Nondestructive Elastic-Wave Tests of Foundation Slab in Office Building

Krzysztof Schabowicz+ and Jerzy Hola

Institute of Building Engineering, Wrocław University of Technology, Wybrzeże Wyspiańskiego 27, 50-370 Wrocław, Poland

The paper deals with the modern nondestructive elastic-wave testing of concrete in foundation slab in a very important office building putunder the ground water level. After the construction water has appeared in lower level of the building. The reason for the tests was to find theplace where water gets for building, and also check the technical condition of the foundation slab to take a decision about the range of repairs tobe done to the foundation slab or about their strengthening. The primary test methods were: ultrasonic Tomography, Impulse Response andImpact-Echo technique. The auxiliary methods were: the electromagnetic method and the ultrasonic method. About 100m2 of foundation slabwere tested at a time. On the basis of an analysis of the results, in particular the tomography images and the distribution of the amplitude-frequency spectra of elastic waves the damage was found. The damage had the form of cracks running along the foundation slab anddelamination of the reinforcement concrete. The depths of cracks were measured, the places of delamination were found, the thickness of slabdid not conform to the design and zones of concrete containing honeycombing were identified. The results of the nondestructive tests and theanalyses were corroborated by exposures. [doi:10.2320/matertrans.I-M2011845]

(Received November 9, 2009; Accepted October 29, 2011; Published January 18, 2012)

Keywords: nondestructive tests, ultrasonic tomography, impulse response, impact-echo, concrete foundation slab

1. Introduction

Today increasingly more often buildings with reinforcedconcrete or concrete slab foundations located much below thegroundwater table are erected. As building practice showssuch foundations often do not conform to the design and haveworkmanship defects. The defects include: slab thicknesssmaller than the design one, deep cracks in the concrete,defective (delamination and porosity) areas in the crosssection of the slab. Because of too long breaks in concreting,delaminations often occur in the lower reinforcement region.Due to the lack of concrete compaction, porosity arises.All the defects may result in damage to and leakage of thefoundations, whereby the rooms located in the basementstorey cannot be normally used.

When there is a high groundwater pressure onto thefoundation it is difficult and risky to check through exposure(e.g., a core borehole) what the actual slab thickness or thedepth of a crack is. Such a method is destructive and poses adanger of damage to the waterproofing insulation. The sameapplies to the location of delaminations and porous areas. Inorder to locate such defects one would have to make manytest pits1,2) which impair the integrity of the slab. Moreover,after the tests the exposed places require repair. These areserious drawbacks.

Considering the above, nondestructive techniques, espe-cially ultrasonic tomography, the impulse-response techniqueand the impact-echo technique, can be highly useful for thetesting of concrete or reinforced concrete slab foundationsaccessible to testing only from one side.3­20) In the authors’opinion, the three state-of-the-art acoustic techniques comple-ment each other and allow one to nondestructively determinethe thickness of a structural component and the depth ofcracks as well as to locate defective areas and theirboundaries. Cases of the comprehensive use of thetechniques, except for the attempt to compare the impulse-response technique and the impulse-echo technique,6) arehard to find in the literature on the subject.7)

This paper presents such a comprehensive application ofthe techniques in situ to a structural component in a civilstructure, i.e. to a defective foundation slab in a special-purpose building. The authors hope that the test results andtheir interpretation as well as the experience gained can beuseful to other researchers in similar test situations.

2. Description of Foundation Slab

The foundation of a large special-purpose building witha basement storey, erected in Poland, was made in the formof a 0.80m thick concrete groundslab reinforced at the topand bottom with grids of steel bars. The groundslab wasfounded 3.00m below the groundwater table. The building’sexternal longitudinal bearing walls and its internal transversewalls were founded on the groundslab. The undergroundpart of the building, sunk into aquifers, was protected againstgroundwater by means of bentonite waterproofing matslaid under the groundslab and on the outer surface of theexternal walls. The slab-wall contacts were protected withspecial sealing tapes. The slab’s top surface was overlaidwith finishing layers of: 0.05m thick polystyrene foam, PVCsheeting and a 0.10m thick concrete floor with a non-sliptopping.

