Engineering, 2013, 5, 881-886
Published Online November 2013 (http://www.scirp.org/journal/eng)
http://dx.doi.org/10.4236/eng.2013.511108
Open Access ENG
Load-Measuring Pot Bearing with Built-In Load Cell
—Part II: Fatigue Performance and Experimental
Temperatur e Correction
Jeong-Rae Cho, Young Jin Kim*, Jong-Won Kwark, Sung Yong Park,
Won Jong Chin, Byung-Suk Kim
Structural Engineering Research Division, Korea Institute of Construction Technology, Goyang-Si, Korea
Received September 4, 2013; revised October 4, 2013; accepted October 11, 2013
Copyright © 2013 Jeong-Rae Cho et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
This paper presents the results of fatigue performance tests performed up to 10 million cycles on a load-measuring pot
bearing with built-in load cell to ve rify its field applicability and proposes an empirical temperature correction formula.
In Part I of this work, various measurement performances of the load-measuring pot bearing were evaluated through
static and dynamic loading tests. Bridge bearings are subjected to the effect of fatigue caused by the repeated applica-
tion of moving loads an d exposed to h arsh site conditions includ ing cold and hot weath ers differently to labo ratory con-
ditions. Accordingly, the durability of the load-measuring pot bearing with built-in load cell shall be secured and the
environmental effects like temperature shall be minimized for its application on field. This study conducted fatigue tests
up to 10 million cycles on a load-measuring pot bearing with the capacity of 1000 kN to examine eventual degradation
of the measurement accuracy with respect to the number of fatigue loading cycles. In addition, the experimental tem-
perature correction procedure is proposed to obtain the temperature correction formula enabling to correct the effect of
temperature on the load measurement.
Keywords: Pot Bearing; Load-Measuring; Load Cell; Fatigue; Temperature Correction
1. Introduction
Cho et al. [1] (referred simply as Part I hereafter) pro-
posed the design concept of a load-measuring pot bearing
with built-in small size button-type load cell inserted in
the base plate of the bearing and examined experimen-
tally its measurement accuracy and the possibility to
achieve dynamic measurement. This study evaluates the
durability of the developed bearing for its application in
real bridge sites and suggests an empirical temperature
correction formula.
Since the load cell presents a load su pporting structure
in which the strain is converted into an electric signal by
a strain gauge attached to the structure, the load cell se-
cures outstanding durability when fabricated as a struc-
ture enabling sufficient protection. Bridge bearings shall
in general secure durable performance for more than 20
years and are subjected to the repeated application of
traffic loads. This means that the load cell inside the
load-measuring pot bearing presented in Part I is exposed
to conditions significantly harsher than those encoun-
tered by ordinary measurement devices. The Highway
Bridge Design [2] requires the bearings to secure fatigue
durability by performing fatigue performance tests up to
2 million cycles. This study co nducted fatigue tests up to
10 million cycles to evaluate the stability of the bearing
itself and its accuracy as load-measuring device with
respect to the number of fatigue cycles accounting for the
more severe conditions.
Besides, the load cell itself is fabricated with a built-in
temperature correction circuit enabling it to measure the
loads regardless of the changes in the ambient tempera-
ture. However, the load-measuring pot bearing developed
in Part I includes an elastomer disk that is extremely
temperature-dependent as central medium transferring
the load to the built-in load cell. This situ ation requires to
evaluate the measurement accuracy with respect to the
temperature and to apply a temperature correction for-
mula if necessary. This study suggests an empirical tem-
*Corresponding a uthor.

