J. Korean Soc. Hazard Mitig Search

CLOSE


J. Korean Soc. Hazard Mitig. > Volume 17(2); 2017 > Article
Ko, Kim, and Kim: Barrier Design for Parking Garage Using W-shape Guardrail Beam and Steel Tube Posts

Abstract

2ton-20km/h의 충격에 대응할 수 있는 주차타워용 빔-지주 타입 난간을 개발하였다. 일반적으로 도로에 많이 사용되는 W형 빔과 철제 원형 관을 지주로 사용하였다. 지주는 콘크리트 슬래브에 앵커로 고정되는데 베이스의 앞에 2개의 M20 앵커를 뒤쪽에는 1개의 M20 앵커를 사용하였으며 지주 간격은 2m이다. 설계된 난간의 횡방향 지지력은 메커니즘 해석법으로 산정하였다. 이를 위하여 낙하물을 이용한 충격시험을 실시하였는데, 이 실험을 통하여 특별한 앵커로 지지된 지주의 동적 횡방향지지력을 구할 수 있다. 본 연구에서는 세 가지 타입의 앵커로 고정되는 지주를 고려하였다. 설계된 난간의 지지하중을 산정하고 이를 이전 연구에서 산출한 소요하중과 비교하였다. 소요하중 이상을 지지력을 확보할 수 있도록 설계된 난간에 대하여 LS-Dyna 프로그램을 이용한 충돌 시뮬레이션과 실차 충돌실험을 실시하여 난간의 구조적 적합성과 충돌차량의 탑승자 안전을 검토하였다.

요지

Beam-post type barrier for parking garage is designed to accommodate 2ton-20km/h impact. It uses conventional W-beam section and circular steel posts. Posts are anchored to the concrete slab with a base anchor system which has 2-M20 anchors in front and 1-M20 at the back, and are at 2m intervals. Total lateral load capacity of the design was estimated by failure mechanism analysis. For the estimation, drop weight tests were performed, which gave dynamic lateral load resistance of a post with specific anchor type. Three different anchor types were considered. Calculated total load capacity was compared with the required impact force from the previous work. For the design, whose capacity was confirmed to exceed the required force, vehicle impact test and LS-DYNA simulation were made to verify the structural adequacy and occupant safety.

1. Introduction

To resolve the limited parking space, parking garages are being built frequently in urban area. As the number of the parkade increases, so does the number of the vehicle falling accidents. Falling accidents involve the failure of the barrier system of a parking garage. In the circumstances, Korean government made it mandatory to design a barrier system in parking garage according to the “parkade regulation” by MOCT. The regulation specifies impact condition of 2ton- 20km/h relating to the design force, which is a bit obscure to convert to a force. For the reason, in the design of roadside barrier, where the impact condition is designated by the vehicle mass and impact speed, vehicle crash test is used to verify the performance of the barrier. Parkade barrier is very similar to the bridge barrier since it is constructed on a concrete slab or steel deck.
Parkade barrier has been designed as a concrete wall type or beam-post type and the design has been made by static forces generally. To design a beam-post type barrier for an impact condition of mass and velocity, equivalent impact force must be known. There is no specific rule to convert the impact condition to impact force. Roughly it can be estimated by Olson Model or impact simulations using Barrier VII. Both the method has limitations since to estimate a force for barrier design, a design must be assumed first. To use the Olson Model for impact force estimation, deflection of a barrier must be known, and to estimate the design force using the Barrier VII simulation, details of a barrier must be known in advance. This is an engineering deadlock. In order to make a rational estimate of the impact force, good assumption of the lateral displacement is necessary in Olson Model and good assumption of the structural details is important in Barrier VII simulation.
In the previous work by Ko et al. (2015) on the design force for parkade design, impact force was estimated using the Olson Model and Barrier VII simulation for a beam-post type barrier with three different base anchor types. Though the estimated force is structure-specific, if the structure type and span length of a design are similar to the type used in the design force estimation, the estimated impact force can be used as a design purpose for slightly modified structures. In this paper, design force will be summarized and a design of parkade barrier using conventional W-beam and steel tubular posts will be made. This type of barrier has been proven to be crashworthy and cost-effective, and is widely used for roadside barrier. Then total resistance of the design will be estimated using plastic analysis of the design. To calculate total resistance, the lateral load carrying capacity of one post is important. So, the lateral load carrying capacity of a post in impact mode was measured from drop weight test. Using the dynamic load carrying capacity of a post depending on different base anchor types, total resistance capacity of a barrier system was calculated for different failure modes and the structural adequacy of the design was confirmed. Then the impact test and simulations were made for 0.9ton-20km/h impact condition and both the structural adequacy and safety of the occupant of the vehicle were studied.

