Journal of Civil Structural Health Monitoring

[Link] (0123456789().,-volV)(0123456789().,-volV)

ORIGINAL PAPER

Evaluating the damage degree of cracking in facades using infrared

thermography
Elton Bauer1 • Patrı́cia Mota Milhomem2 • Luiz Augusto Gimenez Aidar2

Received: 29 April 2018 / Accepted: 5 June 2018

 Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
The study of facade cracking is of great importance in the investigation of pathologies and degradation, as well as in the
rehabilitation processes of buildings. Cracking is one of the major defects in facades; so assessing damage severity is
necessary to define the best maintenance and rehabilitation strategies. The infrared thermography has been increasingly
used as an inspection technique to identify and map facade defects and their severity. The use of hygrothermal simulation
to understand the dynamics of the heat flux allows to observe important references that can help in the interpretation of the
thermogram under different heating or cooling conditions. This study investigates the use of the quantitative passive
thermography to measure the damage degree of cracks in a building facade. The crack depth was evaluated by the Delta-
T values investigated in the three cracked regions. For the width, the variation of the studied temperature in a line
transversal to the crack was used. The analysis was performed at a distance of 33 m from the facade. The conclusions
corroborated the importance of the hygrothermal simulation as a pre-thermography technique, which made it possible to
limit the period of investigation focusing on the phenomena of interest. It is also observed that it is possible to obtain
quantitative values of the degree of damage by thermographic investigation. The measurement of the damage degree, in
turn, brings important contribution to the application of the inspection technique.

Keywords Cracks  Facade  Thermography  Damage  Pathology  Maintenance

1 Introduction rainwater to penetrate [1–3]. The cracked facades submit-

ted to driving rain, develop conditions for other defects
Cracking is a defect of great importance and high incidence such as detachment and efflorescence as well [4–6].
on building facades. It affects the performance of the Most of the models for evaluating building degradation
facade and its systems since it results in loss of integrity of and service life, consider cracking as one of the three major
the elements for allowing degradation agents such as defects regarding the degree of damage inflicted in the
facades [7–9]. Therefore, it can be stated that a high
cracking degree has a decisive influence on reducing the
& Elton Bauer service life of the facade and its systems. It is necessary to
elbauerlem@[Link] detect the cracks and the need for corrective intervention to
Patrı́cia Mota Milhomem elaborate an adequate maintenance plan so the building can
patriciamotamilhomem@[Link] reach the planned service life [10].
Luiz Augusto Gimenez Aidar The causes of facade cracking can either be constructive
luiz.augusto_aidar@[Link] or result from the action of degradation agents on the
1 building. The constructive causes are associated with
Civil and Environmental Engineering Department, Post
Graduate Program in Structures and Construction, Campus inadequacies and failures of material quality and con-
Universitário Darcy Ribeiro, University of Brası́lia, Asa struction processes. It is explained here the question of the
Norte, Brası́lia, DF 70910-900, Brazil drying shrinkage of the cementitious materials, the faulty
2
Post Graduate Program in Structures and Construction, adherence of the bonding of ceramic tiles, and also the
Campus Universitário Darcy Ribeiro, University of Brası́lia, cracking of masonry itself due to the displacements and
Asa Norte, Brası́lia, DF 70910-900, Brazil

