ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Long-Span CFST Arch Bridge Vehicle-Bridge Coupling Vibration - A Practical Review

Overview and Core Findings

This paper by Yao Dunrong and Deng Nianchun tackles a problem I have lived with on every large bridge project I have supplied pipe for: how does a loaded truck crossing a long-span concrete-filled steel tube (CFST) arch bridge actually excite the structure, and when does that excitation become dangerous. The authors used ANSYS to build a finite element model of the Pingnan Third Bridge, then coupled it with a 20-degree-of-freedom KAMAZ-55111 truck model in the multi-body dynamics software Universal Mechanism (UM). The key innovation is the use of UM-generated substructure models at different cutoff frequencies to speed up computation while retaining accuracy. I have sat in too many design review meetings where the analyst proudly shows a beautiful model, and I have to ask the uncomfortable question: did you validate it against anything real. Here the authors did validate, comparing against measured data from the actual bridge, which is exactly what I would demand before signing off on any dynamic analysis that feeds into fatigue or serviceability checks.

The most important practical finding is that road surface roughness is the single dominant factor affecting both vehicle and bridge dynamic response. This is not surprising to anyone who has driven a loaded truck across a bridge deck that has been patched three times since opening. What is surprising is how much the authors quantified it: as the roughness grade improves, the dynamic response decreases dramatically. This tells me that the money spent on bridge deck pavement quality is not cosmetic - it is a structural investment. The second key finding is that vehicle spacing matters. When trucks travel in a convoy, moderate spacing actually reduces the peak displacement response of the arch, because the individual vehicle excitations partially cancel each other. This is the "shaving" and "warping" effect the authors describe. In practical terms, this means that traffic management measures - enforcing minimum headway, controlling convoy spacing - can be a legitimate structural protection measure, not just a congestion management tool.

The third finding is the one that worries me most as a pipe supplier: the short hangers (吊杆) show a significant reduction in impact coefficient as the bridge damping ratio increases, but the mid-span displacement is barely affected. This is a classic example of local versus global response. The short hangers are the most vulnerable members in a CFST arch bridge because they are stiff, short, and carry concentrated loads. I have seen short hanger failures on real bridges, and they are rarely catastrophic but always expensive. The damping ratio of the bridge is largely determined by the concrete filling quality and the steel-concrete interface condition. This is where my world meets theirs: if the concrete pumping leaves voids or poor bond at the steel interface, the damping drops, and the short hangers suffer.

Technical Parameters and Analysis Approach

The table below summarises the key parameters and findings from the study.

Parameter Value / Description Practical Significance
Bridge model Pingnan Third Bridge, CFST arch, modelled in ANSYS Full 3D model with shell elements for steel tube, solid elements for concrete
Vehicle model KAMAZ-55111 truck, 20 DOF in UM Realistic heavy vehicle with suspension dynamics
Coupling method UM substructure at cutoff frequencies 12 and 20 Speed vs. accuracy trade-off; cutoff 12 showed less frequency matching
Road roughness grades Multiple grades from smooth to rough Roughness is the dominant factor on both vehicle and bridge response
Vehicle spacing Varied convoy configurations Moderate spacing reduces peak displacement via "shaving" and "warping"
Damping ratio Varied bridge damping Short hangers benefit most; mid-span displacement insensitive
Frequency matching Vehicle and bridge natural frequencies close Two cases showed amplification; one case (cutoff 12) showed no close match

The cutoff frequency issue deserves a deeper comment. The authors found that when the substructure cutoff frequency was set to 12, the vehicle and bridge frequencies were not actually close, so the amplification was not significant. When the cutoff was higher, the frequency matching became real and the amplification was measurable. This is a subtle numerical point that has a very real physical consequence. If you run a coupled analysis with too coarse a substructure, you can miss the frequency coincidence entirely and conclude that the bridge is safe when it is not. I have been on the receiving end of this kind of error: a contractor once told me the dynamic amplification factor for a pipe rack was 1.05 based on a coarse model, and the actual field measurement showed 1.35. The difference was a missed frequency coincidence. The lesson is simple: always run a sensitivity study on the cutoff frequency, and always validate against field data if available.