Soon after the building was put into service, waterappeared on the floor in one of the rooms in the 100.00and 6.00m wide basement storey. The water leaked at theconcrete floor/longitudinal walls contact. In the part of theroom in which the leakage was most intensive a test pit ofabout 0.50m2 was made in the finishing layers near one ofthe longitudinal walls. The test pit revealed water on thesurface of the foundation slab and in the polystyrene foamlayer (Fig. 1). Therefore a decision was taken to make a large(ca 20.00m2 in area) pit extending over the entire width(6.00m) and length (3.00m) of the room (Fig. 1). After thetopping, the sheeting, the polystyrene foam and the waterwere removed and the groundslab surface was quickly driedover the pit’s whole length (3.00m), a crack became visiblenear the middle of the slab span [Fig. 1(c)]. Water wasleaking out of the crack. In order to determine the length of+Corresponding author, E-mail: k.schabowicz@pwr.wroc.pl

Materials Transactions, Vol. 53, No. 2 (2012) pp. 296 to 302Special Issue on APCNDT 2009©2012 The Japanese Society for Non-Destructive Inspection

the crack, a ca 1.00m wide bandlike pit was made in the lineof the crack. It was found that the crack in the foundation slabextended for about 50.00m. Then it became clear that thewater was leaking through a long fracture which appeared inthe foundation slab.

After the visual inspection the following questions (tomention but a few) arose: what contributed to the fracturepropagating over such a considerable length; how deep wasthe fracture; besides the visible crack, were there any otherdefects contributing to cracking? In order to find answers tothe questions, it was decided to test nondestructively theunilaterally accessible defective groundslab, using state-of-

the-art acoustic techniques, i.e., ultrasonic tomography, theimpulse-response technique and the impact-echo technique.

3. Nondestructive Tests of Defective Foundation Slab

Nondestructive testing of the defective foundation slabbegan in the large pit where, as previously mentioned, theleaking of water through the visible crack was most intensive(Fig. 1). Measuring points were marked on the surface of thegroundslab in columns 1­12 and rows A­E in a grid with a0.50m side. The arrangement of the points and the locationof the crack are shown in Fig. 2.

100 m

6 m

discovered crack

large pit

test pit

3 m

bandlike pit

longitudinal external wall

longitudinal external wall(a)

(b) (c) concretetopping

foundation slab

PVCsheeting

polystyrene foam

crack infoundation slab

Fig. 1 Location and view of pits in foundation slab: (a) schema of pits made in room, (b) test pit, (c) large pit.

Fig. 2 Grid of measuring points for nondestructive tests, marked on groundslab within large pit.

Nondestructive Elastic-Wave Tests of Foundation Slab in Office Building 297

3.1 Ultrasonic tomography testsFirst an ultrasonic tomograph (Fig. 3) was used for testing.

The tomograph is useful for, among other things, determiningthe thickness of unilaterally accessible concrete elements anddetecting inhomogeneities such as air voids and highlyporous zones. Tests were conducted in four bands each0.50m wide (Fig. 2). During testing in a band the tomo-graph’s antenna was always shifted in 0.10m steps in onedirection: from column 1 to 12. The total number ofmeasuring points was about 240. The images of thegroundslab cross sections obtained from the particularmeasuring points were collected in a three-dimensionalmatrix column and three mutually intersecting cross sectionsof the groundslab (images B, C and D) were produced on thebasis of the matrix table. Very similar groundslab thicknessimages were obtained from the tests performed in theparticular bands. An exemplary image for band 2 is shownin Fig. 4. Arrows shows the shape of groundslab’s crosssection.

An analysis of the obtained images showed that:(1) the thickness of the tested foundation slab in the approx.

1.00m wide zones adjoining the longitudinal externalwalls was about 0.80m and conformed to the design,

(2) the foundation slab thickness in the approx. 4.00mwide central zone was 0.36­0.40m and it did notconform to the design,

(3) the thinning of the groundslab thickness from 0.80m to0.36­0.40m in the transition zone was made at an angleof 45°,

(4) the space under the groundslab in its central zonewas probably filled with heterogeneous material, e.g.,construction debris,

(5) there was probably an approx. 0.10m thick levellingconcrete layer at a depth of about 0.80m under thefoundation slab.

3.2 Impulse-response testsThen the impulse-response technique (Fig. 5) was used.

This technique is suitable for, among other things, fastsearching large concrete surfaces and roughly locatingdefective areas in the cross sections of structural components.An elastic wave was nondestructively generated in everysecond measuring point in the large pit by means of aproperly calibrated rubber ended hammer [Fig. 5(a)]. Simul-taneously the signal was registered by a geophone in order toacquire data [Fig. 5(b)]. The data were processed usingdedicated software and maps of average mobility, dynamicstiffness, mobility slope, mobility times mobility slope andvoids index were plotted. The maps were the basis foranalyses of the test results.

