18th World Conference on Nondestructive Testing, 16-20 April 2012, Durban, South Africa

Nondestructive tests aimed at determining the thickness of the concrete

shell of a heat pipe carrying tunnel

Tomasz GORZELAŃCZYK, Jerzy HOŁA, Krzysztof SCHABOWICZ

Wrocław University of Technology, Institute of Building Engineering, Wybrzeże Wyspiańskiego 27

Wrocław, Poland, e-mail: [email protected], [email protected], [email protected]

Abstract
The paper presents the results of nondestructive tests aimed at determining the thickness of the unilaterally ac-
cessible concrete shell of a tunnel carrying heat pipes under one of the widest rivers in Poland. The approxi-
mately 1000 m long tunnel was dug manually using the caisson method. After the tunnel had been in service for
nearly 60 years its load-bearing capacity had to be tested. Considering the way in which it had been built, not
only the thickness of the tunnel shell, but also the likely thickness nonuniformity along the length of the tunnel
had to be determined. The tests were carried out using two independent nondestructive methods: ultrasonic to-
mography and the impact-echo method, which yielded similar results. The authors managed to nondestructively
determine the thickness of the tunnel‟s walls and the variation in this thickness along the tunnel length. The
nondestructive test results were destructively verified in a randomly selected place. The destructive test corrobo-
rated the results of the nondestructive tests.

Keywords: acoustic methods, ultrasonic tomograph, nondestructive testing, concrete

1. Introduction
Today nondestructive techniques are increasingly often used to diagnose various building
structures [1-3]. Owing to their noninvasiveness they are the only choice in some cases.
Acoustic methods are a major group among noninvasive techniques. The former include ul-
trasonic tomography and the impact-echo method [4, 5]. The two noninvasive methods were
employed in this research work to determine the thickness of the concrete shell of a heat pipe
carrying tunnel.
The heat pipe carrying tunnel is nearly 1000 m long and has an inside diameter of about 3.5
m. It runs under one of the widest rivers in Poland, at a depth of ten-twenty meters below the
water table. The tunnel was bored simultaneously from two sides, using a combination of
manual tunnelling and the caisson method, in the 1950s. In the tunnel there are two pipes
(each 900 mm in cross section) carrying hot water from a heat and power plant to households,
and a service platform.
The location of the tunnel and its interior are shown in respectively figs 1 and 2.

Figure 1. Location of investigated tunnel under river Figure 2. Interior of tunnel

After the tunnel had been in service for nearly 60 years the load-bearing capacity of its con-
crete shell had to be assessed. Since the tunnel construction design documentation had been
lost it became necessary to determine the thickness of the concrete shell (accessible from one
side only) as well as the likely (due to the imprecise construction work) changes in this thick-
ness along the tunnel length. Because the tunnel was located much below the groundwater
table no extensive destructive tests, consisting in drilling through the concrete shell, could be
carried out.

2. Description of tests

The ultrasonic tomograph set and the impact-echo set, shown in respectively figs 3 and 4,
were used to test the concrete shell of the tunnel.

Figure 3. Ultrasonic tomograph set Figure 4. Impact-echo set

The tomograph set includes a special multihead ultrasonic antenna and a laptop with dedi-
cated software for recording test results. The antenna with 40 independent dry-contact heads
is used to excite, receive and process ultrasonic signals. The tomograph can be used to test up
to 250 cm thick concrete elements accessible from one side only. The test results are in the
form of images taken in three mutually perpendicular directions [1,4].
The impact-echo set includes measuring heads with exciters (a set of steel balls with different
diameters) and a laptop. An elastic wave is generated in the tested element by striking its sur-
face with an exciter. A dedicated software is used to record (in amplitude-time coordinates)
the graphic image of the elastic wave propagating in the element and to convert it into an am-
plitude-time spectrum by means of the fast Fourier transform. The spectrum is subjected to
further analysis. The impact-echo method exploits the interdependence between frequency fT,
elastic wave velocity in concrete Cp and depth T at which a defect occurs, described by the
relation:

