new friction damper device under earthquake excitation, Engineering Structures, 24 (2002) 365-371. [4] G. R. Reddy, K. Suzuki, T. Watanabe and S. C. Mahajan, Linearization techniques for seismic analysis of piping system on friction support, J. PressureVessel Technology, ASME 121 (1999) 103–108 [5] S.V. Bakre, R.S. Jangid and G.R. Reddy, Response of piping system on friction support to bi-directional excitation, Nuclear Engineering and Design 237 (2007) 124–136. [6] J. C. Anderson and A K Singh, Seismic response of pipelines on friction supports, Journal of Engineering Mechanics Division, ASCE, 102EM2 (1976) 275-291. [7] A. Ravi Kiran, M. K. Agrawal, D. K. Sakhrodia, J. Sahu, K. V. Tale, G. R. Reddy, B. Biswas, R. K. Singh, K. K. Vaze, A. Bhowmick and R. V. Gupta, Re- Evaluation of Equipment of Heavy Water Plant- Kota for Earthquake and Wind Loads, RSD Report, BARC, Mumbai, India, 2014. [8] IS criteria for Earthquake Resistant Design of Structures, part 4 Industrial structures including stack like structures, IS 1893 (part 4), 2005. [9] A. K. Chopra, Dynamics of Structures - Theory and application to Earthquake Engineering, Prentice-Hall, Upper Saddle River, N.J, 2001. [10] American Society of Mechanical Engineers, ASME Boiler and Pressure Vessel Code Section III, Division 1, Subsection–NF–3320, 2007.
