Bundesanstalt für Materialforschung und -prüfung

International Symposium (NDT-CE 2003)

Non-Destructive Testing in Civil Engineering 2003
Start > Contributions >Posters > Tendon Ducts: Print

Detecting Voids Inside Ducts of Bonded Steel Tendons Using Impulse Thermography

Carsten Rieck, Bernd Hillemeier
Technische Universität Berlin, Germany

Abstract

Ducts of tendons are filled with grout of cementitious materials to protect the prestressing steel against corrosion and to ensure adequate bond between concrete and prestressing steel. Poorly grouted ducts reduce the durability of prestressed concrete structures. In this paper an NDT-method is presented to detect voids in the ducts of prestressed concrete using impulse thermography.

Impulse thermography is developed and applied to detect subsurface defects as for example in concrete and masonry structures. For detection, heat is induced by either an internal or external heat source. The transient heat transfer to the surface is then observed using an infrared camera. Compared to defect-free regions, defects appear as areas of different temperatures. Also, deeper defects are observable at a later time and with reduced contrast.

To investigate applicability of impulse thermography towards detection of voids in ducts, three concrete specimens were constructed, using two different types of tendons at various depths (5 cm to 10 cm). Cylindrical, 40 cm-long voids are located inside the ducts. The specimens were heated by different internal and external heat sources.

Due to the heat of hydration of the cement lime inside the ducts, voids with a concrete cover of up to 10 cm could be detected at an early stage. After hydration of the lime, voids can be detected by heating the specimens from the surface with an infrared array. Alternatively, heat from a welding apparatus may be induced through the prestressing steel, using the electric resistance of the prestressing wire.

Preliminary investigations using the FEM-program ANSYS helped in the design of specimens as well as in the experimental work.

This project is sponsored by the Deutsche Forschungsgemeinschaft (DFG).

Introduction

Constructions, which were built in the fifties, sixties and seventies, often show poorly grouted ducts, mainly due to improper grouting technologies and missing quality management systems. Poorly grouted ducts reduce the durability of prestressed concrete structures (fig. 1).

Fig 1: Poorly grouted ducts causing tendon corrosion (photograph by the department of Massivbau, Technische Universität Berlin).

Numerous mechanisms may be responsible for formation of voids in the grout inside the tendons [1 ]: As such, the duct may be improperly vented allowing a pocket of air to be trapped (fig. 2). Further, if the level of grout decreases in the hopper, the grout pump may introduce air to the system. Also, excessive pumping velocities may cause turbulent flow thus entrapping air bubbles in the grout.

Fig 2: Development of an air void at a high point [2].

Voids are mainly found at high points [3]. Impulse thermography may provide an alternative to X-ray testing, impact echo or ultrasonic testing to detect voids inside of ducts (fig. 3).


Fig 3: Experimental set-up to detect voids in ducts with the impulse thermography.

Preliminary Investigations Using the Finite-Element-Method (FEM)

In preliminary investigations, transient heat transfer of a three-dimensional model was simulated using the FEM-program ANSYS [4]. The model is shown in fig. 4. Because of symmetry, only a quarter of the construction was modelled. A constant heat-flux of 5 kW/m2 was loaded on top of the model for 500 s.The other surfaces were adiabatic. The maximum temperature contrast of DT = 0,5 K between a node over the void on the surface of the model and a reference node on the surface was reached after 45 minutes. Further simulations concentrated on different diameters of the duct and the prestressing steel and on different concrete covers. The results of the simulations encouraged to realise experimental investigations.

Fig 4: FEM-model for the transient heat transfer simulation (prestressing steel and duct = purple , air void = red , concrete and grout = turquoise ) [4].

Concrete Specimens with Voids Inside the Ducts

To apply impulse thermography towards detecting voids in ducts, three concrete specimens were built containing two different tendon diameters (57mm and 80 mm) at various depths (5 cm, 7 cm and 10 cm). Cylindrical, 40 cm-long voids are located inside the ducts (fig. 5).

specimen number / diameter of the tendons core grid between ducts and surface
PK 1 5 / 57 mm no
PK 2 5 / 57 mm d = 6 mm, 15 cm x 15 cm
PK 3 3 / 80 mm no
Table 1:

The specimens PK 1 and PK 2 are identical, except the additional core grid with a concrete cover of 30 mm in the specimen PK 2 (fig. 5). The specimen PK 3 has three 80 mm ducts with a concrete cover of 5 cm, 7 cm and 10 cm.