0,96 C p
fT (1)
2 T

Twenty measuring sites in the form of bands were selected in the tunnel, assuming that if
needed, their number and size could be changed.
In each measuring site a shell band 500 mm wide and 1500 mm high was investigated by
means of the ultrasonic tomograph. During the investigation the ultrasonic tomograph‟s an-
tenna was continuously moved (in 100 mm long steps) in one direction. In total there were
300 measuring sites. The images of the cross section in each position were collected in a
three-dimensional matrix table. Three mutually intersecting cross sections (images B, C and
D) of the investigated object would be obtained on the basis of the matrix table. Since the aim
of the tests was to determine the thickness of the concrete tunnel shell, only images B and C
were useful for the analysis.
Also impact-echo tests were carried out in the 300 measuring sites. In each of the sites an
elastic wave was nondestructively generated by means of the exciter mounted on the impact-
echo apparatus and the graphic amplitude image of this wave as a function of time was re-
corded.
Figure 5 shows the distribution of the test bands in the tunnel, including the names of the
cross sections (images) of the investigated object, a system of coordinates tied to the tomo-
graph‟s antenna, and photographs of the apparatus in action.

Names of cross sections:

image D

image C

image B

Figure 5. Distribution of measuring bands in tested tunnel, names of tunnel cross sections, and coordinate system
tied to tomograph antenna

3. Test results and their analysis

Exemplary nondestructive ultrasonic tomography test results (images B and D for measuring
bands 7 and 19) are shown in figures 6 and 7. In the figures‟ top right corner there is a scale
showing the correspondence between the ultrasonic dispersion level in a point of the tunnel
shell and the colour representing this level. The ultrasonic dispersion level indicates that the
physical characteristics of the media in the tested area are different. The arrows mark the
identified outer surface of the tested shell. By relating the position of the arrows to the vertical
axis one can read off the thickness of the tunnel‟s concrete shell.
The chart in figure 8 shows the tunnel shell thickness determined in the individual measuring
bands by analyzing the respective images. The minimum thickness and the maximum thick-
ness of the tunnel shell in a given band are marked black and grey, respectively. Also the av-
eraged tunnel shell thickness in a given measuring band is given in the figure.
It appears from figures 6-8 that the thickness of the concrete shell is not uniform along the
tunnel length, ranging from 380 to 425 mm depending on the measuring band number.

a) b)

Figure 6. Measuring band 7: a) image D, b) image B

a) b)

Figure 7. Measuring band 19: a) image D, b) image B

 430
Average shell thickness d in band
Tunnel shell thickness d in band [mm]

418
420
413
410
408 408 408
410 405
403 403
400
398
398
400 395 395 395
395 395
393
390
388
390

380

370

360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Measuring band no.

minimum thickness in band maximum thickness in band

Figure 8. Concrete tunnel shell thickness determined in particular measuring bands by ultrasonic tomography
method

Figures 9 and 10 show exemplary test results in the form of elastic wave amplitude-frequency
spectra recorded within measuring band 11 by the impact-echo apparatus, and the concrete
shell thickness calculated from relation (1).
The chart in figure 11 shows the tunnel shell thickness determined in the particular measuring
bands by analyzing the individual amplitude-frequency spectra. As in figure 8, black marks
the minimum tunnel shell thickness while grey marks the maximum shell thickness. Also the
averaged tunnel shell thickness in a given measuring band is given in the figure.

9,00E-10
4,43 kHz
8,00E-10
0,96 C p 0,96 3800
7,00E-10 T 411,74 mm
6,00E-10
2 fT 2 4,43
Amplitude

5,00E-10
4,00E-10
3,00E-10
2,00E-10
1,00E-10
0,00E+00
0 10 20 30 40 50 60
Frequency [kHz]

Figure 9. Elastic wave amplitude-frequency spectrum recorded in one of test points in measuring band 11 by
impact-echo apparatus
 9,00E-11
8,00E-11
7,00E-11 4,62 kHz
6,00E-11 0,96 C p 0,96 3800
Amplitude

5,00E-11
T 394,81 mm
2 fT 2 4,62
4,00E-11
3,00E-11
2,00E-11
1,00E-11
0,00E+00
0 10 20 30 40 50 60
Frequency [kHz]