123
 Journal of Civil Structural Health Monitoring

structural deformations of the concrete structure [11–13]. [27, 32, 33]. The disturbance to the heat flux caused by the
The cracking in the regions where the masonry meets the defects allows identifying the defects and measuring the
pillars and beams are high-frequency pathologies espe- damaged area. This disturbance results in different tem-
cially in buildings with low structural rigidity [14]. The perature patterns that are captured and identified in the
causes associated with the degradation agents (solar radi- thermograms [34–39]. Each change in the flux pattern
ation, temperature, driving rain, etc.) translate into thermo- modifies the thermogram and allows identifying the dam-
mechanical stresses on coating systems capable of gener- aged region, even when the defect is not visible on the
ating cracks and detachment. The driving rain causes surface [33, 40, 41]. Variables such as thermal conductiv-
material wetting and non-reversible expansions of the ity, moisture content, presence of structural elements
ceramic materials [15–19]. The humidification and drying influence and may limit the desired analysis [42]. Obvi-
of the facade also generate expansion and contraction ously, this technique is limited regarding depth and is used
variations, capable of damaging the cementitious materials. especially in superficial and subsurface defects. Pavón [27],
All the movements have great potential for cracking since among others [12, 25, 34], studied perturbations caused by
the materials and elements have restricted movement due occluded elements, internal fissures and different materials
to either adherence or by the attachment to the base or inserted in the masonry (wood and polymeric foams). In his
support layer [8, 20]. investigation, he identified facade anomalies in depths of
The ISO 15686-2 standard [21] defines the service life up to 5 cm with good precision.
of the building and its elements taking into account several The main contribution of thermography in the moni-
procedures to define the behavior with respect to degra- toring of facades is the ability to measure and identify
dation. Among the procedures is building inspection, which different anomalies while associating a degree of damage
has the great advantage of directly associating the damage to the element depending on the defect typology, be it tile
degree of the elements with the exposure to the environ- detachment, cracks, paint bubbles or moisture [33, 43, 44].
ment and use. In this case, the evaluation is real and there is Each defect has a different response to heat flux due to
no need to adapt models, simulations or other procedures. either position, size or depth. It is necessary to associate
Studies begin with inspection of the anomalies followed by thermographic patterns depending on the heat flux and
mapping to quantify damage and assign damage degree direction, and the relative temperature of the defect com-
[9, 22–24]. Often the inspection is visual, employing pared to the non-defect surrounding region [26, 27]. The
optical equipment capable of identifying the cracks thermographic monitoring of a crack to identify the degree
[12, 25, 26]. Although it is possible to have a precise of damage is different from what is required to evaluate a
system with very high accuracy, it is not possible to obtain detachment. Thus, the study of the evolution of the thermal
detailed information on the degree of damage, be it the perturbation with increasing surface temperature, ade-
depth or the width of the cracks in the facades. Infrared quately monitored by the thermograms, allows identifying
thermography is a quantitative technique capable of mea- patterns and criteria capable of inferring and classifying the
suring the degree of damage, provided that it is performed degree of damage regarding crack depth and width as well.
quantitatively and with adequate analysis. In this way, the Bauer et al. [29], and Pavón [27] identified in the labora-
monitoring allows to associate not only a mapping of the tory and in the field, thermogram variations that allowed
anomalies, but to consider their gravity [24, 27, 28]. classifying the cracks regarding their depth and width.
Frequently, these studies use the Delta-T, expressed in
Eq. (1), to identify the anomaly and its severity. In these
2 Theoretical background studies, the heat flux in both directions (inward and out-
ward) indicated that deeper cracks have higher Delta-T
The use of thermography as a facade-monitoring technique values (in modulus) during the heating or cooling cycles.
has three main advantages: it allows remote measurements, This differentiated evolution can be used to classify the
real-time evaluation (no processing required), and presents cracking regarding its depth [44].
the results as a thermal image (allowing to identify tem- ðDelta - TÞ ¼ Td  Tnd ; ð1Þ
perature variations in and along the elements) [29]. Ther-
mography measures the infrared radiation emitted by the where Td defect temperature (crack), Tnd temperature
target object and transforms it into temperature using data region without defect, Delta-T thermographic contrast that
characteristic of the material and inspection condition allows the identification of anomalies and their severity.
(emissivity, reflected temperature, etc.) [29–31]. Temper- The crack width must be differentiated from the damage
ature monitoring over time allows assessing the heat flux width. The crack width is defined and measured as the
inward or outward depending on whether the facade sur- opening of the crack whereas damage width is usually a
face is warmer or colder than the interior, respectively larger area showing more severe damage resulting from the

