International Symposium (NDT-CE 2003)Non-Destructive Testing in Civil Engineering 2003
|Start > Contributions >Lectures > Structures 3:|
Evaluation of Concrete Structures by Advanced Nondestructive Test Methods -Impact Echo Test, Impulse Response Test and Radar SurveyYong Hao, Zheng and Kee Ee, Ng SETSCO Services Pte Ltd, Singapore
John Wei, Ong SETSCO Services (M) Sdn Bhd, Malaysia
The authors have frequently involved in the investigation and evaluation of various concrete structures for the past few years in Singapore and Malaysia. Advanced nondestructive test methods including impact echo test, impulse response test and radar survey have been used. In this article, the authors briefly introduce the applications of these methods. Investigation approach and analysis used for different cases are discussed. Test influencing factors are also highlighted. Two case studies on the evaluation of concrete quality of a bridge deck and detection of voids in post tension tendon ducts in pre-stressed slab are introduced here.
1.0 Advanced Nondestructive Test Methods
The use of impact echo, impulse response and radar as diagnostic tools for evaluation of concrete is relatively very new in Singapore and Malaysia. Such methods require skilled operators and experienced interpretation, which is very much lacking in this region. As a result, the awareness of the techniques and their applications have been limited. More works would need to be done to increase the use and appreciation of such techniques, which currently are considered new and advanced here. Some of the main attractions of these techniques are their nondestructive nature and that only one accessible surface is required for the tests. Through advanced computer softwares which facilitate detailed analysis of signals, these techniques are finding their way to users who may not be familiar with the technology themselves. Whilst any over simplification of the analysis tend to increase the application of the tools, it may well lead to misuse or misinterpretation of the methods. Nevertheless, there is an increasing awareness of such tests in the region, especially among the NDT specialists, partly also due to published literature. One of them is the ACI 228.2R.
The authors have used these methods for various investigations and evaluations of concrete structures for the past few years in Singapore and Malaysia. These include: determination of wall or slab thickness and reinforcement profile; evaluation of the concrete integrity of plate-like structures, i.e. slab, walls, bridge deck etc; investigation of water seepage problem through basement and water tank walls; detection of voids in post tension tendon ducts of pre-stressed slab, beams etc.
1.1 Impact Echo Test
The applications of the impact echo technique include: determining both the thickness and flaws in plate-like structure members, such as slabs, walls; detecting flaws in beams, columns and hollow cylindrical structural members; assessing the quality of bond in overlays; and detecting void in post tension tendon duct etc.
One of its main advantages is that it can detect flaw locations as well as the flaw depth. The popularity of this technique has been augmented by well published comprehensive research done by the developers of this technique. The resulting guidebook  has facilitated its use commercially. However, an experienced operator is still required.
1.2 Impulse Response Test
Typical applications of impulse response test are for assessing the condition of large concrete structure members such as floor slabs, pavements, bridge decks, walls, tanks, etc. The impulse response test provide a rapid approach in finding defect areas, such as poor concrete consolidation, poor ground slab support and voiding, delamination caused by steel corrosion.
Robustness, fast output and good repeatability of test results are the main advantages of this test method. However, presently there are not many literatures on this method for applications on superstructure concrete members. Similarly, despite the wide potential of this test method in engineering application, only a few case studies are published by its developer.
1.3 Radar Survey
Radar survey has been successfully used to locate reinforcing bars and post-tensioning tendons, to estimate the thickness of slab, wall, pavement and more recently to locate and identify concrete anomalies and deterioration (i.e. moisture variations, delamination, honeycombing or fractures).
The most attractive features of this technique are its ability to scan large area under investigation in a short time; its high sensitivity to subsurface moisture and embedded metal; its capability to detect both metallic and non-metallic objects.
2.0 Investigation Approach Used for Different Cases
2.1 Determination of Wall or Slab Thickness and Reinforcement Profile
For thickness measurement of a plate-like structure (wall or slab), the following equation is recommended for computation [2), (3]: -
In practice, if the P-wave speed, Cp, could be measured at a known thickness area (sound concrete area), it is more convenient and accurate.