Exemplary test results, in the form of a stiffness map and amobility map, are shown in Figs. 6 and 7. According to the

(c)(b)(a)

Fig. 3 Ultrasonic tomograph: (a) measuring set, (b) measuring antenna, (c) testing procedure.

hammer

geophone amplifier

(b)(a)

Fig. 5 Impulse-response measuring set: (a) set components (b) testedfoundation slab.

(b) (c)

(a)

Fig. 4 Exemplary groundslab thickness image obtained in band 2 by means of ultrasonic tomograph: (a) image C, (b) image D,(c) image B.

K. Schabowicz and J. Hola298

figures, stiffness and mobility are not uniform over the wholetested area of the groundslab. Stiffness is understood as thecotangent of angle ¡ of the mobility curve slope in afrequency range of 0­80Hz. Mobility is an average volatilityof the dynamics vibration. Average value of this parameter ina frequency range of 100­800Hz. In the region of measuringgrid column 5 changes in both stiffness and mobility areobserved. The fact that stiffness in this region is lower than0.1m/N and a large change in mobility [from 10 to about30m/(s·N)] occurs there indicates the presence of a deepcrack in the groundslab. The stiffness map (Fig. 6) showsthat the stiffness of the groundslab is above 0.3m/N betweencolumns 1­3, below 0.2N/m between columns 3­9 andbelow 0.1m/N in the region of column 5. The decrease instiffness can be ascribed to the smaller thickness of thegroundslab in this region. The simultaneous decrease instiffness and increase in mobility in the area delimited bycolumns 3­9 and rows A­E indicate the presence of adelamination in the groundslab’s cross section.

3.3 Impact-echo testsImpact-echo tests were carried out to determine more

accurately the groundslab’s thickness previously estimatedby ultrasonic tomography and to precisely determine theboundaries of the delamination previously located by theimpulse-response technique. For this purpose an elastic wavewas excited in the measuring points shown in Fig. 2 bymeans of an exciter located in the impact-echo apparatushead (Fig. 8). Then using dedicated software and the fast

Fourier transform the signals were converted to obtain anamplitude-frequency spectrum of the registered elastic wave.

A replicable amplitude-frequency spectrum (Fig. 9) wasobtained for the zone near the longitudinal external walls(measuring grid columns 1­2 and 11­12 in Fig. 2).According to this spectrum, dominant frequency fT1

corresponding to groundslab thickness is 2.71 kHz. Usingthe relation given in Fig. 9, the thickness of the testedfoundation slab near the longitudinal external walls wascalculated to be 0.80m (the same value had been obtainedusing the ultrasonic tomography technique).

Replicable acoustic signals, whose amplitude-frequencyspectrum is shown in Fig. 10, were also registered in thecentral zone of the groundslab (measuring grid columns 3­9in Fig. 2). Using the relation given in Fig. 10 the groundslabthickness in this zone was calculated to be 0.36m, which isclose to the result obtained by ultrasonic tomography. Similarnondestructive tests were also carried out along thegroundslab length in the bandlike pit. The tests showed thatover the whole length of the room the groundslab thickness inthe central zone is 0.36m.

Replicable acoustic signals, whose amplitude-frequencyspectrum is shown in Fig. 11, were registered in the slab’sareas delimited by measuring grid columns 2­10 and rowsA­E (Fig. 2). According to Ref. 19), such signals are

1 3 5 7 9 11Columns

Row

s

[m/N]

Fig. 6 Stiffness map for tested fragment of groundslab, recorded byimpulse-response.

Row

s

1 3 5 7 9 11

Columns

[m/(s·N)]

Fig. 7 Mobility map for tested fragment of groundslab, recorded byimpulse-response.

(b)(a)

Fig. 8 Impact-echo measuring set: (a) set components, (b) testingprocedure.

2,71 kHz

11 2

0,96

T

p

f

CT

·

·=

Frequency

Fig. 9 Elastic wave amplitude-frequency spectrum registered near externalbearing wall by impact-echo apparatus.

5,84 kHz

11 2

0,96

T

p

f

CT

·

·=

Frequency

Fig. 10 Elastic wave amplitude-frequency spectrum registered in centralzone of groundslab by impact-echo apparatus.