123
Journal of Civil Structural Health Monitoring

crack. In ceramic coatings, the damage width may con- equipment and sufficiently precise technique to the map-
template a strip of tile detachment that resulted from the ping of building damages.
crack itself [27, 44]. In high-contrast thermograms (high
Delta-T), damage width resulting from cracks may be
estimated by identifying the number of pixels corre- 3 Experimental development
sponding to the thermal disturbance in the thermogram.
From IFOV (instantaneous field of vision) and target dis- The field study consisted of a semi-continuous monitoring
tance, it is possible to estimate the damage width of a section of a facade with significant cracks. The pur-
[27, 38, 39]. pose of this investigation was to measure behavioral
Several difficulties arise while quantifying the damage changes in the thermograms as to define the degree of
degree of a crack by thermographic monitoring. It is damage in the cracks. Passive thermography requires a
noteworthy that the monitoring of facades is complex, due careful study on how the solar incidence occurs as a
to the different elements that compose it. The element function of the solar orientation of the studied facade,
responses to heat flux (solar incidence) and humidity which was defined by the hygrothermal simulation
variations of the elements throughout the day have a performed.
dynamic behavior that changes the thermogram’s fre- The sample consisted of a section of a masonry facade
quently. Thus, there may not be a single criterion for with ceramic tile covering of an east-oriented building in
cracking evaluation, such as a typical Delta-T value. It is Brasilia, DF, Brazil. Figure 1 shows the building as well as
more likely that the study of the dynamic behavior of the three crack regions defined for the study.
temperatures allows understanding and classifying these The first investigation performed during sample selec-
defects. In another approach, the cracking typology, whe- tion consisted of conducting a qualitative thermography to
ther linear or branched, may require differentiated studies show whether the cracks were identifiable in the thermo-
since the disturbance to the heat flux is surely different grams. This initial evaluation is shown in Fig. 2, where the
[16, 45–49]. Thus, thermographic profiles of temperature cracks and other aspects such as possible detachment
evolution are extremely important for the evaluation of the below region 1 are clearly visible.
phenomena involved.
To evaluate and monitor a facade, the heat sources are, 3.1 Monitoring and obtaining results
obviously, meteorological agents, particularly solar radia-
tion. Thus, the monitoring fits the passive thermography, The infrared camera employed was the Flir T-420 with
when controlled heating forms are not used [29, 31]. Also, - 20 to 120 C temperature range, 7.5–13.0 lm spectral
quantitative thermography is used, since the contrast range, 320 9 240-pixel resolution, equipped with a 15
functions and the temperature profiles over time can assign telescopic lens with an IFOV of 0.820 mrad. Relative
a damage degree to the crack [31]. humidity and air temperature were measured by an
Hygrothermal simulation contributes greatly to ther- EXTECH MO297. The distance measurements were per-
mographic monitoring. This tool uses numerical simula- formed using a BOSCH laser, GLM80 model.
tion, based on meteorological data and the characteristics
and properties of the materials, for quantifying the tem-
perature evolution of any element that composes the facade
[15, 49]. Therefore, the heat flux through the facade can be
quantified for the building conditions under analysis. For
example, it is possible to study the temperature gradients
arising from the heat flux along the facade, from the first
moments of heating to its complete cooling. Some studies
have associated the simulation with the thermographic
inspection as to define the best time for the inspection,
when the flux inversion occurs, and when the temperature
gradient has the best thermographic contrasts [27, 49].
This study applied quantitative passive thermography to
evaluate the damage degree of cracks on ceramic coatings
in a field application. The objective was to acquire image