J.-R. CHO ET AL. 882
perature correction procedure to obtain the temperature
correction formula adapted to the load-measuring pot
bearing with built-in load cell.
2. Fatigue Test
Fatigue test performed on bridge structural elements like
the bearing shall be executed by applying a fluctuating
loading up to the number of cycles specified in the de-
sign code considering the vehicle loads with reference to
the permanent loads of the bridge. The Highway Bridge
Design Code specifies the fatigue durability with refer-
ence to 2 million cycles [2]. However, this study con-
ducted fatigue tests up to 10 million cycles accounting
for the more severe conditions [3,4].
Bridge bearings supporting the superstructure (perma-
nent loading state) are basically loaded up to 40% - 60%
of their capacity and experience additional varying loads
due to traffic. Here, the load level induced b y the perma-
nent loads is set to 60% and the fluctuation of the load
due to the vehicle loads is set to 15% to conduct the fa-
tigue test on the 1MN-L1 specimen with capacity of
1000 kN of Part I. Concretely, the loading history of the
planned fatigue test has Pmin = 500 kN and Pmax = 750 kN
as shown in Figure 1. The actual test proceeded by dis-
posing a load cell for the measurement of the load ap-
plied on the bearing and the load-measuring bearing be-
tween two steel blocks as shown in Figure 2, and by
applying the addition al dynamic fatigu e load by means of
an actuator after having applied a constant load through
steel bars loaded by the hydraulic jack of which pressure
has been maintained by means of the hydraulic nut. The
load applied by the hydraulic jack is 500 kN, and the
additional load of about 200 kN is applied by a dynamic
actuator with capacity of 250 kN so as to achieve the
intended load level of 500 kN to 750 kN to be experi-
enced by the bearing. Figure 3 plots the fatigue load ac-
tually applied on the load-measuring bearing and meas-
ured by the load cell disposed above the bearing. The
reason for the application of a pre-load using the steel
bars and the hydraulic jack instead of applying the load
Figure 1. Loading history of fatigue test.
Figure 2. Conceptual scheme of the load-measuring pot
bearing with built-in load cell.
Figure 3. Load applied directly on the specimen during the
fatigue test.
directly using a dynamic actuator with capacity of 1000
kN is due to the durability of the dynamic actuator that
has to withstand 10 million cycles of fatigue loading at a
high level of 500 to 750 kN.
Open Access ENG

J.-R. CHO ET AL. 883
The stability of the bearing itself and its accuracy as
measuring device shall be evaluated according to the
number of fatigue loading cycles through fatigue test. To
that goal, static loading test was performed by applying
the load using the hydraulic jack illustrated in Figure 2
by steps of 500,000 cycles or 1 million cycles after the
application of 2 million fatigue loading cycles. Since this
test uses a hydraulic jack, loading cannot be applied up to
the maxi mum l oad at con stan t speed as done in the static
loading test of Part I. Figure 4 plots the test results with
respect to the major fatigue cycles. The solid lines repre-
sent the values measured by the external load cell in
Figure 2 and correspond to the lo ad applied on the load -
measuring bearing. The dashed lines represent the load
obtained by converting the signals measured by the small
size built-in load cell inserted in the bearing using the
conversion formula for specimen 1MN-L1 derived in
Part I and expressed here in Equation (1).
456.2208 66.4447Ff, (1)
where f is the load (kN) measured by the 5000 kN-load
cell inserted in the bearing; and, F is the vertical load
(kN) applied on the bearing that is the load converted
from the measured load.
Figure 5 plots the error rate obtained by dividing the
error defined as the difference between the actually ap-
plied load and the converted load by the capacity of the
bearing according to the number of fatigue cycles. The
error rate of specimen 1MN-L1 that reached 0.39% in the
static loading test in Part I is seen to increase with larger
number of cycles. The error rate remains below 1% until
4 million cycles and exceeds 1% after 5 million cycles to
reach 2.77% at 8 million cycles but stands within 2% at
Figure 4. Fatigue test results.
Figure 5. Variation of the error according to the number of
fatigue cycles.
the whole between 5 million to 10 million cycles. Con-
sidering that the low accuracy load cell secures an accu-
racy of 1%, these values verify that the error has a level
enabling the application of the load-measuring pot bear-
ing for real measurem e nt.
3. Experimental Temperature Correction
The load cell itself is fabricated with a built-in tempera-
ture correction circuit enabling it to measure the loads
regardless of the changes in the ambient temperature.
However, the load-measuring pot bearing developed in
Part I include an elastomer disk that is extremely tem-
perature-dependent as central medium transferring the
load to the built-in lo ad cell. Following, need is to evalu -
ate the effect of the temperature.
The characteristics of the variation of temperature re-
cords can be overcome by suggesting a temperature cor-
rection formula derived from static loading test con-
ducted at various temperatures in a laboratory maintained
with a constant ambient temperature. In this study, the
experimental temperature correction procedure is pro-
posed to obtain temperature correction formula by per-
forming the test simulating the field conditions as possi-
ble using an apparatus conceived to maintain a constant
load level. Figures 6 and 7 illustrates the testing appara-
tus enabling to maintain a constant load and to apply
loading on the bearing with cap acity of 1000 kN without
particular equipment. This apparatus is identical to the
test setting used for the fatigue test in Chapter 2 except
the absence of the actuator used to apply the additional
dynamic load. Compared to the testing apparatus of
Chapter 2, the capacity of the hydraulic nut is increased
to enable loading by means of a hydraulic jack with ca-
pacity larger than 1 000 kN. The apparatus was fabricated
to prevent as possible the occurrence of eccentricity. This
apparatus allows static loadin g test using a hydrau lic jack
and also to apply constant loading by fixing the hydraulic
nut [4,5].
The specimen adopted for the test is the load-measur-
Open Access ENG

