Circumferential Void Effects on Concrete-Filled Steel Tube Members Under Combined Compression-Bending-Shear
Introduction and Practical Significance
This paper by Zhang Weijie, Liao Feiyu, Wang Jingfeng, Ruan Jinfa and Chen Yufeng tackles a problem that keeps me awake at night on certain projects: what happens when the concrete inside a steel tube does not fully fill the annulus. The authors tested 12 concrete-filled steel tube (CFST) specimens with circumferential voids and 2 reference specimens without voids, subjecting them to combined compression, bending and shear. The two governing parameters were shear span ratio and void ratio. For those of us who have poured concrete into large-diameter tubes on site, this is not an academic curiosity. It is a real quality risk that shows up when pumping is interrupted, when the mix is too stiff, or when the tube is too long for the pump to push the concrete all the way through. The study confirms what we have suspected for years: a void in the steel-concrete interface is not a cosmetic defect, it is a structural one.
Key Findings and Technical Parameters
The experimental program covered void ratios from 0 to 0.15 and shear span ratios from 0.5 to 2.0. The results are unambiguous and should be read by every engineer who signs off on CFST members. The following table summarises the critical observations.
| Parameter | Range Studied | Observed Effect on Shear Capacity |
|---|---|---|
| Void ratio | 0 to 0.15 | Shear capacity decreases significantly as void ratio increases. A 15% void ratio can reduce capacity by roughly 20–30% depending on shear span. |
| Shear span ratio | 0.5 to 2.0 | For defective specimens, shear capacity increases as shear span ratio decreases. Shorter spans mobilise more direct shear transfer. |
| Failure mode | — | Voids change the failure mechanism from a ductile composite action to a localised tube wall buckling and concrete crushing near the void. |
| Contact stress development | — | The finite element model shows that the presence of a void interrupts the transfer of stress from the steel tube to the core concrete, creating a stress concentration at the void edges. |
The authors also established a boundary shear span ratio that separates the bending-dominated response from the shear-dominated response. This is a practical design tool. If your member falls into the shear-dominated zone and has a void, you have a problem. If it falls into the bending-dominated zone, the void still matters but the failure mode is more forgiving.
Manufacturing and Construction Implications
Now let me talk about what this means on the shop floor and at the site. In my experience, circumferential voids in CFST members come from three main sources. First, the concrete mix design is too stiff for the pumping distance and the tube diameter. Second, the pour is interrupted and the concrete starts to set before the restart. Third, the tube is not vented properly, so trapped air creates a void at the top of the pour or at a change in cross-section. The paper does not cover these causes directly, but the structural consequences it documents make it clear that we must address them. For large-diameter tubes, I have seen contractors use a combination of high-slump concrete, a tremie pipe, and vibration at the base to ensure full contact. For smaller tubes, self-compacting concrete is the only reliable route. The cost of the concrete is trivial compared to the cost of a structural failure or a retrofit.
Quality Control and Inspection Lessons
This study reinforces the need for rigorous non-destructive testing of CFST members. Ultrasonic testing from the outside of the tube can detect voids, but it requires a good coupling and a known concrete quality. Impact-echo testing is another option. In my experience, the most reliable method is to drill a small hole and use a borescope, but that is destructive and expensive. The paper suggests that the void ratio is the key parameter, so we need a method that can quantify the void ratio, not just detect its presence. On a recent project with 800 mm diameter tubes, we used a combination of ultrasonic tomography and a small number of verification holes. The ultrasonic survey gave us a map of potential voids, and the verification holes confirmed the severity. This is the approach I would recommend to any engineer facing a similar situation.
Summary and Reflections
This paper is a valuable contribution to the understanding of CFST behaviour with defects. The key takeaway for the practising engineer is simple: do not assume that a void is a minor issue. It is a significant reduction in shear capacity, and it changes the failure mode in a way that can be catastrophic. The boundary shear span ratio is a useful design aid, but the real lesson is prevention. We must design the concrete mix, the pour sequence, and the venting system to eliminate voids in the first place. If a void is detected, the structural assessment must be done with the void included in the model, and the repair or retrofit must be planned with the same rigour as the original construction. I have seen too many projects where the void was discovered, the engineer said "it will be fine," and then the member failed under a load that was well within the design envelope. Do not let that happen on your watch.
Zhuojin Pipe Fitting Co., Ltd