over distances longer than 30 m using common thermo-
graphic equipment (with good resolution, but not of sci-
entific type), seeking to determine the contributions of this Fig. 1 Study sample showing the cracks and the three studied regions

123
 Journal of Civil Structural Health Monitoring

each pixel. Fig. 4b shows the results Td, Tnd, Delta-T, and
crack width (Lcr) in pixels. The crack limits are defined
from the pixel of maximum and minimum inclination
obtaining, then the interval relative to the crack width
[27, 38, 39].
The crack width is determined by Eq. (2).
Lcrack ¼ Lcr  IFOV  L; ð2Þ
where Lcrack is the crack width (mm), Lcr is the crack width
in the thermogram (pixels), IFOV relative to the thermo-
graphic camera lens resolution (mrad), L is the distance
between the thermographic camera and target (m).
The hygrothermal simulation was performed on the
Fig. 2 Thermogram showing the detection of studied cracks
WUFI Pro 5.3 software. This tool allows to simulate the
buildings in relation to several climatic conditions. The
The apparent reflected temperature was determined by
typical meteorological year was developed from 10 years
the corrugated aluminum foil method, according to ASTM
of data collected by the Meteorology (INMET - National
E1862. The used emissivity was 0.85 according to data
Institute of Meteorolgy - Brazil). Several definitions
collected by Bauer and Pavón [29]. After entering these
regarding the configuration of the facade were determined,
data in the thermographic camera, the equipment was
such as material data (density, porosity, specific heat, water
programmed for semi-continuous automatic acquisition of
absorption coefficient, and hygroscopic curve), cardinal
field data at 5-min intervals.
orientation, height of the building, absorptance, etc. For the
Figure 3 shows the schematics for data acquisition. The
purpose of this analysis, the software dynamically simu-
distance between the camera and the target is 33 m with
lates the temperature profiles in the different layers that
35 angle of incidence, considered adequate according to
make up the facade. The material data were those currently
Maldague [31].
used in the local buildings [15]. The simulation was per-
Each thermogram shows the same three regions simul-
formed for a period of 60 days, being studied 3 days for
taneously. The temperatures of the region (Treg), crack (Td),
analysis, at the dates of the field investigation. Care was
and non-defect region (Tnd) were determined for each
taken to select a period without rainfall, similar to the field
region of study. The methodology for obtaining these
conditions encountered. Figure 5 details the model used in
temperatures was adapted from Pavón [27] based on
the hygrothermal simulation, where the masonry is coated
Dufour [38], Lai et al. [39]. From the identification of the
with mortar, and externally on the facade are also the
crack in the thermogram, a temperature line perpendicular
ceramic tiles; also shown schematically are the points
to the crack is obtained (as shown in Fig. 4a). This line is
where the hourly temperatures were obtained along the
exported to Excel to obtain the temperature data for each
days of analysis in the simulation. Tamb is the ambient
line pixel. To eliminate some point distortions, the moving
temperature of the outside air, Tsup is the temperature of the
averages are calculated between the measured values for
exterior surface of the facade, and Tmbl is the temperature
at the interface between the brick and the rendering mortar.
The hygrothermal simulation focused on the evolution
of the surface temperature of the facade (Tsup) and tem-
perature gradient of the coating layer (Grev), according to
Eq. (3)
Grev ¼ Tsup  Tmbl ; ð3Þ
where Grev is the temperature gradient of the coating layer
(C), Tsup is the outside surface temperature of the facade
(C), Tmbl is the temperature at the masonry and coating
mortar interface (C).
Gradient analysis is important as it defines the magni-
tude and direction of the heat flux in the facade. The
identification of some defects requires a greater flux
magnitude.
Fig. 3 Schematics showing the position of the thermographic camera
during thermogram acquisition