Nondestructive Elastic-Wave Tests of Foundation Slab in Office Building 299

registered when the generated ultrasonic wave is reflectedfrom the “bottom” of the tested element and from an internaldefect. Then the signal whose dominant frequency fd isshifted relative to the frequency corresponding to the elementthickness should be analyzed. The value corresponding tofrequency fd in the place where the defect occurs is 7.81 kHz.Using the relation given in Fig. 11, it was calculated that thedefect occurs at a depth of 0.28m in the cross section of thetested foundation slab. An analysis of the results quite clearlyshows that a horizontal delamination occurs in the slab’scross section, in the area (delimited by columns 2­11 and rowA, extending to the nearest transverse wall of the building)in which the spectrum shown in Fig. 11 was obtained. Thedelamination could have formed at the lower reinforcementlevel (situated at this depth) as a result of, for example, a longbreak in the concreting of the groundslab above thereinforcement. The area where the delamination occurs andits boundaries are shown in Fig. 12.

Since water leakage from the crack was most intensive inthe area where the delamination was located it could beconcluded that the defect had contributed to the weakeningof the groundslab and to the initiation of a deep fracture.

Also an attempt was made to determine the depth of thecrack. The tests confirmed that the crack extended to a depthof approx. 0.18­0.21m. Since there was water in the crack,the actual depth must have been greater. Test core boreholesA and B (Fig. 12) drilled to a safe depth, i.e., ca 0.18m,confirmed the presence of the crack down to this depth(Fig. 13). They also showed that the concrete cover on theupper reinforcing mesh bars was too thick, amounting to0.06m. According to the design, the thickness of this covershould be 0.03m.

3.4 Conclusions emerging from test resultsAn analysis of the test results showed that:

(1) the considerable reduction (from the design 0.80 to0.36m) in groundslab thickness at the constructionstage,

(2) the disadvantageous increase (from 0.03 to 0.06m) inupper reinforcement cover thickness,

(3) the presence of the horizontal delamination in thegroundslab cross section at a depth of 0.28m,

(4) the long and deep fracture in the groundslab in themidspan zone,

7,81 kHz

d

p

f

CT

·

·=

2

0,962

Frequency

Fig. 11 Elastic wave amplitude-frequency spectrum registered by impact-echo apparatus in area where delamination occurred.

Fig. 12 Probable area where horizontal delamination is located in groundslab cross section at depth of approx. 0.28m.

Fig. 13 Core boreholes A and B confirming presence of crack extending todepth of approx. 0.18m.

K. Schabowicz and J. Hola300

can substantially reduce the load-bearing capacity of thefoundation slab and so lower the safety of the whole buildingstructure. In order to confirm or refute this conclusion, staticanalyses were carried out using the Robot Structural Analysissoftware.21) The analyses showed that the load-bearingcapacity of the foundation slab was exceeded by 47%relative to the design load capacity. For the groundslabthickness of 0.28m (in the delamination zone), the load-bearing capacity was exceeded even to a higher degree.Considering this high exceedance of the groundslab’s loadbearing capacity, it is no surprising that such an extensivecrack appeared in the foundation slab and groundwater beganto leak through it. It was probably as result of the extensivefracture that the foundation slab’s waterproof insulation wasdamaged and groundwater flooded the floor of the basementroom. Taking into consideration the safety of the buildingstructure, a decision was taken to repair the damagedfoundation slab.

4. Follow-Up Nondestructive Testing of FoundationSlab

As mentioned above, taking into consideration the safetyof the building, a decision was taken to repair the damagedfoundation slab. First the slab was to be sealed and thenstrengthened. For this purpose, all the finishing layers wereremoved from the whole foundation slab.

The fracture (over its whole length) was pressure sealedwith polyurethane resin. For this purpose, injection boreholeswere drilled at every 0.20m along the crack, alternately onboth its sides at an angle of 45° and at a distance of approx.0.20m from the crack. Injection packers were installed,polyurethane resin was injected and the holes were sealed.Also the area in which the horizontal delamination occurredwas sealed and injection boreholes were drilled there to adepth of about 0.28m at every 0.25m in a grid pattern. Thenpackers were inserted and polyurethane resin was injectedunder pressure to seal the injection holes (Fig. 14).

After the groundslab was sealed, the delamination areawas again impact-echo tested in the same measuring points

as before sealing. Replicable signals (amplitude-frequencyspectra) were registered (Fig. 15). The spectra differedsignificantly from the spectrum shown in Fig. 11. Accordingto Fig. 15, dominant frequency fT corresponding to thegroundslab thickness of 0.36 is now 5.81 kHz. This meansthat the delamination has been sealed.

The groundslab was strengthened by increasing its crosssection. This was done by concreting a 0.20m thickreinforced concrete slab on top of the foundation slab.