2. Design Force for a Beam-Post Type Barrier of the Parking Garage

In the previous study (Ko et al., 2015), 0.9ton-20km/h impact condition was converted to a force by means of Olson Model and Barrier VII simulation. To use the formula in Olson Model (Eq. 1), the lateral deflection D of a barrier must be known in advance.
(1)
AvgGlat=V02sin2θ2g[Asinθ0.5B(1cosθ)+D]
where
A: Distance from vehicle’s front end to center of mass
B: Half of the vehicle width (m)
D: Lateral displacement of railing (from Barrier VII simulation)
If the stiffness of the vehicle and rail could be idealized as a linear spring, the average and maximum lateral impact force can be calculated as Eq. 2 and Eq. 3 respectively
(2)
AvgFlat=(AvgGlat)W
(3)
MaxFlat=π2AvgFlat
In the study, rather simply assuming the value, Barrier VII simulation results were used. For Barrier VII simulation, however, more details of a barrier system were needed. To get the design force, following preliminary design composed of W-beam and steel tube posts was made. W-beam has the same sectional and mechanical properties as are used in the standard drawings of Korean roadside barrier. It uses SS400 structural steel and has the sectional area of 18.8cm2. Posts are of ∅101.6×4.2mm (SS400 steel). Even the sectional dimensions of the beam and posts are the same, structural properties of the barrier system will vary significantly depending on the span length between posts and connection method of the posts to the slab. Considering that, Base plate was assumed to have 4 anchor bolts. Since the length of W-beam in the market is 2330mm, the defections of the barrier with span length of 2m, 2.5m, 3.0m were studied using Barrier VII program.
In this study, Barrier VII simulations of 0.9ton-20km/h vehicle impacts head-on to W beam-post barriers with three different base anchor types were run. From the simulations, lateral displacements were measured as in Table 1. Then, the values of the Table 1 were used for impact force estimations by Olson Model and were compared with the lateral impact force measured directly from Barrier VII simulations as in Table 2.
Table 1
Dynamic Defection per Span Lengths
 Span Length (l)   Case I   Case II 
2.0 m 21.80 cm 22.51 cm
2.5 m 26.62 cm 28.30 cm
3.0 m 31.68 cm 32.36 cm
Table 2
Impact Load Using Barrier Ⅶ and Olson Model
Load  Impact Point  Method  Impact Load (KN) for Span Lengths (m) 
2.0 m 2.5 m 3.0 m
Average Case I Barrier VII 24.64 23.37 22.53
Olson Model 22.56 21.79 21.04
Case II Barrier VII 25.04 23.77 22.62
Olson Model 22.45 21.54 20.94
Maximum Case I Barrier VII 50.94 49.53 48.50
Olson Model 35.42 34.22 33.04
Case II Barrier VII 56.45 52.44 49.68
Olson Model 35.12 33.81 32.88
Table 2 summarizes the impact force by Barrier VII simulation and Olson Model for three different span lengths.
In the Table 1 and Table 2, Case I is the case when the vehicle impacts the post at the center of the barrier and Case II is the case when the vehicle hits the center of the beam. So, barrier model of Case I is composed of 5 spans and Case II is composed of 6 spans.