123
Journal of Civil Structural Health Monitoring

Fig. 4 a Temperature line perpendicular to the crack. b Obtaining the temperatures and the width of the crack in pixels

4.1 Hygrothermal simulation

The thermal behavior of the facades is cyclical and is

repeated every day. The surface temperature and sun
incidence are strongly dependent and directly correlated,
e.g., the surface temperature increases with sun incidence
on the facade. The inverse situation is also clear in the
period when the sun ceases to affect the facade, and at
night. Figure 6 shows the simulation result for a period of
3 days during the field investigation dates. It is observed
that both the surface temperature and the ambient tem-
perature show a similar pattern for the days of analysis.
The lower temperature of the facade surface occurs just
before the solar incidence. It should be mentioned that the
eastern facade receives first the solar incidence on the
building, occurring the respective heating.
The incidence of solar radiation on the east facade from
about 6:00 to 11:30 a.m. promotes the heating observed.
The initially negative (night period without sun incidence)
Grev becomes positive, peaking near 3.9 C about
9:00 a.m. Although Grev drops from then on, possibly due
Fig. 5 Representative scheme of the facade used in the hygrothermal
to the more vertical inclination of sunbeams, the heat flux
simulation from outside to inside is positive throughout the period of
sun incidence on the east facade.
4 Results and discussion For comparison purposes, the average temperature data
of region 1, obtained by thermography, were plotted in
The data related to the hygrothermal simulation, the Fig. 6. A significant approximation between the field data
behavior of the studied regions and cracks were analyzed and the simulation data is observed. The thermographic
considering that the objective of this investigation was to data collected between 7:30 and 10:40 a.m. show that the
measure the degree of crack damage using thermography. facade underwent a heating process with a positive Grev
gradient. Pavón [27] states that it is already possible to
identify facade defects during the heating phase for Grev
higher than 0.6 C.

123
 Journal of Civil Structural Health Monitoring

Fig. 6 Temperature and

gradient evolution obtained
from the hygrothermal
simulation

4.2 Behavior of the regions Figure 8 shows the temperature profiles, Td and Tnd, for
each crack studied. These temperatures are close at the
To understand the global phenomena, Fig. 7 shows the beginning with an increasing difference, especially after
evolution of the average temperatures obtained from the 40 min. After this period, the oscillations already observed
thermograms for the three study regions identified in in Fig. 7 occur for both Td and Tnd. The Td temperatures for
Fig. 1. The facade clearly heats up in the first 40 min, after regions 1, 2, and 3 follow a crescent trend, allowing to
which oscillations (heating and cooling) are detected in all conclude that the thermal flux disturbance is different for
three regions simultaneously. These temperature variations each crack, and therefore, the degree of damage must be
can be explained by wind incidence, possible transient different for each case. The relationship between Td and
nebulosity, and the higher solar inclination in the final Tnd individually for each crack (Fig. 8d) shows a linear
investigation stage. Region 1 is evidently warmer than the trend, i.e., the two temperatures (Td and Tnd) respond in the
others, but oscillations occur simultaneously in all regions. same way to the increase of heat. However, each crack