J.-R. CHO ET AL. 884
Figure 6. Setting of temperature correction test.
Figure 7. Hydraulic loading during the temperature correc-
tion test.
ing pot bearing with capacity of 1000 kN fabricated with
identical dimensions and characteristics to 1MN-L1 of
Part I. For the load-measuring pot bearing proposed in
Part I, static loading test shall be performed on all the
fabricated bearings to obtain the applied load and the
conversion formula of the built-in lo ad cell. To go further,
loading test shall be re-executed when the bearings are
reassembled since the contact conditions between the
button of the built-in load cell and the elastomer disk
inside the bearing undergo changes. Accordingly, static
loading test was conducted by loading the fabricated
bearing by the hydraulic jack using the test apparatus
described in Figure 6. Differently to the test using the
actuator in Part I, the loading speed is not constan t in this
test because loading is applied by a hydraulic jack in
which the hydraulic pressure is controlled manually.
Figure 8 plots the test results and the conversion formula
is as follows.
456.2208 66.4447Ff, (2)
where f is the load (kN) measured by the 5000 kN-load
cell inserted in the bearing; and, F is the vertical load
(kN) applied on the bearing that is the load converted
from the measured load. Equation (2) was obtained
through linear regression analysis limited to the loads
larger than 30% of the capacity of the bearing as ex-
plained in Part I. The maximum error for load level lar-
Figure 8. Static loading test and computation of conversion
formula.
ger than 30% is 0.25%. The test was performed at ambi-
ent temperature of 31.3˚C.
Temperature correction test was conducted by main-
taining the load at 1000 kN using the hydraulic jack and
fastening the hydraulic nut. The signals of the load cell
measuring the load applied externally, the signals of the
built-in load cell and the temperature were measured
with respect to time. Slight change of the load from the
constant value of 1000 kN occurred due to temperature-
induced changes encountered by the steel bars, bearing
and load cells. Figure 9 plots the temperature history, the
externally applied load history and the load history ob-
tained by conversion of the measurement of the built-in
load cell. It can be observed that all the histories exhibit
similar trend. However, the error history between the
external load history and measured load history reaches
15 kN and shows regular pattern with respect to the tem-
perature history. Consequently, need is to correct this
error history.
In order to correct the error with respect to the tem-
perature history, need is to examine the factors causing
such error. To that goal, the temperature history and
loading history and, the temperature history and meas-
ured load history were superposed and magnified as
shown in Figure 10. In Figure 10(a), the external load
history and the temperature history exhibit roughly simi-
lar trend but a slight time lag can be observed as well as
some difference in the shape of the sine wave. Such dif-
ferences can be attributed to the characteristics of the
hydraulic nut fixing the steel bars and the elastomer disk
inside the bearing. Besides, in Figure 10(b), the meas-
ured load history obtained by conversion of th e load his-
tory of the small size load cell does not show time lag in
comparison to the thermal load history and their shapes
are also in good agreement. This good correspondence
can be explained by the change of the contact area with
the elastomer disk owing to the instantaneous deforma-
tion of the built-in load cell itself. To summarize, the
error according to the temperature history of the meas-
Open Access ENG

J.-R. CHO ET AL. 885
(a)
(b)
(c)
Figure 9. Temperature correction test results: (a) Tem-
perature history; (b) Load history; (c) Error history.
ured and applied load histories is caused to the instanta-
neous deformation of the built-in load cell with respect to
the temperature, which modifies the degree by which the
load cell is inserted in the flexible elastomer disk and, in
turn, changes the contact area between the load cell and
the elastomer. Based upon this discussion, the conversion
formula enabling to correct the temperature history is
modified and is proposed as follows.
 