Practical Implications for Pipe Manufacturing and Site Work

Now let me connect this to the shop floor and the delivery site, because that is where I have spent my career. The steel pipes for a CFST arch bridge are not ordinary line pipe. They are large-diameter, thick-walled, high-strength pipes - typically Q345, Q420, or even Q460 - and they are often spiral welded or LSAW. The manufacturing challenges are significant:

  1. Welding quality and fatigue performance. The dynamic response of the bridge means that every weld in the arch rib and hanger connections is subjected to cyclic loading. The weld toe is the critical fatigue location. I have seen too many bridges where the welder thought a smooth cap was cosmetic, when in fact it was the difference between a 50-year fatigue life and a 20-year one. The standard for weld profile is not just about penetration - it is about the transition radius at the weld toe. A 1 mm radius versus a 0.5 mm radius can change the stress concentration factor by 20%.
  2. Concrete filling and interface quality. The damping ratio of the bridge depends on the bond between the steel tube and the concrete core. If the concrete is pumped from the bottom and the air is not properly vented, you get voids at the top of the arch. These voids are not just a static strength issue - they are a dynamic damping issue. I have been on site where the contractor used a vibrating poker inside a large-diameter tube and was surprised when the concrete segregated. The lesson is that self-compacting concrete with proper admixtures is not optional for CFST arches; it is mandatory.
  3. Hanger fabrication and installation. The short hangers are the most vulnerable members. These are typically high-strength steel rods or cables with anchorages. The anchorage details are where failures start: corrosion, fretting, and fatigue cracks at the transition from the rod to the anchor. I have seen a hanger fail at the anchor because the contractor used the wrong torque on the nut and the thread was not properly lubricated. The result was a 12-week shutdown and a very expensive repair. The lesson is that hanger installation is a precision operation, not a rough fit-up.
  4. Road surface quality and traffic management. The finding that road roughness dominates the dynamic response means that the bridge deck pavement is a structural element, not just a wearing surface. I have been on projects where the pavement was laid with a tolerance of 10 mm over a 3 m straightedge, and the bridge owner was surprised when the dynamic amplification factor was higher than the design value. The lesson is that pavement smoothness should be specified and measured with the same rigour as the steel fabrication.

Key Lessons and Reflections

The study confirms what I have suspected for years: the most important factor in vehicle-bridge interaction is not the vehicle weight, not the bridge stiffness, but the road surface quality. This is a message that is often lost in the noise of structural analysis. The practical implications are clear: maintain the deck, control the traffic spacing, and avoid frequency coincidence between vehicles and the bridge. The last point is the hardest to control in practice, because you cannot control what trucks a driver chooses to operate. But you can control the bridge's natural frequency by proper design of the arch stiffness and mass distribution, and you can control the vehicle's excitation frequency by enforcing speed limits and traffic management. The combination of these measures is what I would recommend to any bridge owner.

The study also highlights the importance of damping in the short hangers. This is a detail that is often overlooked in design. The hangers are the most fatigue-prone members in a CFST arch bridge, and their response is highly sensitive to the bridge damping ratio. If the concrete filling quality is poor, the damping drops, and the hangers suffer. This is a direct connection between the quality of the concrete work and the fatigue life of the hangers. I would put this in the contract documents as a mandatory quality requirement, not a suggestion.

Finally, the study is a reminder that numerical analysis is only as good as the validation behind it. The authors validated their model against the actual bridge, which is the gold standard. Too many analyses are run without validation, and the results are taken as gospel. I have been in too many meetings where a beautiful finite element model was used to justify a design decision, and the field measurements later showed the model was off by 30%. The lesson is simple: validate, validate, validate. If you cannot validate, be conservative, and be honest about the uncertainty.

In my thirty years in this industry, I have learned that the bridge that fails is rarely the one that was under-designed. It is the one that was over-designed on paper but under-built on site. The dynamic response of a CFST arch bridge is a perfect example. The analysis is complex, but the practical measures are simple: good pavement, good concrete filling, good hanger installation, and good traffic management. Do those four things well, and the bridge will serve its design life. Do them poorly, and no amount of finite element analysis will save you. This paper is a valuable contribution because it quantifies what many of us have known from experience, and it gives the numbers to back up the intuition. I would recommend it to any engineer involved in the design, fabrication, or maintenance of CFST arch bridges, and I would particularly recommend it to the site engineers who are responsible for the concrete filling and the pavement quality. Those are the two areas where the paper's findings have the most practical impact, and where I have seen the most mistakes.