Fig. 7 Evolution of the average

temperatures of the three study
regions

123
Journal of Civil Structural Health Monitoring

Fig. 8 Evolution of Td and Tnd for the three studied cracks

presents a specific correlation with the heat flux showing a clear tendency. Establishing a behavioral profile
disturbance. over time is necessary to reach a conclusion regarding the
It is noteworthy that the visualization and identification degree of damage, instead of treating an isolated value as a
of cracks at the end of the investigation is more complex reference. In laboratory studies, Bauer et al. [33] suggested
since the images become less sharp possibly because all a direct correlation between crack depths and Delta-T val-
temperatures are higher, making it difficult to contrast the ues; the higher the crack depth, the higher the Delta-
crack with other regions. Figure 6 also indicates a decrease T. Thus, it is possible to classify the cracks in relation to
of the Grev gradient, implying a lower heat flux since the their depth. It should be noted that Delta-T fluctuations
whole system is warmer as well. after 40 min follow the same temperature trends observed
Figure 9 shows the thermograms obtained at four dif- in Figs. 7 and 8.
ferent times: 0, 20, 40, and 150 min. It is observed that the Table 1 shows Delta-T, Tnd, and Td values for the
crack is more prominent at 20 and 40 min (Fig. 9a, b) of specific analysis times. The behavior of the cracks is
the heating cycle. At 150 min (Fig. 9d), it becomes already maintained throughout the investigation period. The Delta-
more difficult to visualize the cracks clearly, quite possibly T of each crack evolves during the first 40 min following
due to the higher temperatures of the whole system. the standard observed previously; Delta-T of crack 1 is
Figure 10 shows that Delta-T increases markedly in higher than that of crack 3, which is higher than that of
cracks 1 and 3 in the first 40 min, an increase that is also crack 2. It is also evidenced that Td and Tnd temperatures
observed for crack 2, but much less intensely. Throughout are relatively close and do not vary more than 1 C
the period, the Delta-T was clearly higher in crack 1 between cracks initially. Also, the maximum temperature
compared to crack 3, which was higher than crack 2, values indicate that Td increased more than Tnd, showing a

123
 Journal of Civil Structural Health Monitoring

Fig. 9 Thermograms showing different conditions throughout the inspection. a 0, b 20, c 40, d 150 min

gain of about 9 C compared to near 7.5 C for Tnd; All the studied cracks had the same solar incidence at
obviously the heat flux disturbance influences each situa- the same time, and thus the profiles over time in Fig. 8 are
tion distinctly. similar. Crack 1 shows higher temperatures at the maxi-
The temperature results obtained for the lines perpen- mum point of the profiles (Fig. 11), which was already
dicular to the crack (Fig. 4) are presented in Fig. 11 for the evident in both the Delta-T (Fig. 10) and the monitored
three cracks. The initial thermographic reading shows a temperature evolution (Fig. 8). As the heating process is
higher temperature in the pixels relative to the crack, which theoretically similar, the behavior is the result of thermal
is evidenced by the Delta-T values different than zero at flux disturbances, consequently, of the damage extent. A
that moment (Fig. 10). In the first 40 min of the heating physical explanation for the phenomenon is that at the
cycle, the strong heating translates into increasing tem- beginning of the monitoring, with the increasing surface
perature profiles (perpendicular line) (Fig. 11). Since temperature, the damaged region is colder, generating a
temperatures fluctuated from 40 min (Fig. 7), it was higher contrast Delta-T. As the heating progresses, the
observed that lower temperatures were measured at 55 and Delta-T increases until it stabilizes at a given moment, then
75 min compared to 40 min. As commented, these varia- oscillates according to the incident radiation conditions.
tions are repeated in the three cracks and reflect tempera- Thus, under higher temperature conditions in the system,
ture oscillations by perturbations in the field. It is also the profiles are flatter; (Fig. 9) making it difficult to mea-
shown that the profiles for the crack 3 (Fig. 11c) do not sure both Delta-T and damage width. This behavior causes
present sharp peaks for the readings between 15 and a lack of sharpness in the thermogram, which has also been
40 min, which generate uncertainties in the definition of observed in laboratory research investigating ceramic
the width of the cracks. detachments when it was recommended not to conduct the

123
Journal of Civil Structural Health Monitoring

Fig. 10 Delta-T over time for

the three cracks

Table 1 Delta-T, Tnd, and Td for the studied cracks (C) Regarding damage width, Table 2 shows the results
Delta-T 0 Delta-T 40 Delta-T max
observed in the first 40 min of the period. During this
period, there was clear heating of the study regions.
Crack 1 1.11 1.92 2.18 The crack width (Lcr) was determined in the thermogram
Crack 2 0.21 0.64 0.72 according to the procedure described in Fig. 4 by obtaining
Crack 3 0.64 1.67 1.86 the number of pixels relative to that dimension. The crack
damage width (Lcrack) obtained according to Eq. (2) corre-
Td 0 Td 40 Td max
sponds to the width value of the region in which the heat flux
Crack 1 27.92 36.02 37.30 disturbance operates. Thus, a crack with internal tile
Crack 2 27.31 34.73 36.24 detachment shows a greater width than that observed phys-
Crack 3 27.75 35.50 35.80 ically in the images (Fig. 1). It is observed that the Lcrack does
not evolve over time (neither increases nor decreases)
Tnd 0 Tnd 40 Tnd max
because the width of the heat flux disturbance is measured.
Crack 1 26.81 34.11 34.86 Thus, it can be deduced that Lcrack can be determined at any
Crack 2 27.10 34.08 35.52 moment of the monitoring provided that there is adequate
Crack 3 27.11 33.83 33.95 sharpness in the thermogram. The spurious values in crack 3
show a situation where it was not possible to limit the
inclinations properly, thus making difficult to determine Lcr.
As Lcrack evaluates the extent of the damage and not the
evaluations in moments of maximum temperatures physical width of the crack, a difference in the local heat flux
[29, 33]. that generated this different behavior (ceramic detach-
ments). Perhaps a new inspection with a more adequate
4.3 Damage evaluation angulation would allow better definition.
The results show Lcrack mean values of 61.85, 57.90, and
The relative crack depth can be directly determined from 54.12 mm for cracks 1, 2, and 3, respectively. The large
Delta-T. In the laboratory, it was possible to identify that number of spurious values of crack 3 must be mentioned.
the depth of cracking is directly related to the Delta-T in As shown in Fig. 11c, it is observed that the definition of
both heating and cooling regime [29, 33]. Thus, the anal- the pixel relative to the fissure is critical in several readings
ysis in Table 1 and Fig. 10 shows that crack 1 is deeper performed. The mean values observed are close, reflecting
than crack 3, which is deeper than crack 2. the damage area relative to each crack.