5. Conclusion

Nondestructive ultrasonic tomography, impulse-responseand impact-echo tests of the cracked foundation slab of abuilding founded much below the groundwater table revealedserious workmanship defects in this structural element.Ultrasonic tomography showed that the actual groundslabthickness was much smaller than the design one. This wasconfirmed by the results obtained using the impact-echotechnique, which also yielded a more accurate groundslabthickness. Using the impulse-response method an extensivearea of horizontal concrete delamination was located in thecross section of the foundation slab. This was confirmed bythe results obtained by the impact-echo technique, which alsoidentified the boundaries more precisely. Using the impulse-response and impact-echo techniques it was shown that thevisible crack was deep. The results of the nondestructivetests were used to calculate the actual load-bearing capacity

(a) (b)

(c)

Fig. 14 Room with foundation slab being sealed: (a) general view, (b) close-up of groundslab near crack, (c) close-up of injection packer.

5,81 kHz

T

p

f

CT

·

·=

2

0,96

Frequency

Fig. 15 Elastic wave amplitude-frequency spectrum registered by impact-echo apparatus after foundation slab sealing.

Nondestructive Elastic-Wave Tests of Foundation Slab in Office Building 301

of the defective groundslab and to take a decision as to waysof repairing it.

The results of the tests demonstrate that the three state-of-the-art acoustic techniques are useful for diagnosingunilaterally accessible reinforced concrete plate components.It is recommended to use the techniques jointly since as thisresearch has shown, they complement each other.

REFERENCES

1) EN 12504-1:2009, Testing concrete in structures, Part 1: (in Polish)pp. 1­11.

2) PN-EN 13791:2008, (in Polish) pp. 1­29.3) J. Jerga and M. Pokorny: Civ. Environ. Eng. 3 (2007).4) M. Rucka and K. Wilde: J. Sound Vibrat. 297 (2006) 536­550.5) S. G. Millard, D. Law, J. H. Bungey and J. Cairns: NDT Int. 3 (2001)

625­636.6) A. G. Davis, B. H. Hertlein, M. K. Lim and K. Michols: Proc. Conf. on

Nondestructive Evaluation of Bridges and Highways, (1996) pp. 88­96.

7) J. Hoła, Ł. Sadowski and K. Schabowicz: E-J. Nondestruct. Test.Ultrasonics 14 (2009) 55­62.

8) A. G. Davis: NDT Int. 36 (2003) 185­193.9) American Concrete Institute Report ACI 228.2R-98. Nondestructive

Test Methods for Evaluation of Concrete in Structures, (ACI.

Farmington Hills, Michigan, 1998) pp. 1­62.10) B. H. Hertlein and A. G. Davis: American Concrete Institute Fall

Convention (1998).11) J. Bungey, S. Millard and M. Gratham: Testing of Concrete in

Structures, (Taylor&Francis, London and New York, 2006).12) J. Hoła and K. Schabowicz: ACI Mater. J. 102 (2005) 459­464.13) B. Hoła and K. Schabowicz: Arch. Civil Mech. Eng. 8 (2008) 73­84.14) A. Garbacz: Nondestructive investigations of polymer-concrete com-

posites using stress waves® repair efficiency evaluation (in polish).(Prace naukowe. Budownictwo z. 147, Oficyna Wydawnicza Politech-niki Warszawskiej, 2007).

15) Standard Test Method for Measuring the P-Wave Speed and theThickness of Concrete Plates Using the Impact-echo Method,(American Society for Testing and Materials, 1998).

16) C. G. Petersen, A. Davis and A. Delahaza: Symposium Non-Destructive Testing in Civil Engineering, (NDT-CE, Berlin, 2003).

17) J. Hoła and K. Schabowicz: E-J. Nondestruct. Test. Ultrasonics 13(2008) 1­6.

18) R. Beutel, H. Reinhardt, C. Grosse, A. Glaubitt, M. Krause, C.Maierhofer, D. Algernon, H. Wiggenhauser and M. Schickert:J. Nondestruct. Evaluat. 27 (2008) 59­65.

19) M. Sansalone and W. B. Streett: Impact-echo: NondestructiveEvaluation of Concrete and Masonry, (Bullbrier Press, Ithaca, 1997).

20) J. Hola and K. Schabowicz: Arch. Civil Mech. Eng. 10 (2010) 5­18.21) Software: Robot Structural Analysis, Autodesk, web site http://usa.

autodesk.com/adsk/servlet/pc/index?id=11818169&siteID=123112

K. Schabowicz and J. Hola302