2.2 Evaluation of Concrete Integrity for Plate-like Structures
Generally, if the affected area to be tested was large, impulse response test was the first method used to locate the relevant weak zones for further analysis. The impulse response test was typically carried out at a grid pattern of 0.5m spacing or less. The measured parameter of the impulse response test, i.e. the average mobility could be plotted out onto the layout of grid intervals. Mobility is defined as the surface velocity divided by the applied force. The average mobility is defined as mean value of mobility over the frequency range, i.e. 1 Hz to 1 kHz, which is correlated with the thickness and elastic properties of test member. Poor consolidation and honeycombing structure causes a rise in the average mobility . Since the average mobility is a function of elastic properties of the test members, higher mobility might also indicate lower concrete strength.
Subsequently, impact echo test was carried out to assess the concrete condition at the selected weaker zones for verification. The impact echo test was typically conducted on a smaller grid pattern, i.e. 200mm. Determination of honeycombing or void was based on the analysis of the distribution of amplitude versus frequency spectrum. A typical test spectrum obtained from impact echo test from sound concrete exhibits single large amplitude which corresponds to the P-wave thickness frequency (domain frequency). However, in case of concrete with sub-surface anomalies, similar amplitude will appear at a lower frequency than that of the domain frequency (often accompanied with some amplitude at higher frequencies). Impact echo test is a point testing method that provides test data within a small area under the test.
A case study for evaluation of concrete quality for a bridge deck is introduced in this article.
2.3 Investigation of Voids in Post Tension Tendon Ducts in Slab and Beams
Radar survey was used initially to locate the tendon duct profile with high degree of accuracy. This was followed by the impact echo test along the tendon ducts at typically 0.5m interval. Findings of the impact echo test depend on the analysis of the recorded frequency spectra. A sound concrete slab element exhibits single large amplitude (domain frequency) that corresponds to the thickness frequency. Similarly, for a fully grouted tendon duct, the domain frequency is at slightly lower thickness frequency and a second smaller amplitude peak due to the presence of tendon strands may be observed. Void in tendon duct is normally indicated by the presence of the peak amplitudes at a higher frequency range, accompanied by a shifted domain frequency. However, if the duct is partially grouted or the duct is relatively deep as compared to its dimension, the signal may not be distinct or large in amplitude as those produced by completely empty ducts.
A case study for investigation of voids in post tension tendon ducts of pre-stressed slab is introduced in this article.
2.4 Correlation with Intrusive Testing
Physical investigation could be in the form of coring and extraction of core sample, hacking away a small portion of the concrete, drilling hole and conducting fibrescope inspection, etc. Concrete core samples are preferably to be extracted from locations where NDT results indicate the presence of anomalies in the concrete members. These cores can be subjected to further tests such as visual and microscopic examination and compressive strength test to obtain the physical condition and properties.
It is also a good practice to carry out early verification during the course of investigation to assess the suitability and reliability of the selected NDT methods before embarking on full scale investigation with these tools.
3.0 Test Influencing Factors
Information regarding the structural members to be investigated is necessary and will facilitate the testing and interpretation of the test results. The following information should be gathered:-
4.0 Case Studies
Case1: Evaluation of Concrete Quality of a Bridge Deck
Impulse response test, impact echo test, extraction of concrete core samples and laboratory tests inclusive of compressive strength test, chemical composition analysis and petrography examination were proposed to evaluate the concrete quality in the bridge deck. A "good" area with no reported damage was also selected for testing for comparison purpose. The test area on the bridge deck was near to a kerb side and could be treated as a cantilever slab with thickness varying from 225mm to 900mm. The tendon ducts were evenly spaced at 800mm (centre to centre) running perpendicular to the kerb side.
Impulse response test was first conducted on the bridge deck surface to locate the anomalies for further investigation. In this case, the impulse response test was performed on a grid with intervals of 800mm (parallel to the kerb side and between tendons) and 300mm (perpendicular to the kerb side). Analysis was based on the review of the average mobility and the mobility slope for each test point. Areas with higher average mobility and mobility slope were selected. Figure 3 presents a plot of impulse response test on the affected area on the bridge deck segment (Mobility x Mobility Slope plot).
The average mobility values from the affected area were much higher than that from the "good" area. This indicated that the quality of concrete in the suspected area was not as good as that of the "good" area. The compressive strength test results from extracted core samples (as reported below) confirmed the findings.