3. Design and Strength Assessment

3.1 Lateral load carrying capacity of a post

In the formulation for load carrying capacity of a barrier system, the lateral load Pb causing plastic flexural failure of the post at the ground level takes considerable portion. It varies not only by the size of the post but also by the way it is connected to the slab. It can be obtained from static and dynamic test. Static test has already been made in the previous research for the three different types of base connections. In this research, dynamic drop tests are made and compared with the static test results.
Dynamic drop test will be simulated using LS-DYNA program as well.
In a drop test, the weight falls from certain height and hits the post fixed to a frame. Impact forces were measured by accelerometer on the weight. Also, Motion BLTZ cube4 of 1000frame/sec speed camera was used to supplement the accelerometer and record the response of the post. Fig. 1 shows the LS-DYNA modeling of the post and weight for three different types of base to the steel frame. Fig. 2 shows the drop tower. In both the tests and simulations drop weight was 350kg and height was 4.51m which gave the impact speed of 9.4m/sec.
Fig. 1
Drop Test Model (Type-1, Type-2, Type-3 Anchor)
KOSHAM_17_02_133_fig_1.gif
Fig. 2
Drop Tower and Holding Frame
KOSHAM_17_02_133_fig_2.gif
Type-1 in Fig. 1 uses base plate of size 200×200×16t with 4-M20 chemical anchors and post is pellet welded from top and bottom of the base plate. Type-2 is similar to Type-1 except that it has four rib stiffeners of size 40×72×8t. Type-3 has 2-M20 anchors in front and 1-M20 at the back and in the front side two anchors are 70mm apart each other to maximize the lateral resistance of the post. This allocation of the anchors was determined from the LS-DYNA simulation in the previous research. Both of the posts of Type-2 and Type-3 are welded to the base plate in the same way as Type-1 post.
Following figures show the sequential pictures of drop tests and simulations of three different type base plates and posts. Also force-deflection graphs are followed.
Fig. 3 shows the drop test and simulation result for the post with Type-1 base anchor at 0.1sec time interval. Simulation results well match with the test results. Fig. 3 also shows the force-deflection relations of the drop test. Two drop test results are compared with the two static test results and simulation results. Two static test results and drop test results are averaged. Average lateral resistance force of the drop test was 33.30kN while 23.23kN in static test.
Fig. 3
Drop Test Results for Type-1 Base Anchor
KOSHAM_17_02_133_fig_3.gif
For the posts with Type-2 and Type-3 base anchors, same tests were made and the results are shown in Figs. 4, 5, which gives average lateral resistance force of 35.24kN from drop test and 24.21kN from static test for Type-2 and of 43.25kN from drop test and 27.93kN from static test for Type-2. It can be shown that since, for all the three different type posts, dynamic load resistances are 40% to 50% higher than that of static load resistance, in calculating the impact resistance of a barrier system, drop test results will be used.
Fig. 4
Drop Test Results for Type-2 Base Anchor
KOSHAM_17_02_133_fig_4.gif
Fig. 5
Drop Test Results for Type-3 Base Anchor
KOSHAM_17_02_133_fig_5.gif
In the force-deflection curves, simulation results show some rebound while test results show slips after the deflection of about 150mm. It is due to the fact that while the fillet welded connections between the post and base plate have been broken in the drop test at the deflection of above 150mm, breakage at the connections have not occurred in the simulations since the elements of base plate and tubular post were merged as an uniform body. This model difference did not make a difference in lateral load capacity estimation as Table 3 shows. In the design of barrier system only the test results were used.
Table 3
Laterla Load Carrying Capacity of Posts
Base Plate Anchor Type
Type-1 Type-2 Type-3
Static (Previous Work) Test 23.23 24.21 27.93
Simulation 24.13 25.32 29.41
Dynamic (Drop Test) Test 33.30 35.24 43.25
Simulation 35.26 37.65 44.95
Table 3 summarizes the lateral load carrying capacity Pp of the posts for each different type base anchor from the drop test and simulations. For comparison purpose, capacities from static tests are also included. From the table, it can be shown that the simulations give the similar results as the test values. Also it is noticeable that dynamic load capacities of the posts are 40 to 50% greater than the static results for all the posts with three different base anchor types.