123
 Journal of Civil Structural Health Monitoring

Fig. 11 Temperature profiles of a crack 1, b crack 2, and c crack 3

Table 2 Damage width of the crack

Time (min) IFOV (mrad) Distance (m) Lcr1 (pixels) Lcrack1 (mm) Lcr2 (pixels) Lcrack2 (mm) Lcr3 (pixels) Lcrack3 (mm)

0 0.820 33 2 54.12 3 81.18 2 54.12

5 0.820 33 2 54.12 2 54.12 6 –a
10 0.820 33 2 54.12 2 54.12 6 –a
15 0.820 33 3 81.18 2 54.12 9 –a
20 0.820 33 2 54.12 2 54.12 2 54.12
25 0.820 33 2 54.12 2 54.12 9 –a
40 0.820 33 3 81.18 2 54.12 7 –a
Mean 60.89 57.50 54.12
a
non-significant spurious values

123
Journal of Civil Structural Health Monitoring

5 Final remarks are extremely specific and apply to very specific

conditions and situations.
The Delta-T and Lcrack data allow evaluating the degree of
damage related to this defect, providing severity data of the Acknowledgements The authors are grateful to Capes and CNPq for
investigated sample. This information is significant for the grants and to the Material Testing Laboratory of the University of
drafting the damage map that provides consistent subsidies Brasilia for providing the equipment used in the inspections.
for rehabilitation decisions.
In relation to the investigative process, the use of the
hygrothermal simulation is very advantageous allowing
References
pre-analysis and directions regarding the moment of the 1. Flores-Colen VP, de Brito I, de Freitas J (2009) Methodology for
monitoring. In the investigative process, the conclusions in-service performance assessment of rendering facades for pre-
reached could be obtained by monitoring only the initial dictive maintenance. TG75, W014, W080, W083 &W086, 388
40 min, which differentiated the Delta-T values from the 2. Silvestre J, de Brito JD (2014) Durability of current renderings: a
probabilistic analysis. Autom Constr 44:92–102. [Link]
cracks, thus evaluating their depth relationship, and 10.1016/[Link].2014.04.002
allowing determining the Lcrack damage width. Similar 3. Kanniyapan S, Mohammad G, Nesan IS, Mohammed LJ, Ganisen
investigations in cooling periods potentially have the same AH (2015) Façade material selection criteria for optimising
analysis potential, since the direction of the heat flux is building maintainability. Sci Eng 75(10):17–25
4. Bisegna F, Ambrosini D, Paoletti D, Sfarrab S, Gugliermettia F
reversed. Laboratory studies, therefore, corroborate these (2013) A qualitative method for combining thermal imprints to
statements [27, 33]. emerging weak points of ancient wall structures by passive
The results show that it is possible to monitor defects infrared thermography—a case study. J Cult Herit
and their effects with current equipment and to obtain a 15(2):199–202. [Link]
5. Martı́nez I, Castillo A, Martı́nez E, Castellote M (2013) Physico-
coherent analysis. Equipment with smaller IFOVs will chemical material characterization of historic unreinforced
surely increase analytical accuracy. It is noteworthy that masonry buildings: the first step for a suitable intervention.
the analysis distance was 33 m, quite reasonable for a Constr Build Mater 40:352–360. [Link]
facade inspection. buildmat.2012.09.091
6. Lourenço T, Matias L, Faria P (2017) Anomalies detection in
The study conducted employing passive thermographic adhesive wall tiling systems by infrared thermography. Constr
monitoring to quantify the degree of damage led to the Build Mater 148:419–428. [Link]
following main conclusions 2017.05.052
7. de Souza JS (2016) Evolution of façade degradation—effect of
• To study heat fluxes based on the hygrothermal simula- degradation agents and constituents of the elements. Dissertation,
tion to define the best monitoring allows a great advance Universidade de Brası́lia (UnB), Brası́lia, p 114 (in Portuguese)
8. Silvestre JD, de Brito J (2009) Ceramic tiling inspection system.
towards more accurate surveys to investigate cracks, at
Constr Build Mater 23(2):653–668. [Link]
the precise times when the heat flux disturbance is most buildmat.2008.02.007
evident. This can reduce the monitoring time needed to 9. Gaspar PL, de Brito J (2008) Quantifying environmental effects
investigate the anomalies. on cement-rendered facades: a comparison between different
degradation indicators. Build Environ 43(11):1818–1828. https://
• The degree of heat flux disturbance, evidenced by the
[Link]/10.1016/[Link].2007.10.022
thermographic images acquired, clearly associates the 10. Clemente P (2018) Extending the life-span of cultural heritage
degree of damage, which can be evaluated by the crack structures. J Civil Struct Health Monit 8(2):171–179
depth and damage width, both determined from the 11. Edis E, Flores-Colen I, De Brito J (2015) Time-dependent passive
building thermography for detecting delamination of adhered
thermogram.
ceramic cladding. J Nondestruct Eval 34:1–16
• The method of determining the width of the cracks 12. Rodrigues MFS, Teixeira JMC, Cardoso JCP (2011) Buildings
from the temperatures of the line perpendicular to the envelope anomalies: a visual survey methodology. Constr Build Mater
crack was adequate and allowed measuring the damage 25(5):2741–2750. [Link]
13. Sá G, Sá J, de Brito J, Amaro B (2015) Statistical survey on
dimension in the thermograms. From this, it is possible
inspection, diagnosis and repair of wall renderings. J Civ Eng
to have quantitative evaluations of the affected areas in Manag 21:623–636
the facades. 14. CIB (2014) Defects in masonry walls. Guidance of cracking:
• The temperature distribution pattern in the building, indentification, prevention and repair. CIB Commission W023,
Rotterdam
particularly the facade, is dynamic and subject to
15. Nascimento MLM, Bauer E, de Souza JS, Zanoni VAG (2016)
various phenomena. The different materials and ele- Wind-driven rain incidence parameters obtained by hygrothermal
ments that make up the facade respond to heat flux simulation. J Build Pathol Rehabil 1:1–7
differently. Thus, the interpretation of results requires 16. Barreira E, Almeida RMSF, Delgado JMPQ (2016) Infrared
thermography for assessing moisture related phenomena in
expertise associated with maintenance engineering.
building components. Constr Build Mater 110:251–269. https://
Very complex models are often unhelpful since they [Link]/10.1016/[Link].2016.02.026