Impact echo test was conducted at the same grid points for verification purpose. The extraction of concrete core samples was based on the superimposed findings of both impulse response and impact echo tests.
At one of the test points (near the tendon stressing block), extracted core sample indicated the presence of large void at a depth of about 30mm from the deck surface. The average mobility at this test point was significantly higher than other test points. At another test point, core sample showed honeycombing beneath the rebar layer (see Figure 4) and it was not suitable for compressive strength test. Similarly, impact echo test at this point indicated some anomalies. Figure 5 presents the impact echo test signal at this point.
Based on the concrete core strength results, the average estimated in-situ cube strength (EICS) ranged from 14.5N/mm2 to 30.0N/mm2 with an average of 23.5N/mm2 for the "affected" area while the average EICS for the "good" area was 52.0 N/mm2 ranging from 47.5N/mm2 to 56.0N/mm2.
Based on the findings of impulse response and impact echo tests, the distribution of sub-surface anomalies (voids, delamination and honeycombing) in concrete was generally localised and limited to two test points. Intrusive investigation, such as coring of core sample for visual examination confirmed both NDT findings. In addition to the NDT findings, extensive core compressive strength test revealed the in-situ concrete strength of the deck concrete was generally lower than the required concrete grade.
Case2: Detection of Voids in Post Tension Tendon Ducts of Pre-stressed Slab
In an industrial building construction, it was reported that grouting of the tendon duct was not carried out in accordance with required procedure. As a result, it was suspected that voids could be present in the tendon ducts.
The building was a 10-storey RC structure with post tension pre-stressed concrete slab and beams. The thickness of the pre-stressed slab was 270mm with one way evenly spaced tendon ducts and supported by pre-stressed beams. The spacing of the tendon ducts was about 700mm (centre to centre). The size of the oval shape tendon ducts was about 70mm (width) by 20mm (height) with four 13mm diameter tendons inside. The depth of the tendon ducts (from slab surface) varied from 20mm near supported beams to 220mm at mid span. Figure 6 showed a typical tendon duct profile.
A preliminary inspection using fibrescope was conducted to inspect the presence of voids in the tendon ducts via some existing grouting outlet hoses. The inspection revealed the presence of voids or partially filled grout in a few tendon ducts. These tendon ducts were selected for further detailed investigation. Figure 7 shows the internal view of one tendon duct. Impact echo test was then performed on the slab surface at an interval of 0.5m along the selected tendon ducts.
The thickness of the slab was 270mm. The recorded domain frequency was 6.3 kHz. Based on equation (1), the P-wave speed for the concrete was 3544 m/s.
The expected frequency of P-wave reflections from the presence of void in the tendon duct can be computed from the following equation :
Where, d - the depth of the tendon duct from the surface;
Cp - the P-wave speed;
In this investigation, one tendon duct was noted to be badly affected by poor grouting work. Figure 8 showed the recorded impact echo test signal at test point 7, located 3200mm away from the outlet hose (see Figure 6 for the tendon duct profile and test points). The depth of tendon duct at this point was 140mm. Expected frequency from the void tendon duct was computed to be 12.2 kHz (using equation (2)). In the recorded spectrum, the domain frequency was shifted down from 6.3 kHz to 5.4 kHz and a significant peak was observed at 12.2 kHz indicating the presence of void.
For quality control, impact echo test was also performed on some randomly selected tendon ducts with no reported defects.
Some of the more advanced NDT techniques such as the impact echo, impulse response and radar have been in increasing use in Singapore and Malaysia. So far, these techniques have been successfully employed in investigating concrete problems despite their limited application here. This is contributed in part to the user friendliness of the new equipment and in others, careful use with skilled interpretation. However, verification tests such as intrusive testing are almost always inevitable. In addition, a combination of techniques will often optimise the findings and analysis. This will eventually go a long way in raising the confidence level of these relatively new techniques in this region. Nevertheless, efforts to promote the awareness and appreciation need to be stepped up to further spur the acceptance of these efficacious tools. Notwithstanding however, development of skilled and expert use and interpretation of the tests should receive no less emphasis.