3.2 Mechanism Analysis and Strength of a Design

In order to determine the total resistance force of a beam-post type barrier system, several possible failure modes as in Fig. 6 should be considered. From the crash tests performed failure modes similar to those have been observed. All possible failure modes in the figure will be considered in this study.
Fig. 6
Possible Failure Modes for Beam-post Barrier
KOSHAM_17_02_133_fig_6.gif
The structure is assumed to form plastic hinges as the lateral load to the barrier increase till the point where the barrier becomes unstable due to the development of many plastic hinges, which is called mechanism.
Once a mechanism has formed, the barrier continues to deform without an increase in load. Ultimate moment capacity Mp of the W-beam (SS400) is the plastic section modulus times the yield stress of the steel used. So using plastic section modulus Zp of W beam to be 35.91cm3 and yield stress of 250MPa, Mp equals to be 9.0kN-m.
In order to determine the total resistance force of a system, several failure modes shown in Fig. 6 have been considered. In the figures, Pp is the lateral force required to form a plastic hinge at the base, and ‘wl’ is the lateral force causing hinges on the beam which can be calculated by the following equation.
(4)
W=wl=8Mp/(Ll/2)
where
Mp: Plastic moment capacity of the beam
L: Post spacing on single span
l: Transverse length of distributed vehicle, impact load
So, the total resistance force of each mechanism is the sum of the lateral force W causing the hinges in the beam for each failure mechanism and the lateral force Pprequiredtoformaplastichingeatthebase.
Using the values of the Table 3 and Eq. 4, lateral load capacity are calculated and shown in Figs. 7, 8 and 9 In the figures, required impact force by Olson model and Barrier VII simulation are shown together. The figures show the total lateral load capacities of a barrier with 2.0m span length 2.5m span length and 3.0m span length respectively. In the figures, Type-1, Type-2, Type-3 represents the post base anchor type which was explained in the previous chapter. In the figures “Dynamic” represents total lateral load capacity of the barrier was obtained using Pp from drop test and “Static” represents the capacity using Pp from static test.
Fig. 7
Total Load Capacity (Span Length: 2.0 m)
KOSHAM_17_02_133_fig_7.gif
Fig. 8
Total Load Capacity (Span Length: 2.5 m)
KOSHAM_17_02_133_fig_8.gif
Fig. 9
Total Load Capacity (Span Length: 3.0 m)
KOSHAM_17_02_133_fig_9.gif
From the figures, it can be concluded that the load carrying capacity of 2.0m span length with post of base Type-3 when using the dynamic load carrying capacity has the highest value of 64.2kN which is sufficient to resist the required force Fmax of 50.9kN resulting from the Barrier VII simulation. So, the design of W beam and Post of 2.0m span with Type-3 base anchor was selected for final validation to the 2ton-20km/h impact test. For the validation to the impact, both the LS-DYNA simulation and vehicle impact test were undertaken.

4. Performance Validation by Crash Test and Crash Simulation

The performance of the barrier system designed in the previous chapter was validated by vehicle crash test and LS-DYNA simulation.
The design consists of conventional W-beam and tubular steel posts at 2.0m intervals. W-beam has the same sectional and mechanical properties as are used in the standard drawings of Korean roadside barrier. It uses SS400 structural steel and has the sectional area A of 18.8cm2. Posts are of∅101.6×4.2mm (SS400 steel). Posts are anchored to the concrete slab with Type-3 base anchor system which has 2-M20 anchors in front and 1-M20 at the back. Both the beam and posts were modelled as shell elements.
Impact condition is 2ton-20km/h. Vehicle used for the crash test was Hyundai Dynasty with slight modification in weight to 2.0ton, but in the simulation, Dodge Neon (NCAC 2000) of weight 2ton was used. It consists of 2852 Solids, 122 Beams and 267786 Shells. Despite the model difference, the same weight of the two vehicles gave satisfactory simulation results.
Fig. 10 shows the vehicle model used and Fig. 11 is the barrier model.
Fig. 10
Vehicle Model
KOSHAM_17_02_133_fig_10.gif
Fig. 11
Beam-Post Type Barrier Model
KOSHAM_17_02_133_fig_11.gif
Two posts on the concrete runway were fixed with the base anchor Type-3 and other posts were driven to soil and modeled as such. Fig. 12 shows the barrier constructed for the test. The impact point in the test and simulation is the center of the mid-span. Fig. 13 shows the vehicle and barrier responses to the impact at some important time intervals. Fig. 14 shows acceleration, velocity and deformation of the vehicle after the impact.
Fig. 12
Beam-Post Type Barrier for Test
KOSHAM_17_02_133_fig_12.gif
Fig. 13
Crash Test vs. Simulation
KOSHAM_17_02_133_fig_13.gif
Fig. 14
Acceleration, Velocity, Deformation of Vehicle (Test vs. Simulation)
KOSHAM_17_02_133_fig_14.gif
The acceleration trace from the test was obtained by sampling rate of 10kHz and CFC 180 filter, while in the simulation, 1kHz sampling rate was used to smoothen the trace and see the general shape.
From the figures, the barrier could contain the 2-ton vehicle at 20km/h head-on impact successfully with minor permanent deformation.
The deceleration trace was analyzed for the occupant safety indexes and THIV (Theoretical Head Impact Velocity) and PHD (Post-Impact Head Deceleration) were calculated. THIV from impact test and simulation were 22.4km/h and 12.5km/h respectively and PHD from impact test and simulation were 2.4g and 3.2g respectively, which are well below the safety threshold values of 33km/h and 20g used in the roadside safety design.