Fig 5: Construction drawing for the specimens PK 1 and PK 2 [5].

Thermal Impulse Using the Heat of Hydratation of the Grout

The heat of hydration of the grout is an internal, thermal impulse during the hardening of the cement lime in the ducts. The thermal effect can be observed on the surface with an infrared camera. Voids appear as areas without heating.

Fig 6: Experimental set-up with the infrared camera and the three specimens PK 1 to PK 3 (The numbers at the top are the numbers of thermocouples inside the tendons.)
Fig 7: Temperature distribution 30 minutes after the ducts were filled with cement lime from the top.

The ducts were filled with cement lime (w/c-ratio = 0,4) from the top. The picture of the temperature distribution shows the filled ducts (fig. 7). All ducts could be detected. The heating on the surface could be observed very early (during the filling with the lime). Thermocouples inside the tendons show the early development of the temperature (fig. 8).

Fig 8: Temperature development inside the tendons of PK 3 while the cement lime hardens, measured with thermocouples [5].

Thermal Impulse Using an Infrared Array

After the hardening of the grout, the specimens were heated up from the surface with an infrared array within a time interval of 15 minutes (fig. 9). An infrared camera documented the temperature distribution during the cooling down process.

The voids in the tendons appeared as warmer areas, because of the heat accumulation in the concrete layer between the void an the surface of the specimen. The voids with a concrete cover of 5 cm could be detected very good and with 7 cm fairly good (fig. 10: line L01 and L02). The tendons with a concrete cover of 10 cm were too deep to be detected using the infrared array (fig.10: line L03). The additional core grid in the specimen PK 2 had a negative effect on the detection of the voids.

Fig 9: Experimental set-up with the infrared array.
Fig 10: Temperature distribution on the surface of the specimen PK 3 with 3 tendons of diameter 80 mm in depths of 5 cm, 7 cm and 10 cm (lines L01 to L03), 30 minutes after switching off the heating source.

Thermal Impulse by the Electric Resistance of the Prestressing Wire

The prestressing steel was heated up by a welding apparatus with electric currents between 50 A and 100 A, using the electric resistance of the prestressing wire. The voids inside the ducts can be detected as an area of lower temperatures (fig. 12). The temperature inside the ducts, measured with thermocouples, reached maximum temperatures of 25 °C.

Fig 11: Experimental set-up with the electrodes connected to the prestressing wire.
Fig 12: Temperature distribution on the specimen PK 1. The tendon at the top with a diameter of 57 mm (concrete cover = 5 cm) was heated for 40 minutes with an electric current of 70 A.

Conclusions and Perspective

All voids in the ducts could be detected using the heat of hydration of the cement lime inside the ducts. Even voids in tendons with a concrete cover of 10 cm could be detected at an early stage. The impulse thermography can be used as a non-destructive testing method during the grouting of the ducts.

Heated up by an infrared array or by the electric resistance of the prestressing wire, the 40 cm-long voids can be detected in depths of 5 cm and 7 cm. Tendons with concrete covers of 10cm are too deep to be tested with the impulse thermography at present. An additional core grid between the surface and the tendons had a negative effect on the detection of the voids, especially when internal heat sources are used.

Further investigations concentrate on the optimisation of the electric contacts, the use of lower electric currents over a longer time and the influence of untensioned reinforcement between the surface and the tendons.

References

  1. Plotkin, Steven et al: Grouting of bridge post-tensioning tendons: Training Manual - Department of Transportation. State of Florida, 2002
  2. Knieß, Hans-Gerhard: Verfahren zur Untersuchung von Spanngliedern. Mitteilungsblatt des BAW, Nr. 58, 1986
  3. Matt, P.: Qualitätsgesicherte und überwachbare Spannsysteme im Brückenbau. Forschungsbericht Nr. 192, VSS, Zürich, 1990
  4. Borggreve, Marc: Struktur- und Feuchteuntersuchungen mit der Impuls-Thermografie. Vertieferarbeit am Fachgebiet Baustoffe und Baustoffprüfung an der TU Berlin, 2002
  5. Köhler, Björn: Die Ortung unverpreßter Bereiche in den Hüllrohren von Spanngliedern mit der Impuls-Thermografie. Vertieferarbeit am Fachgebiet Baustoffe und Baustoffprüfung an der TU Berlin, 2003
STARTPublisher: DGfZPPrograming: NDT.net