123
 Journal of Civil Structural Health Monitoring

17. Edis E, Flores-Colen I, De Brito J (2015) Building thermography: characterization of subsurface cracks in concrete. Constr Build
detection of delamination of adhered ceramic claddings using the Mater 24(10):1888–1897. [Link]
passive approach. J Nondestruct Eval 34:1–13 2010.04.014
18. Kominsky JR, Luckino JS, Street NH, Martin TF (2007) Passive 35. Avdelidis NP, Moropoulou A (2004) Applications of infrared
infrared thermography—a qualitative method for detecting thermography for the investigation of historic structures. J Cult
moisture anomalies in building envelopes. Tedford Pond pp 1–11 Herit 5:119–127. [Link]
19. Pérez-Bella JM, Domı́nguez-Hernández J, Cano-Suñén E, del 36. Theodorakeas P, Avdelidis NP, Cheilakou E, Koui M (2014)
Coz-Dı́az JJ, Alonso-Martı́nez M (2017) On the significance of Quantitative analysis of plastered mosaics by means of active
the climate-dataset time resolution in characterising wind-driven infrared thermography. Constr Build Mater 73(1):417–425.
rain and simultaneous wind pressure. Part I: scalar approach. [Link]
Stoch Environ Res Risk Assess 32:1783–1797 37. Junyan L, Qingju T, Xun L, Yang W (2012) Research on the
20. Ferraz GT, De Brito J, De Freitas VP, Silvestre JD (2013) State- quantitative analysis of subsurface defects for non-destructive
of-the-art review of building inspection systems. J Perform testing by lock-in thermography. NDT E Int 45:104–110. https://
Constr Facil 30:1–8 [Link]/10.1016/[Link].2011.09.002
21. ISO 15686-2 (2012) Buildings and constructed assets—service 38. Dufour MB, Derome D, Zmeureanu R (2009) Analysis of ther-
life planning part 2: service life prediction procedures. ISO— mograms for the estimation of dimensions of cracks in building
International Organization for Standardization, London envelope. Infrared Phys Technol 52(2–3):70–78. [Link]
22. Flores-Colen I (2009) Methodology to evaluate the performance 10.1016/[Link].2009.01.004
in service of towed façades from the perspective of predictive 39. Lai WWL, Lee KK, Poon CS (2015) Validation of size estima-
maintenance. Doctoral thesis, Universidade Técnica de Lisboa, tion of debonds in external wall’s composite finishes via passive
p. 541 (in Portuguese) Infrared thermography and a gradient algorithm. Constr Build
23. Souza J, Silva A, De Brito J, Bauer E (2018) Analysis of the Mater 87:113–124. [Link]
influencing factors of external wall ceramic claddings’ service 03.032
life using regression techniques. Eng Fail Anal 83:141–155. 40. Weritz F, Arndt R, Röllig M, Maierhofer C, Wiggenhauser H
[Link] (2005) Investigation of concrete structures with pulse phase
24. Lourenço T, Matias L, Faria P (2016) Anomaly diagnosis in thermography. Mater Struct Constr 38:843–849
ceramic claddings by thermography a review. In 7th International 41. Kobayashi K, Banthia N (2011) Corrosion detection in reinforced
Conference on Safety and Durability of Structures ICOSADOS concrete using induction heating and infrared thermography.
25. Dias JM, Luı́s Matias, Maria J. Henriques, Maria Do Rosário J Civil Struct Health Monit 1(2):25–35
Veiga (2016) thermography (in Portuguese). In 10o Congresso 42. Barreira E, de Freitas VP (2007) Evaluation of building materials
Nacional de Mecânica Experimental, pp 2–4 using infrared thermography. Constr Build Mater 21(1):218–224.
26. Bauer E, de Freitas VP, Mustelier N, Barreira E, de Freitas SS [Link]
(2015) Infrared thermography—evaluation of the results repro- 43. Sciarretta F, Antonelli F, Peron F, Caniglia S (2018) Final out-
ducibility. Struct Surv 33:20–35 comes on the multi-disciplinary long-term monitoring and
27. Pavón E (2017) Criteria and behavioral patterns for evaluation of preservation state investigation on the medieval external façades
ceramic detachments with infrared thermography. Doctoral the- of Palazzo Ducale in Venice, Italy. J Civil Struct Health Monit
sis, Universidade de Brası́lia (UnB), Brası́lia, p 250 (in 8(1):111–133
Portuguese) 44. Bauer E, Pavón E, Oliveira E, Pereira CHF (2016) Facades
28. Kylili A, Fokaides PA, Christou P, Kalogirou SA (2014) Infrared inspection with infrared thermography: cracks evaluation. J Build
thermography (IRT) applications for building diagnostics: a Pathol Rehabil 1(1):1
review. Appl Energy 134:531–549. [Link] 45. Azenha M, Faria R, Figueiras H (2011) Thermography as a
nergy.2014.08.005 technique for monitoring early age temperatures of hardening
29. Bauer E, Pavón E, Pereira CHF, Nascimento MLM (2016) Cri- concrete. Constr Build Mater 25(11):4232–4240. [Link]
teria for identification of ceramic detachments in building facades 10.1016/[Link].2011.04.065
with infrared thermography. Recent developments in building 46. Barreira E, Almeida RMSF (2015) Drying evaluation using
diagnosis techniques 5. Springer, Singapore, pp 51–68 infrared thermography. Energy Procedia 78:170–175
30. Madruga FJ, Ibarra-Castanedo C, Conde OM, Lpez-Higuera JM, 47. Bauer E, Pavón E, Hildenberg A (2015) Errors in the use of
Maldague X (2010) Infrared thermography processing based on thermographic parameters of the mortar and ceramics in the
higher-order statistics. NDT E Int 43(8):661–666. [Link] detection of coating anomalies. In Simpósio Brasileiro de Tec-
10.1016/[Link].2010.07.002 nologia das Argamassas. p 12 (in Portuguese)
31. Maldague XP (2001) Theory and practice of infrared technology 48. Taylor T, Counsell J, Gill S (2014) Combining thermography and
for nondestructive testing. Wiley, NY computer simulation to identify and assess insulation defects in
32. Watase A, Birgul R, Hiasa S, Matsumoto M, Mitani K, Catbas the construction of building façades. Energy Build 76:130–142.
FN (2015) Practical identification of favorable time windows for [Link]
infrared thermography for concrete bridge evaluation. Constr 49. De Freitas SS, De Freitas VP, Barreira E (2014) Detection of
Build Mater 101:1016–1030. [Link] façade plaster detachments using infrared thermography—a
buildmat.2015.10.156 nondestructive technique. Constr Build Mater 70:80–87. https://
33. Bauer E, Pavón E, Barreira E, De Castro EK (2016) Analysis of [Link]/10.1016/[Link].2014.07.094
building facade defects using infrared thermography: laboratory
studies. J Build Eng 6:93–104. [Link]
2016.02.012 Publisher’s Note Springer Nature remains neutral with regard to
34. Aggelis DG, Kordatos EZ, Soulioti DV, Matikas TE (2010) jurisdictional claims in published maps and institutional affiliations.
Combined use of thermography and ultrasound for the

123