Effect of Wind-Induced Unsteady Vortex Shedding, Diurnal Temperature Changes, and Transit Conditions on Truss Structures Supporting Large Highway Signs

Project Details











Iowa Department of Transportation
Iowa Highway Research Board

Principal Investigator
M. Asghar Bhatti
Co-Principal Investigator
Brent Phares

Bridge Research Engineer, BEC

About the research

Highway sign structures are called upon to support a variety of signs including large dynamic message signs (DMS) to better manage traffic flow by providing accurate and timely information to drivers. Overhead truss structures are employed to support DMS cabinets. DMS cabinets are much heavier and have different depths and aspect ratios than typical highway signs. The current American Association of State Highway and Transportation Officials (AASHTO) LRFD Specifications for Structural Supports for Highway Signs, Luminaries and Traffic Signals (AASHTO 2015), which is the main document used for design of sign support structures by state DOTs in the US, do not give clear guidance for estimating wind loads in these situations. This increases the uncertainty in estimating stresses induced in the members of the truss structure supporting the DMS cabinet. Having detailed understanding of stresses caused during the service life of the trusses supporting DMS cabinets is crucial for their safe and economic design. In recent years, there is increasing evidence that the truss structures supporting a variety of large and heavy signs are subjected to much more complex loading than those typically accounted for in the codified design procedures. Consequently, some of these structures have required frequent inspections, retrofitting, and even premature replacement. In order to reliably predict the behavior of these structures, and to design them properly, detailed knowledge of the wind forces is obviously necessary. The first objective of this study is to accurately estimate unsteady wind loads acting on the DMS cabinets and other traffic signs and on the members of the truss structures supporting these signs. The cyclic oscillations of the total wind load associated with vortex shedding behind the signs may be a main contributor to premature fatigue failure. This is because these cyclic oscillations that occur even under steady incoming wind conditions can create a resonance condition.

Besides wind loading, the highway sign structures may be subjected to fatigue induced by stresses caused during the transport of trusses to the site and those caused by large diurnal temperature variations. Thus the second objective is to investigate possible fatigue failure due to vibrations during transportation from fabricator to the site where the truss and DMS cabinet will be deployed. The third objective is to investigate diurnal temperature effects on the fatigue life of structures.

The study was divided into two parts. The computational fluid dynamics (CFD) study, related to the first objective, was conducted by the University of Iowa. The truss monitoring during its transport and monitoring of diurnal temperature effects related to second and third objectives were conducted by Iowa State University. In Part I detailed CFD simulations were conducted to determine the air-induced mean (time-averaged) wind forces on the DMS cabinets and normal traffic signs of different configurations of interest to the Iowa DOT. The time-accurate simulations resolve the large scale turbulent eddies in the wake of the sign and take into account the unsteady wind loads associated with vortex shedding behind the sign. Based on this information, the mean drag coefficients for the DMS cabinets and other traffic signs were estimated. The CFD simulations also provided the time series of the instantaneous drag coefficient, based on which the main variables required to perform a structural fatigue analysis of the support structure can be estimated. A significant finding of this study was that AASHTO 2015 underestimates the wind drag coefficient for signs by as much as 25%. At the same time, the CFD results show that the Minimum Design Loads for Buildings and Other Structures (ASCE/SEI 7-10) recommendations for the design of aluminum sign structures is too conservative. The other main contribution of Part I was to propose a relatively simple procedure to estimate drag forces on the members of the support structure (e.g., truss). The current procedures to estimate wind loads on the members of the supporting structures are based on many simplifying assumptions and are not straightforward to apply for practical cases (e.g., as described in NCHRP Project 17-10(2), various AASHTO and ASCE specifications, and design manuals used by state DOTs). The proposed procedure is much simpler and less confusing than current procedures used by the Iowa DOT. In Part II, a detailed vulnerability assessment of sign support structures during transportation was conducted. To investigate the possibility and extent of damage during transportation, a detailed experimental and numerical study was conducted. One span of an overhead DMS-support truss was instrumented with strain gauges to measure the stress/strain induced during transportation. A numerical model was developed to quantitatively characterize vibration induced by the road profile. Several types of road roughness profiles were considered. Besides the data collected from the field, a detailed finite element model of the complete structure was created to obtain an in-depth understanding of the potential modes of damage and failure. A main finding of the fatigue analysis conducted for the truss structure was that transportation over a few hours can cause fatigue damage similar to up to months of in service loading. A fatigue analysis based on real time field monitoring of DMS support structures under long-term environmental conditions was also conducted. Beside the experimental study, a detailed finite element model was developed to investigate the fatigue life of the most critical parts of structure. Based on both experimental and numerical study, it was found that the diurnal temperature variations do not have a major effect on the truss structures used to support highway signs in Iowa.