5. Conclusions

Lateral load carrying capacity of a tubular post varies considerably depending on whether applied load is static or dynamic. Also it depends heavily on the method how the post is fixed to the concrete slab. Both the static and dynamic tests were made for posts of ∅101.6×4.2mm (SS400 steel) connected to the concrete slab by different connection types. Type-1 uses a base plate of size 200×200×16t and 4-M20 chemical anchors, and post is pellet welded on top and bottom of the base plate. Type-2 has the same base plate and anchors as Type-1 except it has the four rib stiffener of size 40×72×8t. Type-3 has 2-M20 anchors in front and 1-M20 at the back. The lateral load carrying capacity of the post was highest in Type-3 as 43.3kN from the dynamic drop test while 27.9kN from the static test.
Using the lateral load carrying capacity of the post and W-beam, total load carrying capacity of the barrier system with 2.0m span length was calculated from mechanism analysis. The barrier system composed of W-beam and tubular post with base plate Type-3 (two M20 anchors in front and one M20 anchor at the back) gave the total load carrying capacity of 64.2kN which is larger than the required maximum force 50.9kN for 2ton-20km/h impact calculated from the Barrier VII simulation.
Impact performance for the designed barrier system was verified by impact test and simulation. Both the test and simulation gave similar values in safety indices and maximum deflection and were satisfactory.
Conclusively, for 2ton-20km/h design condition, 64.2kN design force is appropriate and the barrier system composed of W-beam and tubular post with base plate anchored by two M20 anchors in front and one M20 at the back has sufficient strength to the impact force and can be used as a typical barrier for parkades in Korea.

Acknowledgements

This research was supported by General Researcher Support Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education, Science and Technology (No.2010-0013232).

References

CEN (2003). EN 1991-2(2003): Eurocode 1: Action on structures – Part 2: Traffic loads for bridges. European Committee for Standardization.
crossref
CEN (2006). EN 1991-1-7: Eurocode 1: Actions on Structures – Part 1-7: General actions – Accidental action. European Committee for Standardization.
crossref
Daniel, Horn (2011). Design of Monopole Bases. Technical Manual 1. http://www.towernx.com.
crossref
Hallquist, J.O (2011). LS-DYNA Theoretical Manual. Livermore Software Technology Corporation, Livemore, CA.
crossref
KRTA (2005). Specification for Bridge Design. Korea Road and Transportation Association, Seoul.
crossref
KSCE (2003). Note on Specification for Bridge Design. Korea Society of Civil Engineers, Seoul.
crossref
LSTC (2011). LS-DYNA Keyword User’s Manual. Livemore Software Technology Corporation.
crossref
MOCT (2010). Parkade Regulation. Ministry of Construction and Transportation, Seoul.
crossref
MOLIT (2014) Roadside Design and Management. Ministry of Land, Infrastructure, and Transport, Seoul.
crossref
Olson, R.M, and Post, E.R (1970) Tentative Service Requirements for Bridge Rail System. National Cooperative Highway Research Program (NCHRP) Report 86, Transportation Research Board.
crossref
Powell, G.H (1973). Barrier ?: A Computer Program for Evaluation of automobile system. U.C Berkeley.
crossref
Ross, H.E, Sicking, D.L Jr, and Zimmer, R.A (1993). Recommended procedures for the Safety Performance Evaluation of Highway Features. NCHRP Report 350, Transportation Research Board. Washington DC.
crossref
TOOLS
Share :
Facebook Twitter Linked In Google+ Line it
METRICS Graph View
  • 0 Crossref
  •    
  • 3,091 View
  • 86 Download


ABOUT
ARTICLE CATEGORY

Browse all articles >

BROWSE ARTICLES
AUTHOR INFORMATION
Editorial Office
1010 New Bldg., The Korea Science Technology Center, 22 Teheran-ro 7-gil(635-4 Yeoksam-dong), Gangnam-gu, Seoul 06130, Korea
Tel: +82-2-567-6311    Fax: +82-2-567-6313    E-mail: master@kosham.or.kr                

Copyright © 2024 by The Korean Society of Hazard Mitigation.

Developed in M2PI

Close layer
prev next