Figure 10. Elastic wave amplitude-frequency spectrum recorded in one of test points in measuring band 11 by
impact-echo apparatus

430 Average shell thickness d in band

Tunnel shell thickness d in band [mm]

420
414

409 409 409

408
410 406
405 406
404 405
404
401
401 400
397
400
394
393 393
392
389
390

380

370

360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Measuring band no.

minimum thickness in band maximum thickness in band

Figure 11. Concrete tunnel shell thickness determined in particular measuring bands by impact-echo method

It appears from figures 9-11 that the concrete tunnel shell thickness is not uniform, ranging
from 387 to 421 mm along the tunnel length.
It should be noted that the tunnel shell thickness values determined in the particular measur-
ing bands by the ultrasonic tomography method do not differ significantly from the ones de-
termined by the impact-echo method.
In order to fully present the results, the averaged tunnel shell thickness values in the particular
measuring bands determined by the two test methods, i.e. ultrasonic tomography and impact
echo, are shown in fig. 12. It appears from this figure that the average concrete tunnel shell
thickness in 20 bands, calculated from the results of the tests carried out using the two meth-
ods, ranges from 389 to 415 mm. Thus it is highly likely that the design thickness of the shell
was 400 mm.
 430
Tunnel shell thickness d in band [mm]

minimum average shell thickness d min in band maximum average shell thickness d max in band
420
415,5

409,0 409,3 408,0 408,3

410 406,8
406,8
403,5 403,8
402,0
399,3 398,5 397,8
400 397,8
394,5 394,8
394,0 393,8

389,5 390,5
390

380

370

360
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Measuring band no.

average thickness in band

Figure 12. Averaged concrete tunnel shell thickness determined in particular measuring bands by ultrasonic
tomography method and impact-echo method

In order to verify the results, the concrete shell was drilled through in measuring band 7. The
tunnel shell thickness in this place was found to amount to 405 mm, which corroborated the
result obtained by the nondestructive methods.

4. Conclusion

By carrying out the nondestructive ultrasonic tomography and impact-echo tests the authors
managed to determine the thickness of the unilaterally accessible concrete shell of the heat
pipe carrying tunnel running under one of the widest rivers in Poland. It is significant that the
two test methods yielded similar shell thickness values in the particular measuring bands. An
analysis of the results showed that the thickness of the shell is not uniform along the length of
the tunnel, ranging from 388 to 415 mm. Hence one can conclude that the design thickness of
the concrete shell was probably 400 mm. The results yielded by the nondestructive methods
were corroborated by drilling through the shell in a randomly selected place.

References

1. A A Samokrutov, V N Kozlov, V G Shevaldykin, „Ultrasonic testing of concrete objects

using dry acoustic contact. Methods, instruments and possibilities‟, The 5th International
Conference “Non-Destructive Testing and Technical Diagnostics in Industry”. Mashinos-
troenie, Moscow, Russia, p.152, 2006.
2. T Piotrowski, A Garbacz, A van der Wielen, L Courard, F Nguyen, „On the relation be-
tween bond quality and impact-echo frequency spectrum‟, Int. Journal for Restoration of
Buildings and Monuments, 16 (4/5), pp 303-314, 2010.
3. K Schabowicz and J Hoła, „Nondestructive elastic-wave tests of foundation slab in office
building‟, 13th Asia-Pacific Conference on Non-Destructive Testing, Yokohama, Japan,
2009.
4. A V Bishko, A A Samokrutov, V G Shevaldykin, „Ultrasonic echo-pulse tomography of
concrete using shear waves low-frequency phased antenna arrays‟, e-Journal of Nonde-
structive Testing & Ultrasonics. vol. 13, 2008.
5. K Schabowicz, J Hoła, D Styś, „Nondestructive elastic-wave tests of concrete in founda-
tion slab‟, 10th European Conference on Nondestructive Testing, Moscow, Russia, 2010.
6. M Sansalone and W Streett, „Impact-echo: Nondestructive Evaluation of Concrete and
Masonry‟, Bullbrier Press, Ithaca, 1997.