0
F
tAaTftt 
(a)
(b)
Figure 10. Comparison of temperature and loading histo-
ries: (a) Temperature history and loading history; (b) Tem-
perature history and measured load history.
where A and C are constants that can be obtained from
the static loading test, here A = 418.93 and C = –51.79.
These values are calculating assuming that the constants
are maintained at definite temperature and during a pe-
riod longer than a definite time. In Equation (3), t0 is the
time lag constant caused by the characteristics of the
elastomer disk determined in the long-term measurement
test. In the figure, t0 = 0.11 day. The constant a has value
representing the change in the contact area according to
the characteristics of the elastomer disk and temperature
change in the long-term measurement test and can be
obtained by linear regression analysis. Here, a = –0.573.
Finally, the conversion formula can thus be rewritten as





418.93 0.5730.1151.79FtT ft (4)
Figure 11 plots the history of the measured load cal-
culated using the conversion formula corrected with re-
spect to the temperature history, the history of the ap-
plied load and the corresponding error history. It can be
seen that the error experiences significant reduction from
15 kN before correction to 10 kN after correction.
C
(3)
Open Access ENG

J.-R. CHO ET AL.
Open Access ENG
886
(a)
(b)
Figure 11. Test results when applying the corrected conver-
sion formula: (a) Applied load history and corrected meas-
ured load history; (b) Error history when applying the cor-
rected conversion formula.
4. Conclusion
Even if the measurement accuracy tended to degrade
gradually with the accumulation of fatigue, the degree of
accuracy was verified to be sufficient for the intended
application. The error history of the developed bearing
exhibited regular variation according to the temperature
history. The application of the conversion formula ob-
tained by linear regression analysis of the external load
and the load measured by the built-in load cell in the
static loading test resulted in an error reaching 1.5% to
2.0% of the total capacity of the bearing. Moreover, the
cause of such error could be attributed to the instantane-
ous deformation of the built-in load cell with respect to
the temperature, which modified the extent by which the
load cell was inserted in the flexible elastomer disk and,
in turn, changed the contact area between the load cell
and the elastomer. Accordingly, the experimental tem-
perature correction procedure is proposed to reduce the
temperature induced error. The modified conversion
formula from the procedure enabled to reduce the error to
approximately 1%. Outstanding sensitivity and accuracy
of measurement could be verified through the fatigue
tests. The load-measuring pot bearing with built-in load
cell was seen to exhibit sufficient field applicability un-
der application of the proposed temperature correction
formula.
5. Acknowledgements
This research was supported by a grant from a Strategic
Research Project (Development of smart prestressing
system for prestressed concrete bridges) funded by the
Korea Institute of Construction Technology.
REFERENCES
[1] J.-R. Cho, Y. J. Kim, J.-W. Kwark, S.Y. Park, W. J. Chin
and B.-S. Kim, “Load-Measuring Pot Bearing with
Built-in Load Cell—Part I: Design and Performance,”
Engineering, Scientific Research Publishing, 2013 (under
Review).
[2] Ministry of Land, Transport and Maritime Affairs, High-
way Bridge Design Code, 2010, in Korean.
[3] Korea Institute of Construction Technology, “R&D on
the Safety of High-Speed Railway Structures,” R&D Re-
port, 2002, in Korean.
[4] Korea Institute of Construction Technology, “Develop-
ment of Stabilization Techniques for Track Construction
System of the High-Speed Railway,” R&D Report, 2003,
in Korean.
[5] Korea Institute of Construction Technology, “Develop-
ment of Stabilization Techniques for Track Construction
System of the High-speed Railway,” R&D Report, 2004,
in Korean.